selected problems of soil tillage systems and operations

Komentarze

Transkrypt

selected problems of soil tillage systems and operations
SELECTED PROBLEMS OF SOIL
TILLAGE SYSTEMS AND
OPERATIONS
Editorial staff:
Piotr F. Borowski
Marek Klimkiewicz
Małgorzata Powałka
Faculty of Production Engineering
Warsaw University of Life Sciences
Warsaw 2010
Reviewers:
Piotr F. Borowski
Jerzy Buliński
Marek Gaworski
Marek Klimkiewicz
Zbigniew Majewski
Zygmunt Owsiak
Małgorzata Powałka
Janusz Wojdalski
ISBN 978-83-928876-6-9
PRINT: WEMA
2
CONTENTS
Chapter 1
The benefits of conservation tillage and its applications in the world
Engin ÇAKIR
4
Chapter 2
Prediction of the effects of tillage in relation to soil organic matter content
Ewa A. CZYŻ, Anthony R. DEXTER
11
Chapter 3
Determination of the soil stability in relation to ecological agriculture
Ewa A. CZYŻ, Anthony R. DEXTER, Olga P. GAŢE
22
Chapter 4
Effect of applied simplifications in tillage systems on biological activity and
POM content in soil
Anna Maria GAJDA
35
Chapter 5
Data mining as a decision-support tool in sustainable agricultural production
Jędrzej TRAJER, Jarosław STANKIEWICZ
47
Chapter 6
Site specific crop fertilization in Haute-Normandie: a tool to improve the
profitability of cereal crop systems
Jean Marc LLORENS, Olivier PERRIN, Odile BOURGAIN
57
Chapter 7
Ecological aspects of soil compaction by wheels of agricultural vehicles
Jerzy BULIŃSKI
77
Chapter 8
Analysis of the working resistance of the working units used in simplified tillage
Krzysztof PIECZARKA
90
Chapter 9
The influence of fertigation on agro-physical properties of arable plants
Ján JOBBÁGY, Ján SIMONÍK, Marek KLIMKIEWICZ
99
Chapter 10
Ecological aspects of energetic plants harvesting
Aleksander LISOWSKI, Adam STRUŻYK, Jacek KLONOWSKI, Tomasz
NOWAKOWSKI, Jarosław CHLEBOWSKI, Michał SYPUŁA, Stanisław GACH
108
Chapter 11
Effect of season on carbon dioxide and ammonia production in broiler housing
Štefan MIHINA, Monika KNÍŽATOVÁ, Jan BROUČEK
123
3
Chapter 1
THE BENEFITS OF CONSERVATION
TILLAGE AND ITS APPLICATIONS
IN THE WORLD
Engin ÇAKIR
INTRODUCTION
To prevent soil degradation and to establish sustainable farming conservation
tillage is necessary especially in the arid areas of the world. Beside this, increasing
the cost of fuel in tillage operations are reducing the time are getting the main factor
for changing the conventional farm techniques. Minimum tillage and direct seeding
are some of the methods that farmers apply recently for a long-term erosion free
farming with minimum fuel cost. Considering the negative effect of intensive farming
in the field, direct seeding becomes more vital for farmers for establishing the nature
of the soil and flora allowing natural plant growth with less plant protection problems.
Conservation tillage was first introduced in the world in 1940, and first no-till
applications were seen in 1960’s. Especially conservation tillage was progressed so
quickly in last thirty years. No-till was applied in the United States of America
2,200,000 ha in 1973/74 and this increased to 4,800,000 ha and 19,400,000 ha in
1983 and 1997, respectively.
If we look at the erosion damage in the world then, we can say that soil
degradation is responsible for making from 2 million ha to 12 million ha or 0.3–0.8%
of the world's arable land, unsuitable for agricultural production every year, with wind
and water erosion accounting for 84% of the soil degradation [den Biggelaar et al.,
2004a]. For this reason, good soil management to protect the soil against erosion in
order to sustain long-term productivity is imperative for meeting the world's future
needs for food and fibre production [den Biggelaar et al., 2004b].
In many studies, conservation tillage saves great deal of fuel and time. This is
very important that most of the cost in agriculture is due to the fuel consumptions.
Köller [2003] found 73% fuel energy save comparing the conventional method in
direct seeding. According to the research findings, fuel consumptions of different
tillage methods were found as 49,4 L ha-1 (100%), 31,2 L ha-1 (63,2%), 28,3 L ha-1
(57,3%), 25,2 L ha-1 (50,9%), 13,3 L ha-1 (27,08%) for plough, chisel, disk harrow,
ridge-tillage and direct seeding respectively.
4
Conventional tillage methods not only cause soil loss but also wear outs the
soil carbon in a long monoculture. In the depth of 0-20 cm, soil organic carbon lost
increases with conventional method [Zinn et al, 2005]. Similar result was found from
Mann [1986]. He found that organic carbon loss was 23% and which amounts 15 Mg
ha-1.
Conservation tillage establishes a good soil environment with a suitable soil
flora. With a long run conservation tillage practices, soil regains its original natural
strength which becomes less sensitive to the compaction due to the heavy load of
machineries.
The aim of this paper is to examine the conservation tillage practices and give
some examples of applied conservation tillage studies in Turkey.
BENEFITS OF CONSERVATION TILLAGE
Traditional soil cultivation systems in the tropics and subtropics, with intensive
soil tillage, will end in soil degradation and loss of crop productivity. This will result in
poverty, exodus of farmers from rural areas, resulting in an increase of city slums and
marginal populations, and finally in social conflicts. If we are to offer the farm family a
chance to survive on the farm and if sustainable agriculture is to be achieved, than
the paradigms of soil use and management must be changed and new farming
practices must be implemented [Derpsch, 2006]. So we have to somehow convince
the farmers about the benefits of conservation tillage by demonstrations and well
extension services. It is vital that new techniques should be adapted to change the
path of bad heading in soil degradation and erosion.
Conservation tillage has mainly four important advantages: Less work, cost
effective, erosion control and improvement in the quality of life. In fact the most
reason anyone can find is to establish sustainable farming for a long term. Some of
the benefits of the conservation tillage can be listed as:
•
Reduced labor requirements
•
Time savings
•
Fuel savings
•
Improved long- term productivity
•
Reduced machinery wear
•
Reduced soil erosion
•
Greater soil moisture retention
•
Improved water infiltration
•
Decreased soil compaction
•
Reduced release of carbon gases
•
Reduced air pollution
Comparing the conventional methods, it is important that tillage is not
necessary for crop production. If we can minimize or apply no-tillage along with good
management, we can slowly form the natural environment for a good sustainable
crop production. For example, instead Burying of plant residues with tillage
implements, we leave them on the soil surface as mulch. Or we can replace the
chemical pest control with biological pest control. We must definitely give up on
monoculture and start using green manure and crop rotations.
By using residue covered conservation tillage, we can reduce the wind and
water erosion to near to zero. Giving up from the intensive soil tillage systems to
minimized or no-till systems, we can create a good soil flora with more available soil
5
moisture, increased water infiltration into the soil and sequestered more carbon in the
soil.
CONSERVATION AGRICULTURAL PRACTICES IN THE WORLD
Last 30 years or so, application of conservation tillage increased in the world
very fast. Starting first with minimum tillage applications, conservation tillage is now
reached to no-till or zero till agriculture in which tillage is not performed. No-till
applications increased in the countries where at beginning it was applied hesitantly
(Table 1-2). The leading countries in the world with the biggest area under no-tillage
are the USA with 19.3 million hectares followed by Brazil with 11.2 million ha,
Argentina with 7.3 million ha, Canada with about 4.1 million ha, Australia with 1
million ha and Paraguay with 790.000 ha of the technology being practiced by
farmers. It is estimated that no-tillage is practiced on about 45 million hectares world
wide.
Table 1. Area under no-tillage in the seventies and eighties (Hectares) [Derpsch
et al., 2006]
Country
1973/ 74
1983/ 84
U.S.A
2,200,000
4,800,000
United Kingdom
200,000
275,000
France
50,000
50,000
Netherlands
2,000
5,000
Japan, Malaysia, Sri
200,000
250,000
Australia
100,000
400,000
New Zealand
75,000
75,000
Brazil
1,000
400,000
Table 2. Total area under no-tillage in different countries in 1996/ 97 (Hectares)
[Derpsch at al, 2006]
Country
Area under no- tillage (ha)
Argentina
4,400,000
Brazil
6,500,000
Canada
6,700,000
Mexico
490,000
Paraguay
500,000
Uruguay + Chile + Bolivia
500,000
U.S.A.
19,400,000
Australia
1,000,000
Others
460,000
Lafond et al. [2005] worked on the effects of three tillage methods; zero,
minimum and conventional methods for three four-year crop sequences in grain yield
over a 12-year period under Canadian growing conditions. Their results indicated that
a one-year non-cereal break crop was enough to alleviate the negative effects of
consecutive cereal crops on winter wheat. Results of this study also supported the
large shifts towards in conservation tillage being observed in the Canadian prairies.
Dam et al. [2005] studied the effect of different tillage practices and corn
residues on soil bulk density, corn emergence rates and crop yields over an 11-year
period for corn (Zea mays L.). They applied no-till, reduced tillage and conventional
6
tillage methods with corn residues and without residues. According to their findings,
bulk density was 10% higher in no-till (1.37 Mg m−3) than in conventional tillage
(1.23 Mg m−3), particularly at the 0–0.10 m depth. They found that No-till with residue
(NTR) possibly had the slowest overall emergence due to the higher surface residue
cover (8.5 Mg ha−1 in 1996) and higher bulk density (1.37 Mg m−3 over the 11 years).
But, they observed that long-term mean dry matter corn yields were not affected by
tillage and residue practices during the course of this study.
Liebig et al. [2004] made a research in which they examined interactive effects
of tillage, crop sequence, and cropping intensity on soil quality indicators for two longterm cropping system experiments in the northern Great Plains. They found that
within the surface 7.5 cm, the continuous crop, no-till system possessed significantly
more soil organic carbon (by 7.28 Mg ha-1), particulate organic matter (by 4.98
Mgha1), potentially mineralizable N (by 32.4 kg ha-1), and microbial biomass carbon
(by 586 kg ha-1), as well as greater aggregate stability (by 33.4%) and faster
infiltration rates (by 55.6 cm h-1) relative to the crop-fallow and conventional tillage
system.
From the results of long-term research made by Megyes et al. [2003], it was
shown that crop production technologies based on reduced/conservation tillage
methods may replace conventional tillage systems and are applicable under the
environmental conditions of Hungary.
Rajan and Khera conducted the field study to estimate the effect of tillage and
different modes of mulch application on soil erosion losses. Treatments comprised
minimum and conventional in the main plots and five modes of straw mulch
applications; mulch spread over whole plot (Mw), mulch spread on lower one-third of
plot (M1/3), mulch applied in strips (Ms), vertical mulching (Mv) and unmulched control
(Mo). Mulch spread over whole plot reduced runoff by 33%. Runoff and soil loss were
5 and 40% higher under conventional tillage conditions. Straw mulching reduced
maximum soil temperature and helped in conserving soil moisture. Minimum tillage
coupled with mulch spread over whole plot was highly effective in reducing soil
erosion losses, decreasing soil temperature and increasing moisture content by
providing maximum surface cover.
Anken and his friends examined how soil tillage systems influence soil quality
especially under the conditions of moist cool conditions in Switzerland. They aimed to
reduce the soil tillage and to check its effect on the yield. They found that reduced
tillage
increased
earthworm
populations,
reduced
Pseudocercosporella
herpotrichoides infection in wheat (Triticum aestivum) and increased plant
colonisation by arbuscular mycorrhizal fungi. They observed that yields for no-tillage
and other ploughless cultivation techniques were on par with those obtained by
ploughing. For direct-drilled maize (Zea mays), the yield was decreased 10% over
the course of 14 years. The no-tillage method did not differ from the others in respect
of bulk density, but it showed an increased preconsolidation stress and hence better
trafficability. Direct seeding method in dry soil conditions with low stubble height had
the minimum working time as 2.9 h ha-1 comparing other methods. Regarding silage
corn yield, although conventional system provided 68.31 Mg ha-1 as the maximum
yield while direct seeding had 67.58 Mg ha-1 yield, there was found no statistical
differences between two methods. The lowest yield was found in direct seeding
method applied in dry soil conditions with high stubble height as 47.08 Mg ha-1. The
researchers also examined the change in organic matter in the soil and found the
most increase 11.8% in direct seeding machine. According to the economic
calculations of the methods, the cost of 422.63 USD ha-1 was the highest value found
7
in conventional tillage. The costs for direct seeding were 321.50 USD ha-1 and
305.33 USD ha-1 in the fields with high stubble height and low stubble height,
respectively.
Yalcin et al. [2003] studied the effect of direct seeding methods on second
crop cotton in Soke region for 3 years. They applied conventional, minimum tillage
and direct seeding methods and they examined the tillage effects on soil compaction,
seedling emergency, plant growth and yield. Methods were applied in dry ad wet soil
conditions. The results showed that conventional tillage method required 11.7 times
more fuel comparing the direct seeding in second crop cotton.
There are many studies in this field which many of them points out the
benefits of application of conservation tillage especially no-till. We know very well
from the experiences that, semi-arid and arid regions should strictly apply the
conservation tillage. According to the resent studies, no-till should be practiced for a
long term sustainable farm. Increasing the application of no-till in the world is a proof
of erosion free conservation tillage.
The reasons why conservation tillage methods especially direct seeding and
no-till are not used by farmers widely can be listed as follows:
•
Not enough information available in this field,
•
They heard conservation tillage but they do not know how to do it,
•
Some of them tried earlier, but left this method and returned to the conventional
tillage because received low yield,
•
There is not any government support,
•
Crop rotation is almost impossible due to low income of the farmers,
•
Direct seeding machines are not available in the market to try.
CONCLUSIONS
The advantages of conservation tillage methods such like low cost, time saving
and sustainable farming will be very much convincing to spread the usage of these
methods. The use of farming techniques providing high income with minimum
damage to environment becomes the main objective of all the farmers in the world.
Increase in the profit directly related with lowering the costs in the farm. The
conservation tillage methods should be practiced not only for profitable farming, but
also for preserving our natural resources and leaving to our next generation in good
conditions.
REFERENCES
1.
Anken T., Weisskopf, P., Zihlmann, U., Forrer, H., Jansa, J., Perhacova K.
2004: Long-term tillage system effects under moist cool conditions in
Switzerland. Soil & Tillage Research 78, s. 111-183.
2.
Anonymous, 2007: Some Agricultural Statisticsi,
3.
http://www.tuik.gov.tr/PreIstatistikTablo.do?istab_id=135
4.
Anonymous, 1991: General Agricultural Census, Village Information Survey
Results, T.C. Government Statistics Institutes DİE, Ankara, Turkey, 1991.
5.
Anonymous, 1991: General Agricultural Census, 2004. T.C. Government
Statistics Institutes DİE, Ankara, Turkey, 2004.
6.
Anonymous 1998: TMMOB Chamber of Meteorological Engineers 1998 Year
Agenda
8
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Aykas, E., Onal, I. 1999: Effects of different tillage, seeding and weed control
methods on plant growth and wheat yield. 7. International Congress on
Mechanization and Energy in Agriculture , 26-27 May Adana- TURKEY
Aykas, E., Yalçın, H., Çakır, E., Gülsoylu, E. 2005: Koruyucu Toprak İşleme
Yöntemleri ve Doğrudan Ekim. E.Ü.Ziraat Fakültesi Dergisi, 42 (3), s. 195 - 207,
ISSN:1018–8851, Bornova-İZMİR.
Cakir E., H. Yalcin., E. Aykas., H. Akdemir., T. Ocel., H. Soya. 2005: Direct
seeding applications on wheat in Odemis Region. Agricultural Machinery
Science Magazine. ISSN 1306-0007. Vol. 1 No. 2 pp 109-114 İZMİR
Çakır E., Aykas E., Yalçın H., 2003: Tillage Parameters and Economic Analysis of
Direct Seeding, Minimum and Conventional Tillage In Wheat. ISTRO
(International Soil Tillage Research Organization) Conference, Brisbane,
Australia.
Çakır E., Yalcin H., Aykas E., Gülsoylu E., Okur B., Delibacak S., Ongun A.R.
2006: The Effects of Conservation Tillage and Direct Seeding Applications on
the Yield of Second Crop Mazie. ISTRO – 17th Conference of the International
Soil Tillage Research Organization, August 28-Sept.3 2006 Kiel, GERMANY.
Çakır E., Yalçın H., Aykas E., Akdemir H., Öcel T., Soya H. 2004: Ödemiş
Koşullarında Buğday Tarımında Doğrudan Ekim Uygulamaları. Tarımsal
Mekanizasyon 22. Ulusal Kongresi. 8-10 Eylül 2004, AYDIN, Bildiriler Kitabı. s.
17-18.
Çakır E., Yalcin H., Aykas E., Gülsoylu E., Okur B., Delibacak S., Ongun A.R.
2006: Koruyucu Toprak İşleme ve Doğrudan Ekimin İkinci Ürün Mısır Verimine
Etkileri: Birinci Yıl Sonuçları. Tarım Makinaları Bilimi Dergisi,. ISSN: 1306-0007
Cilt 2, Sayı 2, Sayfa: 139-146.
Dam R.F., Mehdi B.B., Burgress M.S.E., Madramootoo C.A., Mehuys G.R.,
Callum I.R. 2005: Soil bulk density and crop yield under eleven consecutive
years of corn with different tillage and resiude practices in a sandy loam soil in
central Canada. Soil & Tillage Research 84 (2005) , 41-53.
Derpsch R., Florentín M. & Moriya K., 2006: The laws of diminishing yields in
the tropics. Proceedings on CD, 17th ISTRO Conference, Kiel, Germany,
August 28 - September 3, 2006, p 1218 - 1223
Guo L.B., Gifford R.M., 2002: Soil carbon stocks and land use change: a meta
analysis. GI. Change Biol. 8, 345-360. International Soil Tillage Research
Organization (ISTRO), INFO- EXTRA, Vol. 3 Nr° 1, January 1997.
Korucu T., V.Kirisci S. Gorucu. 1998: Conservation tillage and its application in
Turkey. 18. National Congress on Mechanization and Energy in Agriculture.
Tekirdağ.
Korucu T., V.Kirisci 2003: Economic and technical analysis of using different
tillage techniques on second crop maize. Workshop on Conservation Tillage
and Direct Seeding. İZMİR pp.122-142.
Korucu T., Kirişçi V., Keskin S.G., 1999: Residue management for successful
direct seeding. 7th International congress on agircultural mechanization and
energy 26-27 May, 1999. Adana. Proceediings, p. 232-238.
Kosutic S., Filipovic D., Gospodaric Z., 1999: energy Requirements in maize
using conventional, conservation and no-tillage systems. 7th International
congress on agricultural mechanization and energy 26-27 May, 1999. Adana.
Proceediings, p. 1-6.
Özpınar S., Işık A., 1999: Determination of the effects of the different tillage and
planting methods on cotton yield in the southeast anatolia project area. 7th
9
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
International congress on agircultural mechanization and energy 26-27 May,
1999. Adana. Proceediings, p. 249-256.
Liebig M.A., Tanaka D.L., Wienhold B.J., 2004. Tillage and cropping effects on
soil quality indicators in the northern Great Plains. Soil & Tillage Research 78,
131-141.
Mann L.K., 1986: Changes in soil carbon storage after cultivation. Soil Sci. 142,
279-288.
Onal I., Aykas E.: 1999: Effects of different tillage, seeding and weed control
methods on weed population in wheat production. 7. International Congress on
Mechanization and Energy in Agriculture , 26-27 May Adana- TURKEY
Özsert İ., Kara M.: 1987: Kuru tarım tahıl üretiminde değişik toprak işleme-ekim
sistemleri ve enerji gereksinimleri. 3th International congress on agricultural
mechanization and energy 26-29 October. Izmir. Proceediings, p. 238-245
Prechac-Garcia F., Ernst O., Siri-Prieto G., Terra J.A., 2004: Integrating no-till
into crop-pasture rotations in Uruguay. Soil & Tillage Research 77, 1-13.
Suzer S. 2003: A study on the possibilities of applying reduced tillage methods on
Wheat. Workshop on Conservation Tillage and Direct Seeding. İZMİR pp.108121
Yalçın İ., Doğan T., 2000: Pamuk tarımında değişik toprak işleme yöntemlerinin
toprak özellikleri, işgücü gereksinimleri ve verim parametrelerine etkileri üzerine
bir araştırma. Tarımsal Mekanizasyon 19. Ulusal Kongresi 1 -2 Haziran 2000.
Erzurum. Bildiri Kitabı sayfa 115-120.
Yalçın H., Çakır E., Gülsoylu E., Keçecioğlu G., 2001: Tohum yatağı
Hazırlamada Uygulanan Farklı Toprak İşleme Yöntemleri Üzerine Bir Araştırma.
E.Ü.Ziraat Fakültesi Dergisi 38 (1), s. 71-78, Bornova-Izmir (Makale-Araştırma).
Yalçın H., Çakır, E., Akdemir, H., Öcel, T., Soya, H., 2003: Doğrudan Ekim Ve
Dipkazan Uygulamalarının İkinci Ürün Mısırda Verime Etkileri. Tarımsal
Mekanizasyon 21. Ulusal Kongresi. 3-5 Eylül 2003, KONYA, Bildiriler Kitabı,
S:167-171.
Yalcin H., 1998: A study on investigation of the suitable tillage methods in
second crop maize for silage. Ph.D.Thesis. Institute of Natural and Applied
Science, Ege University, İzmir, p. 136.
Yalcin H., E. Cakir H. Akdemir T. Öcel H. Soya. 2003: No-till planting of maize
as a second crop and subsoil effect on yield. 21. National Congress on
Mechanization and Energy in Agriculture. 3-5 September 2003, KONYA,
Proceedings, pp.167-171.
Yalcin H., Aykas E. 2003: Possibilities of direct drilling of second crop cotton
ISTRO
(International Soil Tillage Research Organization) Conference,
Brisbane, Australia.
Yalçın H, Çakır E. 2005: Tillage Parameters and Economic Analysis of Direct
Seeding, Minimum and Conventional Tillage in Wheat. Journal of Agronomy 4
(4), ISSN 1812-5379, 329-332.
Yalçın H., Çakır E. 2006: Tillage Effects and Energy Efficiencies of Subsoiling
and Direct Seeding in Light Soil on Yield of Second Crop Corn for Silage in
Western Turkey, Soil & Tillage Research, Volume 90, Page:250-255.
Yalcin H., Aykas E., Çakır E. 2006: Reduced Tillage Systems on Second Crop
Cotton in Aegean Region. ISTRO – 17th Conference of the International Soil
Tillage Research Organization, August 28-Sept.3 2006 Kiel, GERMANY.
Zinn Y.L., Lal R., Resck D.V.S. 2005: Changes in soil organic carbon stocks
under agriculture in Brazil. Soil & Tillage Research 84, 28-40.
10
Chapter 2
PREDICTION OF THE EFFECTS
OF TILLAGE IN RELATION TO SOIL
ORGANIC MATTER CONTENT
Ewa A. CZYŻ, Anthony R. DEXTER
INTRODUCTION
It is well known that organic matter is a “good thing” in soil and that it
“improves” it. However, there is very limited information available about the effects of
soil organic matter on soil/machine interactions. In this paper, we present predictions
for the size distribution of clods and aggregates produced by tillage of soils having
different contents of organic matter.
The ideal seedbed has been described as having aggregates < 5 mm diameter
[Braunack and Dexter, 1989; Russell, 1973]. On the other hand, clods (i.e. large
aggregates) are considered to be a problem because they make the soil more
heterogeneous and uneven. Here, we define clods as aggregates larger than 50 mm
diameter. Here we use existing models to predict the soil break-up during tillage, and
we examine the effects of soil organic matter content on the amount of small
aggregates and clods produced during tillage.
This is done in two steps: firstly, we use measured results for the soil water
retention characteristics of a range of soils in an existing theory to obtain values of
the index of soil physical quality, S* [Dexter, 2004]. We then apply these results in an
existing empirical model for the soil crumbling that occurs during mouldboard
ploughing. This gives us predictions of the size distributions of aggregates produced
as a function of the content of organic matter.
We note that all of the results used in this work were obtained using either
undisturbed soil samples measured in the laboratory or real soil in real fields tilled
with real mouldboard ploughs.
MATERIALS AND METHODS
Soils
Soil samples were collected from fields in England, France and Poland that
contained “old” experimental plots (more than 25 years old). The English soils were
11
collected from the Highfield experiment at Rothamsted Research Station at
Harpenden (about 35 km north of the centre of London). The French soils were
collected from two sites: Boigneville (about 60 km south of the centre of Paris at the
Institut Technique des Céréales and Fourrages) and Feucherolles (about 30 km west
of the centre of Paris). The Polish soils were from Grabów (about 20 km west from
Pulawy).
With the English soil, 5 treatments were measured:
(a) permanent fallow which has been cultivated several times each year to kill
weeds.
(b) permanent arable which has been cropped annually with cereals since 1949.
(c) rotation. This site has been maintained since 1949 in a 3-year grass-clover, 3
year cereal rotation.
(d) reseeded grass. This plot was ploughed initially in 1948 and reseeded with
grass shortly afterwards.
(e) permanent grass. This is the original, unbroken grassland that used to cover
the entire field.
With the French soil, the treatments were started in 1970. Two treatments
were measured:
(f)
tilled. This involves mouldboard ploughing annually to a depth of 25 cm.
(g) direct drilled. In this treatment, the soil is disturbed only in the seed row to about
5 cm depth.
In both treatments, the crop rotation is wheat – maize.
With the Polish soil, an experiment on crop rotations and fertilization was
sampled. The experiments were started in about 1975. Two treatments were
measured:
(h) low inputs of organic matter. The crop rotation is: potatoes, winter wheat, spring
barley and maize.
(i)
high inputs of organic matter. The crop rotation is potatoes, winter wheat +
mustard, spring barley + clover, and red clover + grass.
Both treatments had conventional tillage with mouldboard ploughing.
The particle size distributions of the soils were measured by the sedimentation
(hydrometer) method. The contents of organic carbon were measured by wet
oxidation.
The water retention characteristics of the soils were measured using standard
methods. These involved drying soil samples from saturation to a range of pore water
suctions on sand tables with hanging water columns and in ceramic pressure plate
extractors. The range of pore water suctions, h, applied was from 10 hPa to 15000
hPa. This range corresponds to pF1 to pF4.2 where pF is defined as the logarithm
(base 10) of the pore water suction in hPa. For the French soils, samples were also
equilibrated over saturated solutions of KCl, NaCl, NaBr, MgCl2 and LiCl to give
values of pF = 5.34, 5.58, 5.85, 6.18 and 6.47, respectively.
Tillage was done over a range of water contents in the field in Hungary and in
Sweden. Large samples of tilled soil were analysed by sieving. The results were
expressed as the proportions by mass of clods > 50 mm diameter for the Hungarian
soils and proportions > 50, >10 and >5 mm diameter for the Swedish soils. Details of
the field experiments have been published previously [Dexter and Birkás, 2004;
Keller et al., 2007].
12
Complexed and non-complexed organic matter
Crop residues, when incorporated into soil, become decomposed by microbes
to form the soil organic matter. This comprises humic acids and a range of other
compounds. The soil organic matter (OM) interacts with the clay particles (< 2 µm) to
form a complex that is relatively stable in water. It has been found that unit amount
(by mass) of organic carbon (OC) forms a complex with m = 10 units of clay (Dexter
et al., 2008). In this paper, we assume that OM = 1.724 OC.
If the soil has a small content of OC (or OM), then all the OC will be
complexed and some of clay will not be complexed. On the other hand, if the soil has
a large content of OC, then all of the clay will be complexed but not all of the OC will
be complexed. To distinguish between these possibilities, algorithms are necessary
because continuous functions are not possible.
The amount of complexed organic carbon, COC, can be calculated for each
soil using the following algorithm:
clay 
 clay 

[
]
COC = IF OC <
THEN
OC
ELSE
 m 
m 



(1)
and the amount of non-complexed organic carbon, NCOC, using
NCOC = IF [(C − COC ) > 0]THEN [C − COC ]ELSE [0]
(2)
It is possible to quantify the amounts of complexed and non-complexed clay in
a similar way to the amounts of complexed and non-complexed carbon given by
Eqs.(1 and 2), above. The amount of complexed clay, CC, is given by
CC = IF [mOC < clay ]THEN [mOC ]ELSE [clay ]
(3)
and the amount of non-complexed clay, NCC, is given by
NCC = IF [(clay − CC ) > 0]THEN [clay − CC ]ELSE [0]
(4)
The complex (COC + CC) is relatively stable in water, has a low density and is
important in controlling several valuable soil physical properties [Dexter et al., 2008].
The NCOC is associated with soil hydrophobicity (or non-wetting behaviour) as found
by de Jonge et al. [2009]. The NCC is associated with the soil content of readilydispersible clay, RDC (Dexter et al, 2008). RDC is associated with the instability of
soil in water and the hard-setting phenomenon.
Index of soil physical quality, S*
The water retention data, measured as described above, in the range of 1.0 <
pF < 4.2 were fitted to the Groenevelt and Grant [2004] equation:
  −k
0
w = k1 exp

n

  ( pF0 )
13


 − exp − k 0

 ( pF )n






(5)
in which k1, k0 and n are adjustable parameters. This equation has a sound basis in
thermodynamic theory [Groenevelt and Bolt, 1972; Groenevelt and Grant, 2004] and
is for systems in thermodynamic equilibrium. Additionally, for the French soils, the
values of water content obtained over saturated salt solutions were included in the
curve fitting. Eq. (5) states that the water content will be zero at pF0. The value of pF0
has been determined by the authors to be 6.6 [Dexter et al., 2010].
Eq. (5) gives the water content at saturation as
 
k0 

wsat = k1 exp
  ( pF0 )n 


(6)
Eq. (5) has an inflection point (a point where the curvature = zero) at:
 nk 
pFi =  0 
 (n + 1) 
1n
(7)
  −k 

0  − exp − (n + 1) 
wi = k1 exp
  ( pF0 )n 
 n 

(8)
The slope the curve of w plotted against pF has a slope at the inflection point given
by:
 n + 1
dw
− (n + 1)

= −k1 exp
* nk 0 


d ( pF )
n 

 nk 0 
((n +1) / n )
(9)
The slope at the inflection point when w is plotted against ln(h) has been used
as an index of soil physical quality, S. In previous work, S was calculated from fitted
parameters of the van Genuchten [1980] water retention equation. In this present
study, the slope is calculated from the fitted parameters of the Groenevelt and Grant
[2004] equation, as in Eq. (9) above. This procedure gives slightly different values for
the slope at the inflection point which we designate S*:
−1
dw
S* =
*
ln(10) d ( pF )
(10)
In Eq. (10) the minus sign is added simply to make values of the index
positive. There are only small difference in the values of S and S* calculated as
described above. When we compared the values obtained previously for 8 soils, we
obtained:
S = 0.0029 + 0.954S*, r = 0.984, p < 0.0001
(±0.0030) (±0.070)
(11)
A fuller description of S and of the soil physical behaviour associated with
different values of S is given in Dexter and Czyż [2007]. For the present purposes, it
14
is sufficient to know that values of S > 0.035 are associated with good soil physical
properties whereas values of S < 0.035 are associated with poor soil physical
properties. However, the changes in behaviour are progressive and there is no
sudden change at S = 0.035. This value is empirical and is based on observations
and experience of soil behaviour in the field. We have assumed that values of S and
S* are equivalent and that the same categories of soil behaviour are associated with
the same values of S*.
Model for soil crumbling
Tillage experiments were done in the field in Hungary [Dexter and Birkás,
2004] and in Sweden [Keller et al., 2007]. Five different soils were used in Hungary
and four different soils in Sweden. In these experiments, tillage by mouldboard
ploughing was done over a range of different water contents. The structure of the
resulting tilled soil was analyzed by sieving to obtain the size distribution of the
aggregates and clods produced. Also, the values of the index S were obtained for
these soils.
The results showed clearly that the water content given by Eq. (8) is the
optimum water content for tillage. That is, tillage at this water content produces the
minimum amount of clods (> 50 mm diameter) and the maximum amount of small
aggregates. When tillage is done at the optimum water content, then the amounts of
different sizes of aggregates produced may be estimated from Fig. 1. The results for
clod production from Hungary and Sweden were almost identical even though the
soils had very different genetic origins.
1.0
P > x mm
0.8
0.6
x=5
0.4
x = 10
0.2
0.0
0.00
x = 50
0.05
S*
0.10
0.15
Figure 1. Size distribution of aggregates produced by tillage as a function of S*
[Keller et al., 2007]
Here, we focus on the proportion, P(> x) of the tilled soil that is in the form of
clods or aggregates (> x mm diameter) after tillage at the optimum soil water content.
15
RESULTS AND DISCUSSION
Characterization of the experimental soils
The compositions of the experimental soils are given in Table 1. The
complexed and non-complexed components of the soil organic carbon and the clay
as obtained using Eqs.(1-4) are given in Table 2.
Table 1. Compositions of the experimental soils
Site and treatment
sand
silt
-1
(g kg )
(g kg-1)
England, Rothamsted
bare fallow
90
670
cont cereals
130
630
rotation
110
640
re-seeded grass
110
630
perm grass
110
670
France, Boigneville
tilled
80
660
direct-drilled
78
686
Poland, Grabów
low OM input
707
265
high OM input
722
255
clay
(g kg-1)
organic matter
(g kg-1)
240
240
250
260
220
19.0
26.0
36.0
48.0
55.0
260
236
22.9
49.8
28
23
11.2
13.5
Table 2. Contents of organic carbon (OC), complexed organic carbon (COC), noncomplexed organic carbon (NCOC), complexed clay (CC) and non-complexed clay
(NCC)
Site and treatment
OC
COC
NCOC
CC
NCC
-1
-1
-1
-1
(g kg ) (g kg ) (g kg ) (g kg )
(g kg-1)
England, Rothamsted
bare fallow
11
11
0
110
130
cont cereals
15
15
0
150
90
rotation
21
21
0
210
40
re-seeded grass
28
26
2
260
0
perm grass
32
22
10
220
0
France, Boigneville
tilled
13.3
13.3
0
133
127
direct-drilled
28.9
23.6
5.3
236
0
Poland, Grabów
low OM input
6.5
2.8
3.7
28
0
high OM input
7.8
2.3
5.5
23
0
The parameters of the Groenevelt and Grant water retention equation, as
obtained by non-linear curve fitting are given in Table 3. Calculated values of the
index of soil physical quality, S*, are also presented in Table 3. The water retention
curve over the whole range of water contents from saturation to complete dryness for
Boigneville B soil is shown in Fig. 2, as an example.
16
Table 3. Parameters of the Groenevelt and Grant [2004] water retention equation
(Eq.5) and values of the index of soil physical quality, S*
Site and treatment
k1
k0
n
S*
England, Rothamsted
bare fallow
cont cereals
rotation
re-seeded grass
perm grass
France, Boigneville
tilled
direct-drilled
Poland, Grabów
low OM input
high OM input
2.5494
3.0205
0.8456
1.0783
1.0480
7.556
6.642
5.578
4.461
4.555
0.586
0.530
0.887
0.816
0.844
0.0203
0.0233
0.0279
0.0391
0.0396
0.6824
0.4377
6.838
11.920
1.021
2.049
0.0246
0.0474
0.1849
0.2108
12.418
11.894
2.588
2.777
0.0310
0.0406
Boigneville B
-1
w (kg kg )
0.3
0.2
0.1
0.0
1
2
3
pF
4
5
6
7
Figure 2. Water retention curve for the Boigneville B soil. The points shown as
squares were measured using pressure plate equipment, and the points shown as
triangles were measured by equilibration over saturated salt solutions. The smooth
curve is the fit to the Groenevelt and Grant [2004] equation
Predictions of the effects of soil organic matter content on the effects of tillage
When we apply the values of S* given in Table 3 to the results in Fig. 1, we
obtain the predictions given in Table 4. The gives estimates of the proportions of
different size ranges of aggregates obtained by tilling the different soils at their
optimum water contents. It can be seen that, when the content of organic matter is
higher, then the amounts of clods produced is smaller and the amounts of small
aggregates produced is greater.
17
Table 4. Values of S* and predicted proportions, P(<x) of the soil in aggregates or
clods smaller than x mm diameter after tillage by a single pass of a mouldboard
plough at the optimum water content
Site and treatment
S*
P(<50)
P(<10)
P(<5)
England, Rothamsted
bare fallow
cont cereals
rotation
re-seeded grass
perm grass
France, Boigneville
tilled
direct-drilled
Poland, Grabów
low OM input
high OM input
0.0203
0.0233
0.0279
0.0391
0.0396
0.503
0.578
0.692
0.970
0.982
0.239
0.274
0.329
0.461
0.466
0.134
0.153
0.184
0.257
0.261
0.0246
0.0474
0.610
1.000
0.290
0.558
0.162
0.312
0.0310
0.0406
0.769
1.000
0.365
0.478
0.204
0.267
The lines in Fig. 1 that give the proportions of clods > x mm as a function of S*
are given by
P(> 50 mm) = 1.0 – 28.6S*
(12)
P(> 10 mm) = 1.0 – 11.8S*
(13)
P(> 5 mm) = 1.0 – 6.58S*
(14)
We note that the proportion of small aggregates (e.g. < x mm diameter) is
given by
P(< x mm) = 1 – P(>x mm).
(15)
P(<5 mm) = 6.58S*, for values of S* < 0.15
(16)
For example,
Some predictions from the above equations are given in Table 4. As discussed
in the introduction. a good seedbed should be composed of aggregates < 5 mm
diameter. The results in Table 4 show that such a seedbed cannot be produced by a
single pass of a mouldboard plough even when the soil is in good physical condition
(i.e. with a high value of S* ). This is consistent with experience and is the reason
why secondary tillage (e.g. with harrows) is always performed.
To study the effects of the amounts and types of non-complexed material,
NCM, we have defined
NCM = NCOC – NCC/10
(17)
The value of 10 in Eq. (17) is the value of m in Eqs. (1-4). It is needed here to
put NCOC and NCC onto effectively the same numerical scale because OC has
18
approximately 10 times the effect on soil physical properties as clay. The effect of
NCM on S* is shown in Fig. 3. The regression equation is
S* = 0.034 + 0.00096NCM, r = 0.88,
(±0.002) (±0.00019)
p < 0.002
(18)
The value of S* at the intercept (when NCM = 0) is given by Eq. (18) as S* =
0.034 ± 0.002 which is not significantly different from the empirical value of S = 0.035
for the boundary between “good” and “poor” soil physical quality as presented by
Dexter [2004] and Dexter and Czyż [2007]. At NCM = 0, all the clay and all the OM
are complexed.
0.05
0.04
S*
0.03
0.02
-15
NCC/10
-10
-5
NCOC
0
NCM
5
10
Figure 3. Soil physical quality, S*, as a function of non-complexed material, NCM
We have used Eqs. (12, 16, 17, 18) and the results in Table 2 to predict the
proportions of clods (P>50 mm) and small aggregates (P<5 mm) produced by
ploughing as a function of the soil content of non-complexed material, NCM, as
defined in Eq. (17). These predictions are given in Fig. 4. The results for clods are
consistent with the results in Fig. 1 and with the experimental results of Dexter and
Birkás [2004] and Keller et al., [2007]. The results for small aggregates show that a
single pass of a mouldboard plough produces < 30% of small aggregates even with
soil in the best physical condition.
19
Proportion, P
0.5
NCOC
NCC/10
0.4
0.3
0.2
P<5 mm
0.1
P>50 mm
0.0
-15
-10
-5
0
NCM
5
10
Figure 4. Predicted proportions of clods (> 50 mm diameter) and small aggregates
(< 5 mm diameter) produced as a function of the soil content of non-complexed
material, NCM
We conclude that the new measure of non-complexed material, NCM, enables
soils that are poor and rich in organic matter to be analysed on the same rational
basis and physical scale which can then be used to predict the size distributions of
clods and aggregates produced by tillage.
ACKNOWLEDGEMENTS
The authors would like to thank Dr. Guy Richard of INRA, France and Dr. D.S.
Powlson of Rothamsted Research, England for their help.
REFERENCES
1. Braunack, M.V. and Dexter, A.R. 1989: Soil aggregation in the seedbed: a
review. II. Effect of aggregate sizes on plant growth. Soil and Tillage Res. 14,
281-298.
2. de Jonge, L.W., Moldrup, P. and Schjonning, P. 2009: Soil infrastructure,
interfaces and translocation processes in inner space (“Soil-it-is”): towards a
road map for the constraints and crossroads of soil architecture and biophysical
processes. Hydrol. Earth Syst. Sci. 13, 1485-1502.
3. Dexter, A.R. 1988: Advances in characterization of soil structure. Soil and
Tillage Res. 11, 199-238.
4. Dexter, A.R. and Birkás, M. 2004: Prediction of the soil structures produced by
tillage. Soil and Tillage Res. 79 (2), 233-238.
5. Dexter, A.R. 2004: Soil physical quality. Geoderma 120, 201- 239.
6. Dexter, A.R. and Czyż, E.A. 2007: Applications of S-theory in the study of soil
physical degradation and its consequences. Land Degradation and
Development 18(4), 369-381.
20
7.
8.
9.
10.
11.
12.
13.
Dexter, A.R., Richard, G., Arrouays, D., Czyż, E.A., Jolivet, C. and Duval, O.
2008: Complexed organic matter controls soil physical properties. Geoderma
144, 620-627.
Dexter, A.R., Richard, G., Czyż, E.A. and Davy, J. 2010: Clay dispersion from
soil as a function of antecedent water potential. Soil Sci. Soc. Am. J.
(submitted).
Groenevelt, P.H. and Bolt, G.H. 1972: Water retention in soil. Soil Sci. 113, 246253.
Groenevelt P.H. and Grant, C.D. 2004: A new model for the soil water retention
curve that solves the problem of residual water contents. Eur. J. Soil Sci. 55,
479-485.
Keller, T., Arvidsson, J. and Dexter, A.R. 2007: Soil structures produced by
tillage as affected by water content and the physical quality of soil. Soil and
Tillage Res. 92 (1-2), 45-52.
van Genuchten M.Th., 1980: A closed-form equation for predicting the hydraulic
conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44, 892–898.
Russell, E.W. 1973: Soil Conditions and Plant Growth. 10th edition. Longman,
Green and Co., London, 849pp.
21
Chapter 3
DETERMINATION OF SOIL STABILITY IN
RELATION TO ECOLOGICAL
AGRICULTURE
Ewa A. CZYŻ, Anthony R. DEXTER, Olga P. GAŢE
INTRODUCTION
Ecological (organic) farming must protect the environment and at the same
time produce good soil conditions for crop production. Structure is a complex
property of the soil that is related to many agronomic and environmental processes
[Díaz-Zorita et al., 2002] and has been the subject of study for several scientists [e.g.
Dexter, 1988, 1997; Kay, 1990; Gaţe et al., 2004, 2005, 2006a, 2006b; Czyż and
Dexter, 2009]. In ecological agriculture and horticultural production, the soil structure
plays an important role for plant development, soil water balance and soil workability
[Dexter, 1988; Watts et al., 1996b]. Moreover, the soil must have a structure that will
persist in time. The ability of the soil structural units to resist changes after the
application of a stress (i.e. dry or moist fragmentation, compaction, dispersion etc.) is
known as soil structural stability [Dexter, 1988; Kay, 1990; Díaz-Zorita et al., 2002].
Dexter [1988] described two principal types of structural stability: (a) the ability of the
soil to retain its structure under the action of water; and (b) the ability of drier soil to
retain its structure under the action of external mechanical stresses.
In the literature there is a wide range of methods used to assess soil structural
stability which have been developed during the past six decades so that we are now
in a position to give a quantitative, physically-based assessment of structural stability
as influenced by the action of both water and mechanical stresses and over a range
of structural size scales [Watts et al., 1996b]. However, recent studies have shown
that it is the content of readily-dispersible clay which is the most appropriate measure
of structural condition because clay dispersion is indicative of soil instability [Czyż et
al., 2002; Dexter, 2002; Watts and Dexter, 1997; Watts et al., 1996a,b].
Clay particles together with other soil colloids constitute the basic binding
agents of the soil micro-structure [Dexter, 1988, 2002], and therefore the amount of
readily-dispersible clay in soils influences the stability of micro-aggregates and
aggregates [Dexter, 2002]. Furthermore, the content of readily-dispersible clay in soil
has been shown to affect a range of physical and mechanical properties [Kay and
Dexter, 1992; Watts and Dexter, 1997; Watts et al., 1996a], and with complexed clay
22
and non-complexed clay [Dexter et al., 2008; Schjønning et al., 2009] and as a
consequence Kay and Dexter [1990] suggested that the values of dispersible clay
should be part of a minimum data set routinely collected for assessing the impact of
new soil, crop and water management practices on soil properties. Dexter et al., 2008
showed that non-complexed clay is more easily dispersed in water than clay
complexed with OM.
The most important factor that affects the amount of dispersed clay is,
undoubtedly, water. When soil comes into contact with water, the colloids within it
swell causing it to become weaker and more susceptible to damage [Dexter, 1999].
During swelling, clay particles become more widely spaced and finally they can
become dispersed and mobile both within the soil and the wider environment. The
swelling ratio depends on several factors such as the type of cations adsorbed on the
surfaces of clay particles and the quantity and quality of organic matter [Dexter, 1999,
2002]. Calcium ions adsorbed on the clay surfaces favour stability by keeping the
clay flocculated whereas sodium ions encourage swelling and dispersion. The
influence of organic matter on clay dispersion is manifested in two ways. The first is
by acting as a ‘glue’ holding the particles together, and the second is by reducing the
rate of wetting due to its hydrophobic characteristics [Dexter, 1999]. However,
dispersion of clay depends also on the electrolyte concentration of the soil water
[Dexter, 1988, 2002]. Clay dispersion can be prevented by maintaining the electrolyte
concentrations above certain critical values that keep the clay flocculated [Raine and
Loch, 2003]. This principle is used in the reclamation of saline soils by leaching.
The amount of clay that is dispersed may also depend on soil moisture
conditions prior to beginning dispersible clay measurements. Kay and Dexter [1990],
in their study concerning the influence of antecedent water content on the
dispersibility of clay on a red-brown earth from southern Australia, have found that
spontaneously-dispersed clay and mechanically-dispersed clay increased with
increasing antecedent water content (corresponding to potentials ranging from –10 to
–0.3 kPa). The authors also hypothesized that, for the soils that they were using, the
percentages of spontaneously-dispersed clay and mechanically-dispersed clay are
controlled by the specific area of exposed aggregates surfaces and the dispersibility
of clay per unit specific aggregate surface area. The authors have evaluated this
hypothesis using aggregates of different sizes collected from two different
management practices, namely wheat-fallow and continuous pasture rotations. It was
found that the amount of dispersed clay per unit surface area of aggregates
increased with increasing size of aggregates. In addition, the combined effect of
increasing initial aggregate size and increasing antecedent water content was
greatest on the least stable soil under a wheat-fallow crop rotation [Kay and Dexter,
1990].
The amount of clay that disperses may also vary with the amount of energy
applied. Watts et al., [1996a] have quantified the mechanical energy inputs to soil
during tillage in field experiments in order to assess the subsequent aggregate
disruption by measuring the quantity of mechanically-dispersible clay. Their
investigations have shown that increasing the tillage energy resulted in higher clay
dispersion for a given soil water content. Also they reported that mechanical energy
inputs had no significant effect on clay dispersibility when soil water content was less
than the plastic limit. However, when the soil is wetter than the plastic limit, then the
content of readily-dispersible clay increases with both water content and the input of
specific mechanical energy [Dexter, 2002].
23
Changes in dispersible clay due to changes in management practices have
been suggested by Kay and Dexter [1990] to be manifested earliest in the weakest
failure zones. Such zones have a distribution of strengths which is related to water
content and processes which change pore characteristics and/or the cementation
between structural units [Kay and Dexter, 1992]. Cementing materials may be of
organic or inorganic form. Inorganic materials include particles of dispersed clay
which become effective on drying by precipitation or flocculation at intergranular
points at low water contents [Kay and Dexter, 1992].
Soils with high contents of readily-dispersible clay (RDC) in the presence of
water may experience the collapse of their structure with the consequent loss of the
inter-aggregate pores and soil homogenization [Czyż et al., 2002]. Problems
associated with clay dispersion include: anaerobic soil that is unsuitable for plant root
growth, reduced infiltration of water with associated risk of run-off, flooding and
erosion [Dexter and Czyż, 2000; Czyż et al., 2002]. Also, such soils can form crusts
at the surface or can hardset, both of which characteristics are associated with poor
crop emergence and increased energy requirement for tillage [Dexter and Czyż,
2000; Czyż et al., 2002].
The aim of the research was determination of the content of readily-dispersible
clay in soil samples collected from experimental fields at Huta, Kępa and Osiny which
belong to Institute of Soil Science and Plant Cultivation - National Research Institute
(IUNG-PIB) Puławy, and the investigation of the relationships between readilydispersible clay and other soil physical properties.
MATERIAL AND METHOD
Soil samples used in these measurements were collected from ecological
farms at three different locations around Puławy (Huta, Kępa and Osiny) from the
arable layer (0-20 cm) and sub-soil (20-35 cm). Spring wheat and spring barley was
grown in traditional tillage with straw incorporation based on mouldboard ploughing
(to 25 cm depth) and traditional soil management equipment. The samples were
collected at different dates from field plots where spring cereals (spring wheat and
spring barley) were grown. Samples collected for the measurement of readilydispersible clay were stored in sealed plastic bags and were kept at their prevailing
water content until they were needed for analysis.
Particle size distribution and organic matter measurements
The particle size distributions of the soils studied were determined by
Casagrande’s hydrometer method as modified by Prószyński [Lityński et al., 1976].
The organic matter content of the soils was measured by wet oxidation by the Tiurin
method [Ostrowska et al., 1991].
Soil bulk density and water content by oven-drying
Soil samples were collected from the field for determination of dry bulk density,
(Mg m-3), and gravimetric water content (kg kg-1). The soil samples were collected
from soil pits that were dug in the field in the different treatments. For the above, four
replicates were collected from each of the following depth layers: 0-20 cm and 20-35
cm. The depth of sampling was selected to coincide with the depths of layers
corresponding with the different tillage practices. Undisturbed soil cores were taken
24
by pushing stainless steel cylinders of 100 mL vertically into the soil using a hammer,
and for each layer and location 4 replications were sampled. The cylinders were then
closed with stainless steel end caps and were placed in polythene bags to prevent
water loss. Dry bulk density and water content of the soils were measured in the
laboratory by weighing the soil samples before and after drying at 105°C in an oven
for 48 hours. The determinations were done immediately after returning from the field
so that further water loss was avoided. The soil water content values were then
calculated volumetrically and gravimetrically. The dry bulk density was calculated as
the mass of dry soil per unit volume of moist soil.
Readily-dispersible clay (RDC)
The method used for determination of readily-dispersible clay, RDC,
(NTU/(g L)-1), is that described by [Czyż et al., 2002] and Dexter and Czyż [2000],
and is rather similar with that described by Kay and Dexter [1990], Watts and Dexter
[1997] and Watts et al. [1996a,b], but was adapted for Polish sandy soils.
The amount of readily-dispersible clay in water using a standard dilution and
shaking procedure was measured using turbidimetry. For each soil sample 10 subsamples were used. About 5g of soil were weighed and placed in 150ml plastic
bottles and then were shaken with 125ml distilled water in a standardized way (four
complete rotations end-over-end). The bottles were then allowed to stand for 18
hours to allow the larger particles to sediment, leaving only dispersed colloids (in this
case, mostly clay) in suspension. A 30ml sample of this suspension was extracted by
pipette from the centre of each bottle and was transferred to a glass turbidity cell.
Also, 10 sub-samples of the same soil sample were prepared for
determination of the total clay content. In this case, about 18-20g of soil were
weighed and placed in 750-100ml Berzelius glasses and were stirred with 500ml
distilled water for 30min. Then 125ml from this suspension was poured into the
plastic bottles and allowed to stand for 18 hours to allow the larger particles to
sediment, leaving only colloids (clay) in suspension. A 30ml sample of this
suspension was extracted by pipette from the centre of each bottle and was
transferred to a glass turbidity cell.
The turbidity was measured by light-scattering using a Hach 2100 AN ratio
turbidimeter. (Fig. 1-3). Turbidity values are linearly proportional to the concentration
of colloids (clay) in suspension [Dexter and Czyż, 2000]. The turbidimeter readings
were expressed as NTU (Nephelometric Turbidity Units) and were normalized by
dividing by the concentration of the original soil in the water to give NTU/(g L-1). The
mass of soil was corrected to dry mass for this calculation.
0
90 detector
Light path diagram
Transmitted
light
detector
Lamp
Lens
Sample cell
Figure 1. Schematic diagram showing the principle of operation of the ratio
turbidimeter
25
Figure 2. Model HACH 2100 AN Turbidimeter for determination of the content of
readily-dispersible clay, RDC
Figure 3. Photographs of bottles with soil suspensions (top) and measurement the
content of readily-dispersible clay, RDC with HACH 2100 AN Turbidimeter (bottom)
To determine the effects of air-drying of soil samples, some samples were airdried for comparison with samples that had been stored at their field water content.
To test the reversibility of the effects of air-drying, some air-dried samples were
26
rewetted by saturating them and then storing them on a sand table at –40 hPa water
suction for up to two weeks before measurement of content of readily-dispersible
clay.
RESULTS AND DISCUSSION
Physical characterization of the soils used in these investigations is presented
in Table 1.
Table 1. Physical characteristics of the soils used in these investigations
Location Layer*
Sand
Silt
Clay
FAO/USDA Organic
(>50 µm) (50-2µm) (<2 µm) soil texture matter
class
(kg kg-1) (kg kg-1) (kg kg-1)
(kg kg-1)
Huta
top-soil
0.65
0.32
0.03
sandy
0.0122
loam
sub-soil
0.64
0.34
0.02
sandy
0.0017
loam
Kępa
top-soil
0.10
0.67
0.23
silt loam
0.0222
sub-soil
0.11
0.65
0.24
silt loam
0.0170
Osiny
top-soil
0.70
0.26
0.04
sandy
0.0152
loam
sub-soil
0.70
0.26
0.04
sandy
0.0122
loam
Bulk
density
(Mg m-3)
1.574
1.683
1.438
1.614
1.633
1.709
*/ Layer: top-soil (depth 0-20 cm), and sub-soil (depth 20-35 cm)
The amount of readily-dispersible clay was measured on soil samples that had
been stored at the prevailing water content. The mean values of the contents of
readily-dispersible clay, as well as of total clay of the three soils are given in Table 2.
For the soils investigated, the mean content of clay that is readily-dispersible was
found to be 0,416 g/(100g soil). This value is slightly smaller than that found by Czyż
et al., [2002] for 210 Polish soils (0,48 g/100g soil).
In addition, there is a clear tendency of lower stability with increasing depth in
the soil profile. This can be seen by the greater values of the content of readilydispersible clay of sub-soils. Also, it seems that this increase is greater in sandy soils
(Huta and Osiny), where the amount of clay which is readily-dispersible in sub-soils is
almost twice that in top-soils. These significant increases can be attributed to higher
bulk densities and lower organic matter contents of the sub-soils, and probably
because the sub-soils don’t dry as much as top-soils. Watts et al., [1996b] and Watts
and Dexter [1997] have been also reported that the content of readily-dispersible clay
is affected by compaction and organic matter respectively.
According to the classification of sensitivity of Polish agricultural soils to
destruction, as expressed by the content of readily-dispersible clay (g per 100g of
soil), these three soils that were analized are classified as sensitive and highly
sensitive to structural damage.
27
Table 2. Mean values of the contents of readily-dispersible clay, (RDC) and total clay,
(TC) measured on soil samples collected on different dates
Location
Layer*
RDC
TC
RDC
(NTU/(g/L))
(NTU/(g/L))
(g/100g soil)
Huta
top-soil
0.515
5.948
0.260
sub-soil
1.736
7.911
0.439
Kępa
top-soil
1.234
78.678
0.359
sub-soil
1.774
94.864
0.428
Osiny
top-soil
1.863
19.505
0.366
sub-soil
1.125
14.889
0.647
Mean
1.375
36.966
0.416
s.d.
0.520
39.219
0.130
s.e.
0.164
12.402
0.041
*/ Layer: top-soil (depth 0-20 cm), and sub-soil (depth 20-35 cm)
Because the soil from Kępa had a significantly higher clay content than the
two sandy soils from Huta and Osiny, the regression analysis of the results in the
following discussion was done only for the latter two soils.
A range of organic matter contents produced in soils by different management
practices had a significant effect on the dispersibility of clay particles. The results
which are plotted in Fig. 4 show a negative correlation between the amount of
dispersible clay and the organic matter content of the soils. The regression line from
Fig. 4 is given by the following equation:
RDC = 1.408 – 0.766 · OM
r2 = 0.62;
p = 0.00031
(1)
(±0.210) (±0.161)
where RDC is the content of readily-dispersible clay (g per 100g of soil) and OM
is the organic matter content (%).
Soil hydrophobicity can be imparted by organic matter and this can reduce
both the wetting rates and the subsequent soil disruption by slaking and dispersion
[Dexter, 2002]. Lower values of the organic matter content have been found by Watts
et al. [1996b] and Watts and Dexter [1997] to be associated with greater amounts of
dispersed clay in Romanian soils and UK soils respectively. Organic matter not only
stabilizes soil through reducing the content of RDC, but it also reduces the sensitivity
of soil to destabilization by mechanical energy inputs [Watts et al., 1997].
A similar tendency of soil destabilisation (as measured by turbidimetry) due to
lower contents of organic matter has been reported also by Dexter and Czyż [2000]
in a Polish sandy soil.
28
readily-dispersible clay (RDC),
g/(100g soil)
2,25
2,00
1,75
1,50
1,25
1,00
0,75
0,50
0,25
0,00
0,0
0,2
0,4
0,6
0,8
1,0
1,2
1,4
1,6
1,8
organic matter (OM), %
Figure 4. Variation of the content of readily-dispersible clay as a function of organic
matter for the sandy soils from Huta and Osiny
Another aspect that we analized was the relationship between the amount of
readily-dispersible clay and the total clay content of these soils, both being measured
by using the same technique (turbidimetry).
The mechanical energy inputs to soil during tillage can destabilize soil
structure through increasing the content of readily-dispersible clay. If the soil
disturbed by tillage contains more clay then there is an increase in the specific
surface area of aggregate surfaces that will be exposed to further disruption.
Moreover, these energy inputs by tillage can weaken the bonds between clay
particles, especially when tillage is performed on soils that are wetter than the plastic
limit [Dexter, 2002]. Fig. 5 shows the variation of the amount of readily-dispersible
clay as a function of the total clay content, and the regression line is given by the
equation:
p < 0.0001
(2)
RDC = - 0.868 + 0.142 · TC
r2 = 0.84;
(±0.334) (±0.017)
where RDC is the content of readily-dispersible clay and TC is the total clay
content, both expressed as NTU/(g/L).
Usually, large values of the dispersibility means that the clay is loosely bound
in the soil aggregates and that the soil is susceptible to dispersion and cementation
during wetting and drying processes [Elmholt et al., 2000; Dexter, 2002].
Reorientation and hardening of the dispersed clay minerals can result in a dense and
mechanically-strong soil [Kay and Dexter, 1992]. With these mechanisms, the result
is the destruction of the existing soil structure, and as a consequence, this readilydispersible clay is expected to contribute to the agricultural problems which were
described in the introduction.
29
readily-dispersible clay (RDC),
NTU/(g/L)
3,5
3,0
2,5
2,0
1,5
1,0
0,5
5
10
15
20
25
30
total clay (TC), NTU/(g/L)
Figure 5. Effect of total clay content on the amount of readily-dispersible clay in
sandy soils from Huta and Osiny
As we mentioned earlier, the soil water content is the most important factor
affecting structural destabilisation as quantified by the amount of readily-dispersible
clay. The values of readily-dispersible clay after the measurements were completed
were plotted in Fig. 6, against their respective initial water contents and the analysis
of the results showed that there is a consistent increase in readily-dispersed clay with
increase in initial soil water content which can be described by:
RDC = 0.063 + 3.006 · wi
r2 = 0.54;
p = 0.00178
(3)
(±0.078) (±0.768)
where RDC is the readily-dispersed clay (g per 100g of soil) and wi is the initial
soil water content (kg per kg of soil).
readily-dispersible clay (RDC),
g/100g soil
0,55
0,50
0,45
0,40
0,35
0,30
0,25
0,20
0,06 0,07 0,08 0,09 0,10 0,11 0,12 0,13 0,14
-1
water content (wi), kg kg
Figure 6. Effect of prevailing water content on the dispersibility of clay particles of
sandy soils
30
The great sensitivity of readily-dispersible clay to water content is consistent
with observations on an Australian red-brown earth [Kay and Dexter, 1990] and two
soils in the UK [Watts et al., 1996a; Watts and Dexter, 1997].
High intensity tillage systems can increase the exposed surfaces of the
aggregates to further disruption under the action of water. Watts et al., [1996a]
showed that aggregates collected following tillage yield larger quantities of
dispersible clay than those collected prior to tillage. Also they found that these
differences appear to be larger the higher the water content at the time of tillage.
An important aspect concerning the determination of readily-dispersible clay is
related to the soil sample condition at the time of measurement. Gaţe [2004], Gate et
al., [2006a, 2006b] found that the contents of readily-dispersible clay measured on
air-dry soil samples were around 20 times smaller than when measurements were
made on samples that had been stored at field water content. Fig. 7 illustrates this
phenomenon, and the regression line is given by the following equation:
RDCwi = 20.5 RDCdry – 1.9
r2 = 0.66;
p = 0.008
(4)
(±5.6)
(±1.5)
where RDCwi is the readily-dispersible clay measured at field water content, and
RDCdry is the readily-dispersible clay measured on air-dried soil samples, both
in units of NTU/(g L-1).
RDCwi, NTU/(g/L)
10
8
6
4
2
0
0,0
0,1
0,2
0,3
RDCdry, NTU/(g/L)
0,4
0,5
Figure 7. Comparison of the values of readily-dispersible clay (RDC) obtained after
different conditions of sample storage. Values on the x-axis were obtained with soil
samples that had been air-dried whereas the values on the y-axis were obtained with
samples that had been stored at field water content [Gate, 2004]
If the soil is allowed to air dry, the cementing materials (e.g. dispersed clay)
will precipitate or flocculate at intergranular points at low water contents [Kay and
Dexter, 1992]. If this material deposition occurs near or at the ends of microcracks,
then the micro-cracks will be stabilized and this will result in an increase of the tensile
strength of the soil [Kay and Dexter, 1992]. Also, as soil dries, it is compressed by
effective stresses due in part to the pore water pressure and in part to the surface
tension in the water menisci. Such compression is non-reversible. Therefore, to
determine the physical properties of soil as they occur in the field, samples must
never be allowed dry to lower water contents than they have ever experienced in the
field.
31
The irreversibility of the effects of drying is shown by the fact that re-wetting of
air-dried soil for up to two weeks had no significant effect on the content of readilydispersible clay.
Fig. 8 illustrates the effect of different tillage systems on soil physical quality
[Czyż and Dexter, 2009]. The relationship between the index of soil physical quality,
S, and the content of readily-dispersible clay (RCD) has been published earlier by
Gaţe et al., [2004]. RDC has been found to be negatively-correlated with a measure
of soil physical quality, S [Gaţe, 2006]. In other words, soil with a small content of
RDC is usually of good physical quality (i.e. with high S) whereas soil with a large
content of RDC is usually of poor physical quality (i.e. with low S). Soils with high S
have been found to be more friable [Dexter, 2004], easier to till [Keller et al., 2007]
are not hard-setting [Dexter, 2004], have lower penetration resistance [Dexter and
Czyż, 2000] and have higher saturated hydraulic conductivity [Han, 2008]. Also Czyż
and Dexter [2009] showed on samples collected from a long-term field experiment on
a farm in Rogów with a silt loam soil, that reduced tillage and no-tillage reduced the
amount of RDC especially in the top layer in comparison with traditional tillage.
Reduced tillage and no-tillage also decreased the index of soil physical quality, S,
after the first year in comparison with traditional tillage. However, after 4 years, the
non-inversion plots (reduced tillage and no-tillage) showed some improvement in soil
physical quality and soil stability in water. Reduced tillage decreased the quantity of
readily-dispersible clay (RDC) and therefore increased soil stability, especially in the
top layer 5-10 cm in both soils in comparison with traditional tillage [Czyż and Dexter,
2009; Dexter and Czyż, 2000; Gaţe et al., 2004].
0,06
0,05
Values of S
0,04
0,03
0,02
0,01
0,00
2003
2004
2005
TT
RT
NT
2006
Years
S values = 0.035
Figure 8. Effects different tillage systems (traditional, TT; reduced, RT; zero-tillage,
NT) on soil physical quality values, S, at 5-10 cm depth in the first four years of use of
different tillage systems [Czyż and Dexter, 2009]
CONCLUSIONS
The turbidimetric method for measurement of clay dispersion is easy and
rapid. Because the amount of readily-dispersible clay has implications for a range of
agricultural and environmental problems, it can be a useful method for describing of
some important aspects of soil physical quality.
32
The amount of readily-dispersible clay (measured by turbidimetry) increases
with decreasing soil organic matter. This rate of increase rises with increasing soil
water content. The soils with greater total clay content have a greater specific surface
area and this can result in more clay being dispersed by low energy inputs.
Soil samples for measurement of RDC must never be allowed to dry to lower
water contents than they have experienced in the field or the physical properties will
be significantly and irreversibly changed.
Several authors shown that the water-dispersible clay content is smaller when
the soil OC content is larger [e.g. Watts and Dexter, 1997; Czyż et al., 2002].
Increasing amounts of readily-dispersible clay (RDC) have been shown to
contribute to destabilization of soil structure. This dispersible clay is expected to have
negative effects on other physical properties, such as soil aeration, water infiltration
and soil strength.
REFERENCES
1.
Czyż, E. A., Dexter, A. R. and Terelak, H. 2002: Content of readily-dispersible
clay in the arable layer of some Polish soils. In: Sustainable Land ManagementEnvironmental Protection. (Eds. Pagliai, M. and Jones R.). Advances in
Geoecology, 35, s. 115-124.
2.
Czyż, E.A. and Dexter, A.R. 2009: Soil physical properties as affected by
traditional, reduced and no-tillage for winter wheat. Int. Agrophysics, 23(4),
s. 319-326.
3.
Dexter A.R. 2004: Soil physical quality: Parts I, II, III. Geoderma, 120, s. 201239.
4.
Dexter, A.R. and Czyż, E. A. 2000: Effects of soil management on the
dispersibility of clay in a sandy soil. Int. Agrophysics, 14, s. 269-272.
5.
Dexter, A.R. and Czyż, E.A. 2007: Applications of S-theory in the study of soil
physical degradation and its consequences. Land Degradation and
Development, 18(4), s. 369-381.
6.
Dexter, A.R. 1988: Advances in characterization of soil structure. Soil Till. Res.,
11, s. 199-238.
7.
Dexter, A.R. 1997: Physical properties of tilled soils. Soil Till. Res., 43,s. 41-63.
8.
Dexter, A.R. 1999: Some thoughts on soil structure research. In: Proceedings of
the Soil Structure/Carbon Workshop. (Eds. Reynolds, W. D., Drury, C.F. and
Tan, C.S.), Leamington, Ontario, 23-25 August, pp. 11-23.
9.
Dexter, A.R. 2002: Soil Structure: The Key to Soil Function. In: Sustainable
Land Management-Environmental Protection. (Eds. Pagliai, M. and Jones, R.).
Advances in Geoecology 35, s. 57-69.
10. Dexter, A.R., Richard, G., Arrouays, D., Czyż, E. A., Jolivet, C. and Duval, O.
2008: Complexed organic matter controls soil physical properties. Geoderma
144, s. 620-627.
11. Díaz-Zorita, M., Perfect, E. and Grove, J.H. 2002: Disruptive methods for
assessing soil structure. Soil Till. Res., 64, s. 3-22.
12. Elmholt, S., Debosz, K., Munkholm, L.J. and Schjønning, P. 2000: Biotic and
abiotic binding and bonding mechanisms in soils with long-term differences in
management. Proceedings from Workshop on Soil Stresses, Quality and Care,
Aas, Norway, 10-12 April. DIAS Report No. 38. The Danish Institute of
Agricultural Sciences, s. 53-61.
13. Gaţe O.P., Czyż, E.A. and Dexter, A.R. 2004: Effects of readily-dispersible clay
on soil quality and root growth. In: Plant Growth in Relation to Soil Physical
33
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Conditions (Eds. J. Lipiec, R. Walczak, G. Jόzefaciuk), 48-56, Institute of
Agrophysics, Lublin, Poland.
Gaţe O.P., Czyż, E.A. and Dexter, A.R. 2005: Soil physical quality as affected
by management practices. Mat. Conf. Eur. Geophysical Union, 25-30 April,
Vienna, Austria.
Gaţe O.P., Czyż, E.A. and Dexter, A.R. 2006a: Assessment and prediction of
soil physical quality for Polish arable layers. Mat. Int. Conf. Soil protection
strategy-needs approaches for policy support. 9-11 March, IUNG-PIB, Puławy,
Poland.
Gaţe O.P., Czyż, E.A. and Dexter A.R. 2006b: Soil physical quality, S, as a
basis for relationships between key physical properties of arable soils. In: Soil
management for Sustainability (Eds. Rainer Horn, Heiner Fleige, Stephan Peth,
Xinhau Peng). Advances in GeoEcology, 38, s. 102-109, Catena Verlag:
Reiskirchen (ISBN 3-923381-52-2).
Han Han, Gimenez D. and Lilly, A. 2008: Textural averages of saturated
hydraulic conductivity predicted from water retention data. Geoderma, 146, 121128.
Kay, B.D. and Dexter, A.R., 1990: Influence of aggregate diameter, surface
area and antecedent water content on the dispersibility of clay. Can. J. Soil Sci.,
70, s. 655-671.
Kay, B.D. and Dexter, A.R. 1992: The influence of dispersible clay and
wetting/drying cycles on the tensile strength of a red-brown earth. Aust. J. Soil
Res., 30, s. 297-310.
Kay, B.D. 1990: Rates of Change of Soil Structure Under Different Cropping
Systems. Advances in Soil Science (Ed. Stewart, B. A.), Springer-Verlag, 12, s.
1-52.
Keller, T., Arvidsson J. and Dexter, A.R. 2007: Soil structures produced by
tillage as affected by water content and the physical quality of soil. Soil Till.
Res., 92(1-2), s. 45-52.
Lityński, T. and Jurkowska, H. and Gorlach, E. 1976: Agricultural chemical
analyses, soils and fertilizers. (in Polish). PWN Press, Warsaw, Poland, 330 pp.
Ostrowska, A., Gawliński S. and Szczubiałka, Z. 1991: Methods of Analysis and
Etimation of Soil and Plant Properties. (in Polish). Prot. Envir. Inst. IOŚ Press,
Warsaw, Poland, 333 pp.
Raine, S.R. and Loch, R.J. 2003: What is a sodic soil? Identification and
management options for construction sites and disturbed lands. In: Roads,
Structures and Soils in Rural Queensland. 29-30 July, Toowoomba.
Queensland Department of Main Roads, Brisbane, 14 pp.
Schjønning P., de Jonge L.W., Olsen, J.E. and Greve, M.H. 2009: Organic
farming effects on clay dispersion in carbon-exhausted soils. ICROFS news,
November/2009. Archived at http: //orgprints.org/16503
Watts, C.W. and Dexter, A.R. 1997: The influence of organic matter in reducing
the destabilization of soil by simulated tillage. Soil Till. Res., 42, s. 253-275.
Watts, C.W., Dexter, A.R. and Longstaff, D.J. 1996a: An assessment of the
vulnerability of soil structure to destabilisation during tillage. Part II. Field trials.
Soil Till. Res., 37, s. 175-190.
Watts, C.W., Dexter, A.R., Dumitru, E. and Canarache, A. 1996b: Structural
stability of two Romanian soils as influenced by management practices. Land
Degradation & Development, 7, s. 217-238.
34
Chapter 4
EFFECT OF APPLIED SIMPLIFICATIONS IN
TILLAGE SYSTEMS
ON BIOLOGICAL ACTIVITY AND POM
CONTENT IN SOIL
Anna M. GAJDA
INTRODUCTION
Soil microbial biomass (SMB) and particulate organic matter (POM) are
considered as sensitive indicators of the direction of soil quality changes caused by
many environmental factors, among others applied farming systems. Because of their
short turnover time (< 1 year) they respond rapidly to all conditions alter soil
environment. It has been shown that microbial biomass carbon (MBC) comprises 1 to
5% of the total organic C [Anderson and Domsch, 1989; Sparling, 1992]. Also, it has
been estimated that 90% of the CO2 evolved from soil comes from the activity of soil
microorganisms. This illustrates the extremely important role of soil microorganisms
that they play in soil environment [Doran, 1987; Smith and Paul, 1990]. Particulate
organic matter (POM) and other size-defined fractions as density-flotation fractions of
soil organic matter (SOM), collectively known as light fractions of SOM, have been
shown to be dynamic and more sensitive to change than total SOM [Masto et al.,
2006; Liebig at al., 2004; Gajda and Martyniuk, 2005]. For these reasons POM has
been intensively studied as an indicator of changes in soil quality in response to
various soil management systems, particularly soil tillage intensity and crop rotations
[Liebig et al., 2004; Cambardella et al., 2001].
The aim of conducted research was to assess the effect of reduced tillage
system on microbial biomass C and N content, number of nitrate-forming bacteria,
dehydrogenase activity and particulate organic matter fraction as compared to the
applied conventional tillage system.
MATERIALS AND METHODS
The studies were conducted on the long-term field experiment located on a
private farm in Rogow (Lublin voivodeship) on a silt soil and at the IUNG-PIB
Experimental Station in Grabow (Masovian voivodeship) on a sandy loam soil in the
35
years 2006-2008. The compositions of the soils were analyzed by standard methods.
Reduced tillage RT
Conventional tillage CT
Photo 1. Field experiment at private farm in Rogow (Lublin voivodeship) on a silt soil.
Conventional (CT) and reduced tillage (RT) systems
Photo 2. Experimental Station of IUNG-PIB in Grabow (Masovian voivodeship) on a
sandy loam soil. Conventional (CT) (read color) and reduced tillage (RT)
(green color) systems
At both experimental sites winter wheat was grown under two tillage systems:
conventional tillage (CT) based on the moldboard plough and traditional soil tillage
equipment and reduced tillage (RT) based on soil crushing-loosening equipment and
rigid-tyne cultivator. Surface of experimental fields was mulched with chopped wheat
straw under both tillage systems. The winter wheat was fertilized according to Polish
agrotechnical recommendations.
36
Two tillage systems:
traditional (conventional)
tillage (CT) (surface covered with
chopped wheat straw) based on
mouldboard ploughing (to 25
cm depth) and traditional soil
management equipment, and
reduced tillage (RT) with surface
mulching (chopped wheat straw)
based on soil crushingloosening equipment and a
rigid-tine cultivator (to 10 cm
depth)
Photo 3. Management equipment used on fields under CT and RT systems
Soil samples were taken randomly during each vegetation season from 0-15;
15-30 cm depths. To determine soil water content samples were dried for 24 h at
105ºC. The samples were stored in plastic bags that were then stored in a cold room
(4ºC) no longer than 10 days. These procedure ensured that the samples remained
moist and had well preserved their biological activity.
Photo 4. Collecting of soil samples at Rogow
For analyses requirements, all samples were passed through a 2 mm sieve
and stored fresh at 4ºC in a refrigerator. During two weeks all analyses required fresh
37
soil were completed. Analyses of microbiological properties of soil included: microbial
biomass C content using the fumigation-incubation (F-I) method [Jenkinson and
Powlson, 1976] with modifications given in Martyniuk et al. [2001], the rate of CO2 of
evolution from soil using the titration method, microbial biomass N content using the
fumigation-extraction (F-E) method [Jenkinson and Powlson, 1976] modified by
Kenney and Nelson [1982], most probably number (MPN) of nitrate-forming bacteria
according to McCrady [1980], soil dehydrogenase activity system [Casida et al.,
1964], and content of particulate organic matter (POM).
CT
RT
Photo 5. Estimation of activity of dehydrogenase system in soil
Photo 6. Microorganisms isolated from soil under winter wheat
For POM analysis representatives air dried (a. d.) soil samples were dispersed
with 100 ml of 5-g L-1 sodium hexametaphosphate and shook for 18 h on reciprocal
shaker. After shaking soil was passed through a 500 µm sieve and then through a 53
µm sieve. Particulate organic matter (POM - 0.053 mm fraction) quantities in the soils
were determined using Cambardella and Elliott [1992] modified method in which
POM was estimated by the Loss-On-Ignition procedure (LOI) described by Schulte
[1988], Schulte and Hopkins [1996] and Gajda et al. [2001].
Once a year estimation of total N content (Dumas method – FP-528-LECO)
and organic C content were performed in the certified Central Laboratory of Institute
of Soil Science and Plant Cultivation, National Research Institute (IUNG-PIB) at
Pulawy, Poland.
All analyses were performed using fresh or air dried soil (according to the
requirements of used methodology) in 3 replicates of each sample. Obtained results
were statistically analyzed using ANOVA method. Differences at P<0.05 were
38
considered as signifinant.
Details of the compositions of the studied soils are given in Table 1.
Table 1. Some physical properties of the studied soils
Site
Soil
Sand
Silt
g/100g
g/100g
Rogow
Silt
14.0
81.0
sandy
loam
Grabow
68.0
28.0
Clay
g/100g
5.0
4.0
C org.
g/100g
1.24
1.23
These are originally different soils but with the same contents of organic carbon (Corg.).
RESULTS AND DISCUSSION
The highest content of soil microbial biomass C was estimated in RT soil taken
from Rogow at the 0-15 cm depth – 430 µg CO2-C/g d. w. of soil (on average). Soil
managed conventionally contained a smaller amount of microbial biomass C pool
(about 80%) than the RT soil. Microbial biomass C contents were smaller at 15-30
cm depth than at 0-15 cm depth in both CT and RT soil by about 25% and 30%,
respectively (significant at α = 0.05) as shown in Fig. 1A. There were no statisticallysignificant effect of tillage system and soil depth observed on microbial biomass C
content in sandy loam soil ES in Grabow (Fig. 1B).
Figure 1. Microbial biomass C content in soils under winter wheat grown in CT and
RT tillage systems in Rogow (A) and Grabow (B)
a, b, c – values marked with different letters are statistically significant at P >95.0%
The evolution of CO2 from soil was influenced by both the tillage systems and
soil depth. The largest amounts of CO2 evolved were measured with RT soil sampled
at 0-15 cm depth in Rogow – (310 µg CO2-C/g d. w. of soil, on average). Evolution of
CO2 at 15-30 cm depth was about 30% less than at 0-15 cm depth (significant at α =
0.05) in the RT system, but in the CT system the differences between soil depths
were not statistically significant (Fig. 2A). Similar trends in CO2 evolution were
observed in soil taken from experimental fields in Grabow, but the values obtained
were less than 50% of those obtained with the Rogow soil (Fig. 2B). Also Balota et al.
[2003], Doran and Parkin [1996], Marinari et al. 2006 and Masto et al. [2006]
published a higher results of microbial biomass C content and the rate of CO2
evolution in soil under reduced as compared to the conventional tillage system.
39
Figure 2. The rate of CO2 evolution from soils under winter wheat grown in CT and
RT tillage systems in Rogow (A) and Grabow (B). Explanations as in Fig. 1
An influence of tillage system on the microbial biomass N content in the
studied soils was also observed. The greatest content of microbial biomass N was
measured in RT soil in Rogow in both studied layers - 0-15 and 15-30 cm, 33.7 and
28.0 µg N/g d. w. of soil, respectively. In the CT soil the estimations of microbial
biomass N were only about 20% (on average) of those measured with the RT soil. In
the RT soil, an effect of soil depth (significant at α = 0.05) was observed. In the
surface layer estimations of microbial biomass N were 20% greater than in the
subsurface soil (Fig. 3A). The results with the Grabow soil showed similar trends to
those obtained with the Rogow soil, but the values in the RT soil in Grabow were only
about 20% of those obtained in Rogow (Fig. 3A, B). There were no statisticallysignificant effect of soil depth on the size of the microbial biomass N pools in the CT
soil at both experimental sites (Fig. 3). Similar results has been published by Lalande
et al. [2005], Calderon [2000] and Gajda and Martyniuk [2005].
B
40
b
biomass N,
ug N/g d. w. soil
biomass N,
ug N/g d. w. soil
A
40
35
30
25
20
15
10
5
0
c
a
a
CT
30
RT
20
10
a
b
a c
0
0-15 cm
0-15 cm
15-30 cm
soil depth
15-30 cm
soil depth
Figure 3. Microbial biomass N content in soils under winter wheat grown in CT and
RT tillage systems in Rogow (A) and Grabow (B). Explanations as in Fig. 1
At both experimental sites, tillage system and soil depth affected the number
of nitrate-forming bacteria (significant at α = 0.05). Higher counts of nitrate-forming
bacteria (about 1.5 times, on average) were found in the surface layer of
conventionally-managed soil in Rogow and Grabow, as compared with the 15-30 cm
layer, however in the RT soil the differences between analyzed soil depths were
greater and counts of nitrate-forming bacteria in 0-15 cm layer were 2.5 times larger
than in the 15-30 cm layer. In general, CT and RT soil taken in Rogow and Grabow
contained similar numbers of nitrate-forming bacteria in both depths ranges. The
results showed significantly greater nitrification strength of RT soil than soil managed
40
conventionally at both experimental sites (Fig. 4).
Figure 4. Most probably number (MPN) of nitrate-forming bacteria (Nitrosomonas
spp., Nitrobacter spp. and others) in soils under winter wheat grown in CT and RT in
Rogow (A) and Grabow (B). Explanations as in Fig. 1
The activity of soil enzymes can be used as a sensitive indicator of changes in
soil biological activity and fertility in response to various soil management practices
[Doran, 1987; Bending et al., 2004; Marinari et al., 2006]. In our studies, the effect of
soil tillage system on enzymatic activity of the soils was determined by assessing the
activity of dehydrogenase systems, which provides correlative information on the
biological activity and microbial populations in soil as well as indicates the rate of
organic matter oxidation [Gajda et al., 2001; Martyniuk et al., 2001; Gajda and
Martyniuk, 2005]. The activity of the dehydrogenase system in the studied soils was
influenced by both the applied tillage systems and soil depth. The greatest activity of
dehydrogenases was measured in RT soil in Rogow in the layer - 0-15 cm - 1517,6
mm3H2/100g d. w. of soil, and 2131,4 mm3H2/100g d. w. of soil, in 2006 and 2008,
respectively. Similar trends were observed in the layer 15-30 cm. In the CT the
measurements of dehydrogenases activity were only about 60% (on average) of
those measured with the RT soil. In the RT soil, an effect of soil depth (significant at
P<0.05) was observed, in both studied years. In the surface layer the measurements
of dehydrogenase activity were about 25% (on average) greater than in the
subsurface soil in both experimental years (Fig. 5A). The results with the Grabow soil
showed similar trends to those obtained with the Rogow soil, but the values in the RT
soil in Grabow were only about 35% of those obtained in Rogow (Fig. 5B). Similar
results have been presented by other authors, e.g. Gajda et al. [2000]; Martyniuk et
al. [2001]; Marinari et al. [2006].
41
Figure 5. The dehydrogenase system activity in soils under winter wheat grown in CT
and RT in Rogow (A) and Grabow (B). Explanations as in Fig. 1
According to Liang et al. [2002] the pool of labile organic matter fraction is
several times greater than that of soil microbial biomass. He also reported the great
importance of labile organic matter fraction in the turnover of organic matter in
agricultural soils as a source of readily mineralizable substrate for soil
microorganisms. In our studies, the quantity of POM was influenced by both the
tillage system and soil depth. In the layer 0-15 cm the measured quantity of POM
fraction in the Rogow soil was significantly higher (25% on average) in RT soil as
compared to the POM content in CT soil. Similar trends in POM quantity were
observed in the Grabow soil (Fig. 6).
Figure 6. The average POM content in the 0-30 cm layer of Rogow and Grabow soils
under winter wheat grown in CT and RT. Explanations as in Fig. 1
After three year studies the decrease of POM content in CT soil was noticed in
both experimental sites. Content of POM in CT soil measured in 2008 was about
30% less, on average, than that measured in 2006. In soils under RT the increase of
quantities of POM was significant at both experimental sites. In the Rogow soil
42
measured content of POM in 2006 was about 22% lower as compared to 2008.
Obtained results in the Grabow soil showed similar trends to those in the Rogow soil,
but the values of POM content in Grabow were about 30% (on average) lower than
those obtained in Rogow (Fig. 7). Also, several authors reported the rapid depletion
of the labile fraction of OM in soil as result of cultivation practices and the rapid
increase when soil was not disturbed by tilling [Marriott and Wander, 2006]; Sleutel et
al. [2006]; Wander and Yang [2000].
Figure 7. Changes in POM content in the 0-15 cm layer of soils under winter wheat
grown in CT and RT systems in Rogow and Grabow. Explanations as in Fig. 1
At both experimental sites the quantity of POM especially in soils under RT
was significantly lower in the 15-30 cm layer than that in surface soil (0-15 cm) by
about 20-30%, on average. In general, there were no statistically-significant
difference in the quantity of POM in studied depths of CT soils (Fig. 8). Similar results
were obtained by Albrecht et al. [2000]. They also noticed that tillage created an even
distribution of POM throughout the soil profile, but the decrease of tillage intensity
increased the POM fraction, especially in the upper soil depths.
Figure 8. The effect of soil depth on the quantity of POM in Rogow (A) and Grabow
(B) soils under CT and RT systems
43
CONCLUSIONS
1)
2)
3)
In general, at both experimental sites the size of microbial biomass C and N
pools were significantly greater in soil under reduced tillage than in soil
managed conventionally in both depths of soil profile: 0-15 and 15-30 cm.
The rate of CO2 evolution, number of nitrate-forming bacteria and activity of
dehydrogenase system reached the highest values mostly in soil under reduced
tillage at both experimental sites.
During three-year studies the decrease of POM content in soil under
conventionally managed soil was noticed. In Rogow and Grabow soils under
reduced tillage the significant increase of POM fraction content was observed.
ACKNOWLEDGEMENTS
The author would like to thanks to Polish Ministry of Education for the award of grants
AGROGAS 17/E-184/SN-019/2006 and Project No. 6ZR7 2006C/06735.
REFERENCES
1. Albrecht S. L., Fauci M. F., Skirvin K. W., Bezdicek D. F. 2000: Organic matter
fractions from pacific northwest soils: depth and tillage effects. Columbia Basin
Agricultural Research Annual Report Spec. Rpt. 1012, pp. 19-23, Oregon State
Univ. in cooperation with USDA-ARS, Pendleton, OR
2. Anderson J.P.E., K.H. Domsch 1989: Ratios of microbial biomass carbon to
total carbon in arable soils. Soil Biology and Biochemistry, 21, pp. 471-479
3. Balota E. L., A. Colozzi-Filho, D.S. Andrade D, R.P. Dick, 2003: Microbial
biomass in soils under different tillage and crop rotation systems. Biol. Fertil.
Soils, 38, pp. 15
4. Bending G.D., Turner M.K., Rayns F., Markx M.C., Wood M. 2004: Microbial
and biochemical soil quality indicators and their potential for differentiating areas
under contrasting agricultural management regimes. Soil Biol. Biochem., 36,
1785-1792
5. Calderon F. J., L.E. Jackson, K.M. Scow, D.E. Rolston, 2000: Microbial
responses to stimulated tillage and in cultivated and uncultivated soils. Soil Biol.
Biochem. 32, pp. 1547
6. Cambardella C. A., E.T. Elliott 1992: Particulate soil organic matter changes
across a grassland cultivation sequence. Soil Sc. Soc. Am. J., 56, pp. 777
7. Cambardella C. A., A.M. Gajda, J.W. Doran, B.J. Wienhold, T.A. Kettler 2001:
Estimation of Particulate and Total Organic Matter by Weight Loss-On-Ignition.
In: R. Lal, J. M. Kimble, R. F. Follett, B. A. Stewart (Eds.), Assessement
Methods for Soil Carbon, CRC Press LLC, Boca Raton, FL, pp. 349-359
8. Casida L.E., D.A. Klein Jr., T. Santoro 1964: Soil dehydrogenase activity. Soil
Science, p. 98
9. Doran J.W. 1987: Microbial biomass and mineralizable nitrogen distribution in
no-tilled and plowed soils. Biology and Fertility of Soils, 5, pp. 68-75
10. Doran J. W., T.B. Parkin, 1996: Quantitative indicators of soil quality: A
minimum data set. In: Doran J. W., Jones A. J. (Eds.), Methods for Assessing
Soil Quality. SSSA Special Publication 49, Madison, WI, pp. 25-37
11. Gajda A. M., J.W. Doran, T.A. Kettler, B.J. Wienhold, J.L. Pikul Jr., C.A.
Cambardella 2001: Soil quality evaluations of alternative and conventional
management systems in the Great Plains. In: R. Lal, J. M. Kimble, R. F. Follett,
44
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
Stewart B. A. (eds) Assessement Methods for Soil Carbon, Lewis Publishers,
CRC Press LLC, Boca Raton, FL, USA, pp. 381-400
Gajda A.M., S. Martyniuk, 2005: Particulate organic matter and microbial
biomass C contents in soils with different mechanical structure. Pam. Pul. 40,
pp. 49
Gajda A., S. Martyniuk, A. Stachyra, B. Wróblewska, S. Zieba 2000: Relations
between microbiological and biochemical properties of soil under different
agrotechnical conditions and its productivity. Polish J. Soil Sci. 33, pp. 50
Jenkinson D.S., D.S. Powlson 1976: The effect of biocidal treatments on
metabolism in soil V. A method for measuring microbial biomass. Soil Biology
and Biochemistry, 8, pp. 209-213
Keeney D.R., D.W. Nelson 1982: Nitrogen: Inorganic forms. In: Page A.L. et. al.
(eds.) Methods for analysis. Part 2. 2nd ed. Agron. Monogr. 9. ASA and SSSA,
Madison, WI
Lalande R., B. Gagnon, R.A. Chapman, G.M. Barnett 2005: Soil microbial
populations, activity, and community structure in continuous corn or forage
systems under organic or inorganic fertilization in eastern Canada. Canadian J.
Soil Sci., Vol. 85, No. 1, pp. 27
Liang B. C., B.G. McConkey, J. Schoenau, D. Curtin, C.A. Campbell, A.P.
Moulin, G.P. Lafond, S.A. Brandt, H. Wang 2002: Effect of tillage and crop
rotations on the light fraction organic carbon and carbon mineralization in
Chernozemic soils of Saskatchewan. Canadian J. of Soil Sc., 83, pp. 65-72
Liebig M. A., D.L. Tanaka, B.J. Wienhold 2004: Tillage and cropping effects on
soil quality indicators in the northern Great Plains. Soil Till. Res., 78, pp. 131
Marinari S., R. Mancinelli, E. Campiglia, S. Grego 2006: Chemical and biological
indicators of soil quality in organic and convewntional farming system in Central
Italy. Ecological Indicators, 6, pp. 701-711
Marriott E. E., Wander M. 2006: Qualitative and quantitative differences in
particulate organic matter fractions in organic and conventional farming system.
Soil Biol. Biochem. 38, 1527
Martyniuk S., Gajda A., Kus J. 2001: Microbiological and biochemical properties
of soil under cereals grown in the ecological, conventional and integrated
systems. Acta Agrophysica, 52, pp. 185.
Masto R. E., P.K. Chhonkar, D. Singh, A.K. Patra 2006: Changes in soil
biological and biochemical characteristics in long-term field trial on a subtropical inceptisol. Soil Biol. Biochem. 38, pp. 1577
Schulte E. E. 1988: Recommended soil organic matter testes. In:
Recommended Chemical Soil Test Procedures for the North Central Region.
North Cen. Reg. Pub. No. 221, Bull. No. 499, North Dakota Ag. Exp. Stn., North
Dakota State Univ., Fargo, ND, 58105, pp. 29-32
Schulte E. E., Hopkins B. G. 1996: Estimation of soil organic matter by weight
loss-on-ignition. In: Soil Organic Matter: Analysis and Interpretation. SSSA
Special Publication No. 46. Soil Science Society of America, Madison, WI, pp.
21-31
Sleutel S., De Neve S., Nemeth T., Toth T., Hofman G. 2006: Effect of manure
and fertilizer application on the distribution of organic carbon in different soil
fractions in long-term field experiment. Europ. J. Agronomy, 25, 280
Smith J.L., E.A. Paul. 1990: The significance of soil biomass estimations. In:
Bollag J.M. and Stotsky G. (eds.), Soil Biochemistry, Marcel Dekker, New York,
6, pp. 357-396
45
27. Sparling G.P. 1992: Ratio of microbial biomass to soil organic carbon as a
sensitive indicator of changes in soil organic matter. Australian Journal of Soil
Research, 30, pp. 195-207
28. Voroney R.P., Paul E.A.1984: Determination of KC and KN in situ for calibration
of the chloroform incubation method. Soil Biology and Biochemistry,16 pp. 9-14
29. Wander M. M., Yang X. 2000: Influence of tillage on the dynamics of loose- and
occluded- particulate and humified organic matter fractions. Soil Biol. Biochem.,
32, 1151
46
Chapter 5
DATA MINING AS A DECISION-SUPPORT
TOOL IN SUSTAINABLE AGRICULTURAL
PRODUCTION
Jędrzej TRAJER, Jarosław STANKIEWICZ
INTRODUCTION
Sustainable agricultural production is a system of farming allowing for the
development of a farm while maintaining ecological safety and taking the
interrelations between the farm and the surroundings, i.e. environment and the
infrastructure into consideration [Sawa 2008]. Such a system of farming can be
implemented by means of ensuring soil fertility, appropriate selection of species and
varieties of plants, fertilization and plant protection. It can be achieved by means of
appropriate management, conscious use of self-regulation mechanisms in the
ecosystems and applying methods and tools available due to scientific and technical
progress. Farmers’ knowledge and skills are of primary importance in the farm
development understood in this way, and their effectiveness may be improved by
means of using appropriate methods supporting decision-making. Following
sustainable farming principles requires a series of adjustments in the field of
organization and technology, which are established on the basis of various criteria for
the evaluation of agricultural production sustainability. It may apply to various aspects
such as economic, social or ecological ones, which are analysed in relation to the
intensity of production. The problem is complex and multi-dimensional. Therefore,
advanced tools such as data mining should be used to analyse it. These methods
allow for extracting knowledge from large datasets using statistics and artificial
intelligence methods.
The paper presents an example of using data mining for the analysis of black
currant yield. Poland is one of the major soft fruit (especially black currant) producer.
According to the data provided by the Central Statistical Office (GUS) [Statistical
Yearbook, 2009] Poland’s share in the global production equals 40% and it is equal
to 70% in the European Union. Approximately 24% of farmers grow and produce
black currant, and the share of black currant plantations in the total area of orchards
and berry fruit plantations equals 35%. Unfortunately, the structure of the production
is fragmented and the size of most plantations does not exceed 1 ha (approximately
94%), and there are only a few thousand large and very large plantations. Such high
47
fragmentation leads to low specialization of production and results in treating black
currant cultivation as less important and an additional source of income. However,
there are also large, highly specialized farms in Poland, and their efficiency is much
higher. Despite such a large share in the global production, Polish producers of black
currant are still waiting for the stabilization of the market and strive with basic
problems: unstable price and finding a market for their product. Yield is one of the
major determinants in agricultural production, it determines additional profits as well
as additional costs. Its value depends on numerous external factors, both humandependent and human-independent, e.g. on weather conditions. A thorough analysis
of this quantity may allow for making fast and appropriate decisions that are essential
for more efficient sustainable farming.
RESEARCH METHODOLOGY
Data mining makes use of advanced methods of analysis (statistics, artificial
intelligence, databases), which make it possible to extract knowledge from large
datasets using computers, which would otherwise be difficult [Berry and Linoff 2000].
Analyses are made by means of created models and formulated classification
problems. The examples of such analyses are presented below.
Black currant production analysis indicates that important factors which
determine the yield include: variety – a very important factor as each variety is
characterized by different yielding, disease resistance, flowering season/ maturation,
which, in turn, is connected with another factor – minimum temperature during
flowering season. If frost (even light frost) occurs during flowering season, which
occurs quite often in this climate, as in the past few years, the yield may be
considerably reduced as the majority of flowers die and fall. Another weather-related
factors are total precipitation, especially in June, the month of the most intensive
growth of fruit. However, the total precipitation itself does not provide sufficient
information as single heavy rainfall is not desirable. Therefore, the number of rainy
days was also taken into consideration. Sun exposure that has influence on the
speed and uniformity of maturation was also taken into consideration in the model.
Other significant factors, which to a greater extent depend on human, include:
fertilization N, P, K as well as chemical protection against various kinds of diseases
and insect pest. Neglecting these two factors may also seriously affect plants or lead
to big loses that influence the yield not only in one particular year but also in
successive years due to the fact that different kinds of insect pest may affect plants
and damage them regularly. Another factor that has influence on the yield is
plantation age, and the yield itself as the output data. Data for the analysis were
obtained from horticulturists [Stankiewicz 2010], and weather data were obtained
from the Institute of Meteorology and Water Management in Warsaw.
Cluster Analysis
Generalized EM method was used for the initial cluster analysis of data in
order to identify homogeneous groups of cases, which may be analyzed in detail.
Clustering using EM method [Stanisz 2007] is a modification of a statistic method of
k-means method, involving the determination of a certain number k of homogeneous
groups (clusters, concentrations), and the determination of the centre of each group
as well as assigning a certain number of objects to them. It also involves calculating
the probability, with which the objects match particular groups. In this method, the
48
distance between each object and the centre is calculated for each object, and the
objects are assigned to individual groups depending on the shortest distance from
the centre. After the objects are assigned for the first time, the centres are updated. It
is done by calculating the means in each group. After that, the procedure is repeated,
and the objects are updated; their distances from the centres are verified, and, if
need arises, they are moved to other groups. The following steps are repeated until
the shortest distances (differences) within each group are the shortest possible and
the distances between the groups are the longest possible.
Multilayer perceptron – MLP
For the isolated groups it is possible to create models describing
dependencies between input variables (predicators) and the output variable, which
make it possible to analyse the influence of selected factors determining the yield.
Artificial neural networks, MPL type [Osowski 1996] were used for that purpose.
Figure 1 shows the network topology.
Figure 1. Multilayer perceptron (MLP) topology
The neuron on the right represents the output variable – yield. The neurons on
the left represent the input neurons – quantity variables: (starting from the top)
plantation age, fertilizers quantities (N – nitrogen, P- phosphor, K – potassium), the
number of chemical treatments (herbicides, fungicides, and insecticides), minimum
temperature in the flowering season, total precipitation, and number of days with
precipitation in the period of fruit maturation and sun exposure, the latter one being of
qualitative type (it represents three examined varieties). The number of neurons in
the hidden layer is determined by means of testing various network topologies,
decreasing the number of them, starting with 2n+1 until the learning quality does not
worsen. Such maximum number of the neurons results from the Kolomogorov
theorem [Osowski 1996], which says that a neural network can describe any
nonlinearity using a multi-layer perceptron, in which the hidden layer has 2n+1
neurons (where n is the number of neurons in the input layer). Error backpropagation algorithm was used for training the network. This algorithm involves
appropriate modification of weights so that the minimum error be obtained, here in
the approximation task. In order to improve the quality of the created models the
number of input variables was reduced, and those values which did not increase the
error significantly were omitted. The models may be used to determine yields for
different values of input variables and to determine the influence of changes in the
49
predicators’ values on the yield value (two- and three-dimensional charts may be
created).
Self-organizing map – SOM
One of the most efficient tools supporting decision-making is a self-organizing
map [Kohonen 1988] representing individual cases classified in the whole group in
the graphical form. It makes it possible to assess them both individually and within a
group with regard to the selected predicators as well as make clusters mapping
similar cases according to other analysis criteria. Other tools do not provide such
options.
A map of features was created using Kohonen network, which has an input
layer representing predicators selected for the analysis, as well as a radial output
layer in the form of a topological map, fig. 2.
Figure 2. Kohonen network topology
The training process takes place without a teacher using the neighbourhood
algorithm, in which similar cases are distributed closely on the topological map.
Having finished training stage, individual cases are assigned to the localization on
the topological map, and they may be assessed in relation to the predicators and
according to the adopted criteria. New instances may also be analysed, which are
interpreted based on the location in the previously identified area on the map.
RESEARCH RESULTS
Cluster analysis using EM method allowed to distinguish three homogeneous
groups in 93 cases that were analysed, table 1. The symbols of the columns
represent: NrP – number of case, P – probability with which a given case belongs to
a given cluster, NrS – the number of cluster. There were so few cases in cluster
number 2 that the cases it contained were assigned to cluster number 3 for the
purpose of building neural models. The probabilities with which cases were assigned
to clusters were so high that they confirmed the fact that the right number of clusters
had been selected. Two groups were identified as a group with a black currant
plantation, which is more than 8 years old (cluster 1), in the second one – younger
black currant (cluster 3).
50
Table 1. Cluster analysis results for the black currant yield
NrP
P
NrS NrP
P
NrS NrP
P
NrS
3 32 1.0000
3 63 0.9996
3
1 0.9999
0.9988
3
0.9999
3
0.9999
3
2
33
64
3 34 0.9730
3 65 0.9999
1
3 0.9998
3 35 1.0000
2 66 0.9999
1
4 0.9997
3 36 0.9999
3 67 0.9999
1
5 0.9996
3 37 0.9998
3 68 0.9999
1
6 0.9997
2 38 0.9987
3 69 0.9999
1
7 0.9929
2 39 1.0000
2 70 0.9999
1
8 0.9601
3 40 1.0000
2 71 0.9988
1
9 0.9998
3 41 0.9996
3 72 1.0000
2
10 0.9979
3 42 0.9993
3 73 0.9999
1
11 0.9999
3 43 1.0000
3 74 1.0000
1
12 0.9999
3 44 1.0000
3 75 1.0000
1
13 0.9999
3 45 1.0000
3 76 1.0000
1
14 0.9990
3 46 1.0000
3 77 1.0000
1
15 0.9998
3 47 0.9999
3 78 1.0000
1
16 0.9997
3 48 1.0000
3 79 1.0000
1
17 0.9999
0.9984
3
0.9999
3
0.9999
1
18
49
80
3 50 0.9999
3 81 0.9999
1
19 0.9996
3 51 0.9999
3 82 0.9999
1
20 0.9998
3 52 0.9317
3 83 0.9999
1
21 0.9999
3 53 0.9998
3 84 0.9999
1
22 0.9998
2 54 0.9992
3 85 0.9999
1
23 0.9999
2 55 0.9995
3 86 0.9976
1
24 0.9659
3 56 0.9846
3 87 1.0000
2
25 0.9999
3 57 0.9999
3 88 1.0000
1
26 0.9994
3 58 0.9998
2 89 1.0000
1
27 1.0000
3 59 0.9997
3 90 1.0000
1
28 1.0000
3 60 0.9999
3 91 1.0000
1
29 1.0000
3 61 0.9999
3 92 0.9999
1
30 1.0000
0.9999
3
0.9996
3
1.0000
1
31
62
93
Neural forecast models were build for these groups. For cluster 3, with
younger black currant, MLP9:16:1 model was obtained, in which input variable 3
(fertilization with phosphor) was omitted as it was considered insignificant. Sensitivity
analysis shows the importance of other predicators, table 2. It is possible to observe
that the variety which grows faster is an important factor for younger plantations. It
was also shown in the sensitivity analysis, where the age of plantation was the
second most important factor. In the first group, the program selected the number of
days with precipitation as the most important factor, which reflects the reality as
regular precipitation boost growth, which leads to higher yields.
Similarly, for cluster 1 (plantation older than 8 years) MLP8:16:1 model was
obtained, in which variables 1 and 8, representing age of plantation and the number
of days during growing season respectively, were reduced.
Sensitivity analysis (table 3) for this model shows a significance of the
selection of black currant variety as well as appropriate fertilization.
51
Table 2. Sensitivity analysis of the black currant yield model for cluster 3 (plantation
up to 8 years old)
Table 3. Sensitivity analysis of black currant yield model for cluster 1 (plantation over
8 years old)
Both models were positively verified on the basis of satisfactory values of
regression statistics, table 4.
Table 4. Regression statistics for neural models of black currant yield
Training set
Validation set
Regression
statistics
up to 8
over 8 years up to 8 years over 8 years
years old
old
old
old
Error ratio
0.557
0.399
0.533
0.299
Correlation
0.831
0.919
0.851
0.960
52
The values of regression statistics of the standard deviation ratio of the errors
and Pearson correlation coefficient prove that the model for the plantation over 8
years old has slightly better quality. In order to improve the quality of both models, a
bigger set of training cases should be used and appropriate predicates should be
selected.
The designed models allow for the assessment of the influence of values of
each predicator on the value of the yield. Figures 3,4 and 5 show three dimensional
graphs showing the influence of significant predicators for the model of young
plantation.
Figure 3. The influence of precipitation and fertilization on the yield of black currant,
plantation up to 8 years old
Figure 4. The influence of sun exposure and the number of treatments on the yield of
black currant, plantation up to 8 years old
53
Figure 5. The influence of age plantation and the minimum temerature on the yield
of black currant, plantation up to 8 years old
Fig. 3 shows that fertilization in years with higher precipitation give better
results. On the other hand, fig. 4 shows, that higher yield can be obtained when the
number of treatments and sun exposure are higher. It is possible to notice that the a
bigger number of treatments during increased sun exposure results in lower yield. It
may be caused by damages during treatment. Fig. 5 shows relation between age
plantation and minimum temperatute during vegetation.
Topological map classifying cases according to selected predicators (here a
Kohonen network was created for a plantation up to 8 years old, fig. 6.) may provide
valuable information supporting decision-making concerning black currant production.
Figure 6. Kohonen network topology for the classification of the yield from a
plantation up to 8 years old
The following predicators were used (from the top): fertilization with nitrogen
and potassium, variety (one out of three), the number of protective treatments as well
as the number of days with precipitation and sun exposure during maturation. 64
cases were used for the training purpose, which were properly distributed on a self-
54
organizing map, SOM, fig. 7, when the process of training was completed. This map
has a size of 8x8 and shows distribution of 64 cases, in relation to selected
predicators, and, additionally, according a new criterion, it shows areas (clusters)
representing different yield values (from the smallest to the largest respectively, light
colour to dark colour, white colour means that there is not a unique assignment).
no
1
1 54,56
2
50
2
3
2,8
4
18,24
6
22
3
60,61,
64
4
59,62
33
41,42,
47
5
11,12
36
37
6
14
16
7
13
30
46
8
27,29
28,32
48
5
6
7
58
8
53
55
49,51,
52
38
57,63
34,40
1,7,9
15
35,39
3,4
10
19
4345
23
17,20,
21,25,
26,31
Figure 7. Self-organizing map SOM presenting black currant yield for plantation up to
8 years old
The map is the basis of prediction for the new cases, yield value may be
identified by reading their location on the map.
CONCLUSIONS
1.
2.
3.
Efficient sustainable agricultural production management requires a lot of
experience and detailed knowledge, which is difficult or even impossible to
obtain without sophisticated tools for decision support, such as data mining
methods.
The presented results of research on black currant yields using selected data
mining methods confirmed that they have a potential as regards obtaining
knowledge that would facilitate efficient decision-making process in sustainable
agricultural production.
The selection of the analysis method depends on the goal, and the results of
research using the selected method may often be the basis for further research
as the example of research on black currant yields confirms. In this example,
cluster analysis allowed to distinguish homogeneous groups of cases, which
could further be analysed in detail – using modelling and classification.
55
REFERENCES
1. Berry, M., J., A., Linoff, G. 2000: Mastering data mining. New York: Wiley.
2. Biuletyn Państwowej Służby Hydrologiczno-Meterologicznej, Rok 2002-2007.
Miesięcznik Instytutu Meteorologii i Gospodarki Wodnej, Warszawa.
3. Kohonen T. 1988: Self organization and associative memory. Springer Verlag,
New York.
4. Osowski S. 1996: Sieci neuronowe w ujęciu algorytmicznym. WNT, Warszawa.
5. Pluta S., Żurawicz E. 2006: Nowe Odmiany i Technologie Upraw Krzewów
Jagodowych. Instytut Sadownictwa i Kwiaciarstwa, Skierniewice.
6. Rocznik statystyczny 2009. Główny Urząd Statystyczny, Warszawa.
7. Sawa J. 2008: Próba oceny zrównoważenia procesów produkcji rolniczej.
Inżynieria Rolnicza. Nr 2 (100).
8. Stanisz A. 2007: Przystępny kurs statystyki. Tom 3. Analizy wielowymiarowe.
StatSoft, Kraków.
9. Stankiewicz J. 2010: Analiza czynników determinujących plon czarnej porzeczki
z wykorzystaniem data mining. Praca magisterska WIP SGGW, Warszawa.
56
Chapter 6
SITE SPECIFIC CROP FERTILIZATION IN
HAUTE-NORMANDIE: A TOOL TO
IMPROVE THE PROFITABILITY OF
CEREAL CROP SYSTEMS
Jean Marc LLORENS, Olivier PERRIN, Odile BOURGAIN
INTRODUCTION
Site specific crop management (SSCM) can potentially by used for numerous
technical actions applied to crops: management of fertilizer, pesticides, seeding
rates, or irrigation. Efficient use of agro-chemicals is beneficial for farmers as well as
for the environment. However, in most cases, the agro-environmental interest
conflicts with economic profitability [Atherton et al., 1999] or socio-economic
resistance to innovation [Robert, 2002].
It appears even more important to have a macro view of the farm to better
evaluate the economic efficiency of such techniques. Taking into account the cost of
both the acquisition of information and the equipment used, profitability should be
evaluated for aggregated crop requirements of the production systems. SSCM is
more than a simple technology. It increases the relevance of decision making criteria
of agricultural advisers [Swinton and Lowenberg-Deboer, 1998] as well as
environmental considerations [Melakeberhan and Avendano, 2008].
Taking into account the current fluctuation of raw material costs and prices,
the assessment of the economic profitability of new techniques such as precision
agriculture is becoming more and more important. Previous studies in HauteNormandie have shown that SSCM has no negative effects on yields [Duval et al.,
2007, Bécu et al., 2008]. Most of the time, the technical results obtained were
partially evaluated from an economic point of view: limited to increased profit due to
fertilizer reduction or gross profit margins. Moreover, these studies were carried out
on specific crops, mainly wheat [Duval et al., 2007] or maize [Koch et al., 2004]
and/or specific types of input, such as nitrogen [Dailey et al., 2006, De Vuyst and
Halvorson, 2004 ; Koch et al., 2004 ; Link et al, 2006 ; Lobell, 2007 ]. However, very
few studies have been undertaken to analyse global crop systems [Sartori et al.,
2005] or to evaluate the profitability of SSCM [Haefele and Wopereis, 2005], due to
the complexity of the work involved.
57
The methodology that we used in our region, allows different thresholds of
profitability concerning investments of the SSCM equipment required in relation to
soil heterogeneity levels to be assessed. We applied it to specific cropping systems
in Haute-Normandie involving heterogeneous soil and variable agricultural surface
areas. Furthermore, the methodology together with the specialised software has
proved its efficiency to assist farmers, agricultural advisers and also regional policy
decision-makers [Flemming and Adams, 1997, Link et al., 2006]. It has already been
used in a context of watersheds subject to runoff and erosion to quantify anti erosive
agricultural measures to be taken [Bourgain and Michaud, 2006]. Following this first
successful experience it was decided to extend to the analysis of the profitability of
PA under local pedoclimatic conditions.
The three main aims of this study are:
•
To elaborate a methodology including farming systems based on field sampling
and modelling.
•
To define relevant technique-economic indicators to try to estimate the
profitability of precision agriculture for the total economic activities of the farm.
•
To apply this methodology to carry out simulations. Variables being soil
heterogeneity and agricultural practices (with or without PA).
MATERIAL AND METHOD
Agricultural context: Typology of crop systems
Our final scale of investigation is the farm system as an economic unit. One part of
the economic and technical data is taken at the plot scale [Duval et al., 2007; Bécu et
al., 2008] then it is aggregated at the level of the farm. Three crop systems (cereals,
sugar beet-flax and potatoes) were selected (Table 1).
Table 1. Typology of crop systems [Rosace, 2005] used in economic simulations with
Olympe for precision agriculture in Haute-Normandie *Full Time Labour Unit
Systems
Cereals
Main caracteristics
Labour*
Average
area (ha)
Farm Unit
Type
1
95
(70-120)
C1a
2
145
(120-180)
C2
2
240
(180-350)
C3
1
110
(80-140)
BL1bis
2
200
(140-350)
BL2bis
2 or
more
150
(more than
90)
PT
more than 80% of
arable land in
cereals, oleo
proteagineous +
fallow
20 to 25 % of arable
Sugar beet
land in sugar beet
– flax
and flax
Potatoes
More than 10 ha of
potatoes (25 ha
average)
58
Each system is composed of the rotation of seven or eight different crops. In
fact, these types of farms represent 40% of the farms in Haute-Normandie. They are
defined as being without milk production and a limited surface area of permanent
pasture (less than 15% of the cultivated surface area). The turnover for livestock is
less than 20% of the total.
The six cropping systems (Table 1) represented 87% of the arable farms in
Haute-Normandie i.e. 2800 farms. This typology is the synthesis of six years of
observations [Rosace, 2005]. Standard economic information taken from various
sources was used to represent the farms in the region [Rosace, 2005, Agreste, 20082009, Teyssier, 2007, Chambre d’agriculture, 2008].
Pedoclimatic context:
The trials were carried out in fields on commercial farms in Haute-Normandie,
NW France. The soils of these fields are mainly silty over a layer of clay and flint. The
principal heterogeneity is that of depth and the quantity of stone. In each field deep
and shallow zones were determined by apparent soil electrical conductivity (ECa).
The amount of stone can be up to 50% in some shallow soils, on the other hand it is
almost absent in deep soils. The experimental plots were distributed in three
geographic zones of increasing pluviometry from the south east to the North West of
the region (between 550 and 700 mm/y).
Characterization of soil heterogeneity
The ECa was mapped in situ, using a mobile single-dipole Geonics EM38 unit
with a mobile GPS-based system. This survey consisted of a grid of ECa
measurements arranged in rows (2.8 measures per metre, each measure is the
average of 13.9 readings). The rows were 12 metres apart, giving 1,667 readings par
hectare. Each reading penetrates to a depth of approximately 1.35 m. In general,
shallow soil readings were lower than 15 mS/m and deep soil readings were up to 35
mS/m.
Modelling of crop response and the obtaining of technical optimums of nitrogen
In our experimental procedure to modelling, there were nine N levels (9) and
two replications for each level. The trials were carried out in 2005 and 2006.
For each trial data, we have selected the best model by minimizing the root
mean square error [Makowski et al., 1999]. The different models tested are as
follows:
1) ADAS + Plateau [Min (B ; a – b N – c 0.99N)] ; B = maximum yield, N = N
applied, a, b, c = parameters. (Laurent 2006, personal communication)
2) Linear + plateau [Min(c ; b + a N)] ;
N = N applied, a, b, = parameters, c =
maximum yield
3) Exponential (a - (a - b) e- c N ) ; N = N applied, a, b, c = parameters
4) Quadratic (a N² + b N + c), N = N applied, a, b, c = parameters
5) ADAS (a – b N – c 0.99N).N = N applied, a, b, c = parameters (Laurent 2006,
personal communication)
Methodology to determine the number of PK sampling
The choice criteria of the plot on which the methodology is based are; the
knowledge of the evolution of the regrouping of the land, the knowledge of the yield
and conductivity heterogeneity and the knowledge of the crop successions.
59
We constituted a grid sampling which each sample was fifty meters apart from
is neighbour (Figure 1) which corresponds to a frequency of 4 samples per hectare
(4pha). This grid sampling, allowed us to test statistically 4 different methods of
sampling for a frequency of 1 sample per hectare (1pha) and 2 different methods for
a frequency of 2 samples per hectare (2pha) (Figure 1).
Figure 1. Plot grid sampling (4 per hectare, 4pha). () 2 per hectare (2pha_V1) (+) 2
per hectare (2pha _V2)
The soil samples were carried out after harvesting in our experimental
agricultural systems. This means a period of time sufficiently distant from the field
inputs to make sure this has no effect on our interpretation of the analytical results.
The analyses were carried out on 500 g of soil, at each sampling location; a
composite soil sample (18 subsamples of about 30 g) was systematically collected in
a two circle pattern across the area, taken with a A3H Pro 15® ground auger. The
nutrient soil rates were measured by the Olsen method for phosphorus [Olsen et
Sommers, 1982] and by the exchangeable-K for potassium.
Technical referential:
The yields of wheat and oil seed rape (OSR) were taken and analyzed using
the results of modeling of earlier agronomic trials [Duval et al., 2007 and Bécu et al.,
2008]. For the other crops (Barley and peas), yields were estimated using expert
advice.
As fertilizer recommendation methodology for site specific crop management
(SSCM) the COMIFER method was used for the calculation of P and K fertilizer
recommendations [Llorens, 2001]. This consists of, first, multiplying the nutrient rate
(kg of P2O5 or K2O / ton) contained in the plant by the yield (ton/ha). Then, to
calculate a coefficient (C) which is the combination of four factors [Llorens, 2001]: the
crop requirement, soil rate, recent history of fertilization and residue use from the
previous crop.
D (kg/ha) = B (kg/t) * Y (t/ha) * C
Where D is the fertilizer amount, B the P or K content in plants, Y the crop yield
and C the multiplying coefficient.
60
Recommendations for nitrogen fertilization concerning wheat and oil seed rape
are mainly calculated by the balance-sheet methods used in France [Meynard et al.,
1997].
Quantity of recommended nitrogen = crop requirement – soil reserve
However, for oil seed rape soil reserve, a supplementary factor must be taken
into account: that is the quantity of nitrogen absorbed by the plant at the end of
winter, estimated by the weighing method.
Calculation of the economic criteria
The partial budget method
The partial budget method is used in order to determine the economic interest
of fertilization [Swinton et Lowenberg-DeBoer, 1998]. This methodology aims at
determining if the adoption of precision agriculture has a positive effect on the
profitability of the farming organisation (Cordonnier P. et al., 1970). The principle is
simple: to determined the balance (∆) between gains and losses following the
adoption of precision agriculture.
LOSSES
GAINS
A : supplementary expenses C : supplementary products
B : eliminated products
D : eliminated expenses
∆ = GAINS – LOSSES = (C+D) – (A+B)
If ∆ > 0, adoption of precision agriculture creates an increase in profit and is
thus favourable to the farmer. If ∆ < 0, adoption of precision agriculture creates a
decrease in profit and is thus unfavourable to the farmer
The direct margin method
The selected economic factors were production costs, specific fertilization
materiel costs and direct profit margin (gross margin – specific fixed costs). In our
case, Crop Product indicator was established according to the following definition:
Crop Product (CP) = yield*sale price + direct crop subsidy.
Variable costs and specific fixed costs for fertilization (VCSFCF) = inputs
(fertilizers, seeds, pesticides) + traction equipment+ fertilizer spreader + localisation
equipment (GPS, software, electronic box) + maintenance cost of the localisation
equipment + soil analysis).
Direct Profit Margin (DFM) = CP – VCSFCF
Inputs, traction and soil analysis costs were calculated per hectare of crops.
The other ones were distributed over the total crop land. Therefore, the later will be
less significant as the surface area increases.
Economic simulations
Using Olympe (economic simulator software) we were able to work at different
scales: field, crop system and farm (Figure 2). Thus we simulated the economic
profitability of SSCM equipment in the most common crop systems in HauteNormandie.
Olympe is a decision support software with various functionalities. It can be
used as a database, an accounting calculator and a simulator. Different simulations
can be done for a cropping system and be compared. In our study, they concerned:
•
soil heterogeneity of nitrogen, phosphorus and potassium,
61
•
•
yield potential of the fields, evaluated by the soil apparent electro conductivity
(ECa) which represents in Haute-Normandie (France) soil water reserve [Duval
et al., 2007] which is the main limiting factor,
surface area variability of the cropping system (cereal production system in our
study).
Figure 2. “Olympe” an economic research decision making tool which is both a data
base and a simulator [Le Bars et al., 2005]
In order to simulate the SSCM equipment profitability, the calculation
hypotheses were the following: Fertilizer amounts (N, P and K), SSCM fixed costs
concerned site-specific or non site-specific (spreader, tractor, GPS and software,
maintenance, and P and K soil analysis) and yield spatial variability (based on ECa).
In order to illustrate our methodology in this paper, only cereal production
systems results (Table 2) will be presented for 3 farm sizes (95 ha, 140 ha and 245
ha).
Table 2. Average Crop distribution for crop systems in Haute Normandie (%) and
average crop distribution for cereal production systems used for simulation (ha). SB =
Sugar beet, OSR = oil seed rape, Pt = potatoes, WB = winter barley, Pa = pasture, F
= fallow, P = Pea
Crop distribution (ha)
Cereal
wheat SB flax Pt
Pa
OS
P
WB F
R
C1a (95 ha)
49
6
17
10
13
C2 (145 ha)
75
9
26
15
20
C3 (240 ha)
125
14
43
24
34
62
RESULTS
Area evolution of crop management since 1947
The field area (21 ha) has changed considerably since 1947. At this time,
three different zones should have been identified (Figure 3). In zone 1 (about 14 ha),
the area was cropped in large plots of cereals. Zone 2 (about 3 ha) was partially
wooded. In zone 3 (about 4 ha), the plots were essentially cropped with cereals, the
areas are smaller than in zone 1 (Figure 3). From 1947 to 2003 the areas show a
constant enlargement. In 2003, there was only one field, but we can still determine
the three zones on the conductivity map (Figure 4). We have three aerial
photographs (1947, 1960 and 2003) to monitor the area evolution of our field since
1947.
Figure 3. Area evolution of our field since 1947 (21 ha in 2003) in Upper-Normandy
(NW France). Three different zones in 1947: [1] large plots (cereals). [2] Wooded
area. [3] Small plots (cereals)
Heterogeneity of yields
The results show that over the six years, with a N level calculated by the
balance sheet method, the average yield on deep soil is always superior to shallow
soils (Table 2). The loss in yield on shallow soils can reach 30% (from 9.07 to 5.9 t/ha
for wheat in 2004).
63
Table 3. Aggregated average yields of wheat and oil seed rape observed on the trial
plots for one rate of N calculated using the balance-sheet method for the period 2003
to 2008
year
Average yields of wheat (t/ha)
Average yields of OSR (t/ha)
Shallow soil
Deep soil
Shallow soil
Deep soil
2008
9.53 (± 0.06)
11.47 (± 0.6)
4.18 (±0.23)
5.77 (±1.5)
2007
7.23 (± 0.68)
8.83 (± 1.14)
3.40 (±0.41)
3.42 (±0.34)
2006
5.87 (± 0.1)
8.69 (± 0.0,7)
2.71 (±1.28)
3.7 (±0.98)
2005
7.76 (± 0.67)
9.09 (± 0.93)
4.05 (±0.73)
4.13 (±0.52)
2004
5.90 (± 0.79)
9.07 (± 1.17)
4.76 (±0.81)
5.13 (±0.81)
2003
7 (± 1.25)
8.98 (± 0.91)
The amount of annual rainfall and its distribution during the year may cause
differences of yield of wheat between the two types of soil ranging from one to three
times greater on the deepest soils (1.33 t/ha in 2005 to 3.17 t/ha in 2004).
Conductivity variability in relation to soil depth and yields in our pedoclimatic
conditions
In relation to the depth soil variations, the apparent electroconductivy (ECa)
and the wheat yield map are very similar in our field (Figure 4).
Figure 4. Conductivity (mS/m) and wheat yield map (t/ha) in our field
Indeed, in 1947 and 1960 the areas where the soil was shallow (Zone 2) were
wooded (Figure 3). The yield is the lowest in the shallow soil (Zone 2) because it is
the less fertile area (Figure 4). In deep soil (zone 1) the yield can be twice as high as
in zone 2 (7 to 12 t/ha) and in intermediate soils (zone 3) the yield is slightly higher
than in zone 2.
Modelling the yield as a function of the N application rate N
To obtain the technical optimum, the first step was to model the yield as a
function N application rate from the data obtained in 2005 and 2006 for wheat and
64
OSR. The N levels simulated by the models were 0, 50, 100, 150, 200, 250, 300 and
350 kg/ha of NN (Figure 5).
The type of model used to best explain each variation in yield as a function of
N application rate N varies according to the soil depth and the pluviometry of the year
studied.
The analysis of the curves shows that for each plot, no matter which year or
crop, the maximum yield is obtained in the deep soil. The differences are more
pronounced in 2006 than in 2005, particularly for wheat (Figure 5b). In fact, in 2006
the water deficit was more apparent which translates into lower maximum yields in
2006 than 2005.
a) wheat 2005
8
6
4
.
yield (t/ha) .
10
10
yield (t/ha)
b) wheat 2006
8
4
2
2
0
100
200
300
Nitrogen level (kg/ha)
0
400
100
200
300
400
Nitrogen level (kg/ha)
c) oil seed rape 2005
6
d) oil seed rape 2006
6
.
5
4
yield (t/ha)
yield (t/ha) .
6
3
2
1
5
4
3
2
1
0
0
0
100
200
300
Nitrogen level (kg/ha)
0
400
100
200
300
400
Nitrogen level (kg/ha)
Figure 5. Increase in yield of wheat and OSR in relation to different levels of N
application on shallow soils (-----) and deep soils (____) in Haute-Normandie (France)
in 2005 and 2006. Each symbol represents one field experiment
Heterogeneity of P (P2O5) and K (K2O) soil rates
The P and K soil rate maps interpolated by the Kriging method (Figure 6) are
inversely related to wheat yield maps (Figure 4). When the yield is high the soil rate is
low. The crop requirement reduced the soil rate during a period of time when the
management was uniform.
65
Figure 6. P and K soil rate maps: P2O5 in mg/kg (a); K2O in g/kg (b).
Geostatistic interpolation was performed by the Kriging method
Minimum soil samples to describe the P and K soil rates heterogeneity
We quantified the spatial structure of the P and K soil rates and we produced a
prediction for each chosen soil sampling frequency. The most accurate map was
produced with a frequency of four soil samples per hectare. This is consistent with
the result shown in literature [Bruchou and Mary, 2007]. But this sample frequency is
too expensive for the farmers. For the models produced from the one sampling data
per hectare we demonstrated that there is no spatial dependence [Becu et al., 2008].
For the models produced from the two sampling data per hectare, the spatial
variability of P and K soil rates is satisfactorily described, we created the interpolated
maps which we used to calculate the fertiliser recommendation that will be applied by
SSCM. Thus, two samples per hectare are sufficient to describe the soil rate
variability in our experimental conditions [Becu et al., 2008].
Economic interest of precision agriculture from 2003 to 2008
We calculated an estimation of the evolution of the economic interest of
precision agriculture from 2003 to 2008, a period marked by strong price fluctuation
for agriculture inputs and outputs (Figure 7). The price references used come from
"AGRESTE-la statistique agricole" (for use of price index and quantity in calculation
of technical efficiency, Balk 1998). The price of inputs, related to the harvest of year
N is an average of the monthly prices from 01/06/N-1 to 31/05/N. for this period, it
was considered there was no modification of the technical procedures concerning the
fertilizer application.
66
140
Input prices (€)
120
100
80
60
40
20
Ju ne 2 00 7
Ju ne 2 00 6
Ju ne 2 00 5
Ju ne 2 00 4
Ju ne 2 00 2
Ju ne 2 00 3
0
Ammonium Nitrate
Super phosphate
Potassium chlorate
Fuel
Figure 7. Price evolution of some principal agricultural inputs from 2003 to 2008
Expense in partial level in Partial Budget (€/ha)
We calculated the profitability during a period from 2003 to 2008 based on
fertilizer recommendation for SSCM on a oil seed rape and wheat crop rotation Oil
seed rape –wheat-wheat. The gains connected to the use of SSCM are mainly due to
a lower consumption of fertilizer. Indeed, on average, for the period 2003-2008, 50%
of the gains were created by a reduction in supplies in N, 23% by a reduction of
supplies of P and 27% by a reduction of supplies in K. However, the investment
necessary for SSCM practice (SSCM equipment and information cost) involve
economic losses which represent in the partial budget 41% of the total estimated
gains.
These observations vary according to the year under study. Comparative
studies carried out for the period 2003-2008 demonstrate a constant increase in the
balance of the partial budget (∆). This evolution represents the increase of the
interest of the SSCM in preference to uniform fertilization (Figure 8). This increase
has been linked to the rise in the price of agricultural inputs.
280
260
∆
240
220
Uniform fertilizer expenses
200
180
160
SSCM Fertilizer expenses +
SSCM specific expenses
140
120
100
2003
2004
2005
2006
2007
2008
Modulated f ertilisation
Unmodulated f ertilisation
Figure 8. Evolution of the partial budget balance between 2003 and 2008
67
Therefore, in the context of the increasing cost of agricultural inputs we
demonstrated the rising interest in the use of precision agriculture [Llorens and
Perrin, 2009], in particular towards the end of the study period where the interest rose
39% in one single year (18,6 €/ha in 2007 to 25,7 €/ha in 2008).
Economic profitability simulations using the direct margin method
To generate the data base (Figure 2), we have classified the soils into three
levels of yield potential. For each yield level, we have defined three intrafield
heterogeneity rates for each nutrient (N, P and K, [Table 4]).
Table 4. Hypothesis concerning three levels of N, P and K heterogeneity of HauteNormandie’s soils (medium, minimum and maximum) in relation with three levels of
crop yield potentials
Soils of HauteLow potential
Medium potential High potential
Normandie
17%
58%
25%
maximum
Nitrogen
12%
68%
20%
medium
8%
74%
18%
minimum
30%
15%
55%
maximum
Phosphorus medium
13%
37%
50%
5%
70%
25%
minimum
15%
20%
65%
maximum
Potassium medium
9%
41%
50%
0%
70%
30%
minimum
These rates are based on the results most commonly observed in HauteNormandie in a 6-year field investigation (70 farmers and over 9000 ha analysed)
undertaken by the GRCETA27 [Becu, unpublished data]. The fertilizer amounts are
calculated from these rates.
Whatever the soil heterogeneity and the crop system surface area (CSSA), the
direct profit margin (DPM) obtained through SSCM is higher than that of uniform
application (Figure 9). The direct margin gains calculated using Olympe vary from 32
to 47 euros per ha (Figure 9). We can see that an increase of farm surface area and
soil heterogeneity raises the profitability of site-specific farming compared to uniform
farming. These results were obtained from experimental data [Duval et al., 2007,
Bécu et al., 2008], expert advice and representative farm data (typical systems of
Haute-Normandie).
68
Figure 9. Direct margin gains linked to the precision agriculture use in the cereals
systems in Haute-Normandie (NW France)
From an environmental point of view, SCCM allows a significant reduction in
sprayed amounts for all heterogeneity levels. This difference increases according to
the CSSA. For example, for nitrogen, sprayed amounts diminished by more than 1500
kg for a 95 ha CSSA, more than 2300 kg for a 145 ha CSSA and more than 3880 kg
for a 240 ha CSSA.
Even if fixed costs are higher in SSCM (€2075 as opposed to €470), whatever
the farm surface area, the profits generated by the reduction in amounts of sprayed
fertilizer and traction operation efficiency are higher than outlay. Indeed, the traction
material is more efficient in terms of working time in SSCM than in UA which
contributes to a higher profitability (Table 5). A larger surface area is covered each
hour by SSCM material allowing farmers to double their application.
69
Table 5. Global direct margin (in euros) of the three cereal crop systems (95 ha [a], 145
ha [b], 240 ha [c]) in SSCM and UA related to different rates of silty soils heterogeneity
in Haute-Normandie (France)
cereals
oil seed
rape
Crop product
peas
other
TOTAL CROP PRODUCT
fertilizer
Variable
seeds
costs
chemicals
other
Specific fixed costs
TOTAL SPECIFIC
COSTS
DIRECT PROFIT
MARGIN
DIRECT PROFIT
MARGIN GAIN
cereals
oil seed
rape
Crop product
peas
other
TOTAL CROP PRODUCT
fertilizer
Variable
seeds
costs
chemicals
other
Specific fixed costs
TOTAL SPECIFIC
COSTS
DIRECT PROFIT
MARGIN
DIRECT PROFIT
MARGIN GAIN
Cereal crop system 95 ha (a)
Precision Agriculture
Uniform
Minimum
Medium
Maximum
Application
Heterogeneity Heterogeneity Heterogeneity
75215
76620
76395
76395
13090
8500
9100
105905
16251
8387
14825
2166
470
13310
8670
9100
107700
13943
8387
14825
1642
2075
13200
8670
9100
107365
13244
8387
14825
1642
2075
13200
8500
9100
107195
13097
8387
14825
1642
2075
42099
40872
40173
40026
63806
66828
67192
67169
3022
3386
3363
Cereal crop system 145 ha (b)
Precision Agriculture
Uniform
Minimum
Medium
Maximum
Application
Heterogeneity Heterogeneity Heterogeneity
115225
117080
117135
117135
20020
12750
13900
161895
24864
12783
22665
3314
470
20350
13090
13900
164420
21329
12783
22665
2511
2075
20240
12920
13900
164195
20373
12783
22665
2511
2075
20350
13090
13900
164475
20069
12783
22665
2511
2075
64096
61363
60407
60103
97799
103057
103788
104372
5258
5989
6573
70
cereals
oil seed
rape
Crop product
peas
other
TOTAL CROP PRODUCT
fertilizer
Variable
seeds
costs
chemicals
other
Specific fixed costs
TOTAL SPECIFIC
COSTS
DIRECT PROFIT
MARGIN
DIRECT PROFIT
MARGIN GAIN
Cereal crop system 240 ha (c)
Precision Agriculture
Uniform
Minimum
Medium
Maximum
Application
Heterogeneity Heterogeneity Heterogeneity
192695
196235
195785
195505
33110
20400
23080
269285
41418
21140
37700
5521
470
33660
20740
23080
273715
35573
21140
37700
4179
2075
33550
20740
23080
273155
33889
21140
37700
4179
2075
33550
20740
23080
272875
33363
21140
37700
4179
2075
106249
100667
98983
98457
163036
173048
174172
174418
10012
11136
11382
DISCUSSION
Influence of plot regrouping and agricultural practices on the heterogeneity of rates of
P and K and the availability of these nutrients for crops.
The regrouping of land and agricultural practices modify the methods of use of
the cultivated areas. In our example (Figure 3), the cropping history of this plot since
1947, partially explains the three resulting soil rates zones. These zones create soil
conditions which currently influence the availability of these nutrients on crops.
Indeed, according to Marschner [1995] the conditions of nutrient uptake by
plants are: the root interception (soil volume displaced by root volume), the nutrient
diffusion in the soil solution towards the roots (nutrient transport along the
concentration gradient) and the mass flow transport of bulk soil solution along the
water potential gradient (driven by the transpiration). The effect of these three
conditions is considerably reduced in zone 2 (Figure 3) where the soil is shallow, as
the roots have problems in developing in this soil, the yield is less, therefore
transpiration and consequently mass flow is reduced. The reduction of these two
parameters lowers the nutrient uptake and induces an accumulation of P and K in
this zone, more noticeable for P which is less mobile in soil [Fardeau et al., 1991,
Lewis and McGechan, 2002, Mollier et al., 2008] than is K [Holmqvist et al., 2003].
This element can be leached in this zone.
If, on this area, homogeneous fertilization is applied, the calculation of P and K
fertilizer amount is based on the lowest soil test found in zone 1 which is the most
representative of the plot (the largest area, Figure 3). These low rates of P and K
would imply fertilizing advice beyond that required for zone 2. Thus, the nutrient mass
balance is extremely positive because the amount added by the fertilizer is higher
71
than that removed by the plant during harvesting, thus, little by little, the rate of this
zone will increase creating considerable heterogeneity in this part of the field. This is
more important for P than K due to difference in mobility previously mentioned.
Minimum samplings per hectare to be carried out to describe heterogeneity
applicable for precision agriculture
In this study, the variability of P and K rates can be considered as permanent
variables in soil [Bruchou and Mary, 2007] which is relevant to the dynamic of these
nutrient in the soil [Lewis and McGechan, 2002; Holmqvist et al., 2003]. A study
based on 8 soil parameters and 10 plots (Piraux, 2005) and a far wider grid sampling
than ours (10 samples par hectare) has shown that in a same field the variability and
the spatial structure of the pedological parameters of interest can vary considerably
from one plot to another: some seem generally more variable and better structured
than other.
In our case, models of variograms for sampling frequency of two samples per
hectare allowed us to demonstrate mathematically that we have sufficiently shown
the heterogeneity of P and K in relation to the precision of the site specific crop
management material available. Two analyses per hectare is an operational sampling
for SSCM in our pedoclimatic conditions.
Assessment of the environmental profitability
The assessment of farm input use and efficiency are on the agenda in many
countries with intensive agricultural production, as a response to the increased
interest in the environmental performance of different farming systems. Input–output
accounting systems can facilitate farmers and advisors to improve the environmental
performance of both crop and livestock production [Halberg et al., 2005].
In this survey, thanks to the underlying Kriging method we could adjust SSCM
maps allowing the reduction of the excess or potential lack of P and K [Becu et al.,
2008]. In zone 2 (Figure 3) where there is an excess of P, the environmental risk of
eutrophication is low because there is no river close to this field. But if it were not the
case, the use of SSCM would be an appropriate technique for reducing P
environmental impact. To complete this physicochemical approach it would be
interesting to test the influence of SSCM on soil microbiology: reducing the lacks and
the excess of P and K within the field could considerably change the structure of the
microbial population and thus the nutrient mobility towards the root [Deneux-Mustin et
al., 2003].
Uniform management implies surplus which could lead to P accumulation in
soils, making them long-term diffuse sources of P loss to water [Hooda et al. 2001].
Soil available P has been considered as an important indicator in both evaluating the
soil P supply capacity and determining the P fertilization rate and P loss risk from
runoff [Liu et al., 2007].
Economic interest of precision agriculture in fertilisation management
In a context where world agricultural supply prices rise significantly, SSCM can
provide an efficient solution. However, the drop in petrol prices of the second
semester 2008 could result in a decrease in the price of fertilizer in 2009; so far not
observed.
Whatever the situation the products (yield*price) are higher in PA than in
uniform application (UA). Fertilizer utilization Costs is systematically lower when PA
is practised [Bourgain et Llorens, 2009].
72
The net gains in terms of direct profit margins vary between 32 and 47 €/ha.
These variations, favourable to the site specific management of inputs, are created
through an optimisation of yields in addition to more efficient management costs.
Other technical criteria such as the composition of fertilizers could have been
taken into account. In UA farmer can use compound fertilizers (binary [PK] or ternary
[NPK]). In this case, they can reduce the number of tractor operations (by 2 PK
fertilizer and by 3 for NPK) which limit mechanisation costs. However, in the study
region farmers are still advised to use simple fertilizers because it is easier to adapt
the dose to the plant needs (Bourgain and Llorens, 2009). Indeed, the proportions in
PK rates in the soil can be quite different from those of commercial compound
fertilizer. In this paper we have chosen to simulate the real use of fertilizer by
farmers: the objective being to generate decision-making criteria which can be
applied by agricultural advisers.
The results of our simulation show that the interest of SSCM for cereal
production systems is higher when the farm surface area is larger [Bourgain and
Llorens, 2009] which confirms our previous partial results [Duval, et al., 2007 and
Bécu et al., 2008]. This is due to the distribution of the fixed costs on a larger surface
area. The logic of these results is established but our methodology and simulator
provide precise estimations. This could provide a convincing argument for these
application techniques [Jochinke et al., 2007] which reconcile economy and
environment and allow their assessment whichever production system chosen.
Within the perspectives of this research we plan to study these results, in
terms of optimal profitability using a mathematic model based on the production
function. Another follow up study could draw up an environmental criteria simulation
by using Olympe in terms of the reduction of the impact of fertilizers in particular
concerning water resources (especially for nitrates and phosphates).
Concerning the agricultural policy-makers, the optimisation of fertiliser use
thanks to the modulation of inputs can create an optimisation of protein production
and thus an increase in profits for producers wishing to improve the protein level
[Bongiovanni et al 2007]. From the industrial point of view, development of such
models could have an effect on the choice of new material to be developed. For
policy-makers, the focus is mainly on the environmental aspects associated with the
economic viability of this sector in order to provide efficient management of water
resources [Rejesus and Hornbaker, 1999].
CONCLUSION
The methodology used supported by the simulator models and experimental
trials has shown the economic interest of PA in the pedoclimatic context of HauteNormandie. Thus applying PA reduces input costs and increases profit margins
without affecting yields. This is particularly relevant at a time of increasing input
prices and global food insecurity.
In follow up studies we will simulate all systems and practices under different
economic situations. We will also simulate environmental externalities such as
potential pollution from excess application of nitrogen or phosphorus, and thus
enabling us to offer a wider range of decision-making criteria to agricultural advisors
and policy makers.
73
Acknowledgements
This work could not have been carried out without the experimental field trials
undertaken by two French farmer groups (GRCETA de l’Evreucin and DEFISOL 27).
We should, first of all, like to thank the advisors of GRCETA Vincent Debandt and
David Mahieu for their comments and their expert agronomic advice. We should also
like to thank Maxime Bécu and Charles Duval from Défisol27 for their technical data
concerning soil description and crop systems which was essential for this research.
Finally, a special mention for Hazel Favre and Bill Edmonds for helping us to find the
right words to create this document in English.
BIBLIOGRAPHY
1. AGRESTE, 2008 et 2009: bulletins de conjoncture de août 2008 et janvier 2009,
2. Atherton B.C., Morgan M.T. Shearer S.A., Stombaugh T.S., Ward A.D. 1999:
Site-specific farming : a perspective on information needs benefits and
limitations. Journal of soil and water conservation 54 (2), 455-460.
3. Bécu M., Duval C., Debandt V., Taïbi S. et LLorens J-M. 2008: Influence of the
past history of the addition of fertilizers on the intra field variability of the rate of
P and K in the soil: potential of site-specific fertilization. Ageng 2008 International Conference on Agricultural engineering & Industry Exhibition.
Hersonissos – Crete 23-25 June 2008, Greece.
4. Bongiovanni, R.G., Robledo, C.W. and Lambert, D.M. 2007: Economics of sitespecific nitrogen management for protein content in wheat. Computers and
Electronics in Agriculture 58: 13–24.
5. Bourgain O., Michaud M. 2006: Evaluation of the economic impact of
agricultural practices aiming at limiting muddy flows : farmers from the Plateau
du Neubourg (France). Workshop « the socio-economic aspects in the
management of soil erosion » Strasbourg, France 7th – 8th April 2006.
6. Bourgain O. and Llorens JM. 2009: Methodology to estimate economic levels of
profitability of precision agriculture : simulation for crop systems in HauteNormandie. In EFITA Conferences 09, Wageningen Academic publishers p. 9198.
7. Bruchou C., Mary B. 2007: Analyse statistique de caractéristiques permanentes
et non permanentes du sol d’une parcelle agricole. In : Guérif M. et King D.
Agriculture de précision. Editions Quae, Versailles p 147-164.
8. Chambre d'agriculture Normandie, BCMA, CUMA, 2008: Coûts d'utilisation
prévisionnels des matériels agricoles.
9. Cordonnier P., Carles R., Marsal P. 1970: Economie de l'entreprise agricole.
Editions Cujas, France, pp. 307-320.
10. Dailey, A.G., Smith, J.U. and Whitmore, A.P. 2006: How far might medium-term
weather forecasts improve nitrogen fertiliser use and benefit arable farming in
the England and Wales? Agriculture, Ecosystems and Environment 117: 22–28.
11. Deneux-Mustin S., Roussel-Debet S., Mustin C., Henner P., Munier-Lamy C.,
Colle C., Berthelin J., Garnier-Laplace J., Leyval C. 2003: Mobilité et transfert
racinaire des éléments traces : influence des microorganismes. Tec & Doc
Lavoisier, Paris.
12. DeVuyst, E. A. and Halvorson, A. D. 2004: Economics of Annual Cropping
versus Crop–Fallow in The Northern Great Plains as Influenced by Tillage and
Nitrogen. Agron. J. 96:148–153.
74
13. Duval C., Debandt V., Eveillé J-P., Mahieu D, Taïbi S., LLorens JM. 2007:
Influence of the pedo-climatic variability in Haute-Normandie (NW France) on
the intra field spatial variability on yields of wheat and oilseed rape. In Précision
agriculture 07, Wageningen Academic publishers p. 87-94.
14. Fardeau J-C., Morel C., et Boniface R. 1991: Cinétique de transfert des ions
phosphate du sol vers la solution du sol : paramètres caractéristiques.
Agronomie, (11) p. 787-797.
15. Fleming, R. A. and Adams, R. M. 1997: The Importance of Site-Specific
Information in the Design of policies to Control Pollution. Journal of
environmental economics and management 33: 347-358.
16. Haefele, S.M. and Wopereis M.C.S. 2005: Spatial variability of indigenous
supplies for N, P and K and its impact on fertilizer strategies for irrigated rice in
West Africa. Plant and Soil 270: 57–72.
17. Halberg N., Verschuur G., Goodlass G. 2005: Farm level environmental
indicators; are they useful? An overview of green accounting systems for
European farms. Agriculture, Ecosystems and Environment (105) p.195–212
18. Holmqvist J., Øgaard A.F., born I. O., Edwards A.C. , Mattsson -L., Sverdrup H.
2003: Application of the PROFILE model to estimate potassium release from
mineral weathering in Northern European agricultural soils. European Journal of
Agronomy (20) p.149-163.
19. Hooda P.S., Truesdale U, V.W., Edwards A.C., Withers P.J.A. Aitken M.N.,
Miller A., Rendell A.R. 2001: Manuring and fertilization effects on phosphorus
accumulation in soils and potential environmental implications. Advances in
Environmental Research (5), p.13-21.
20. Jochinke, D. C., Noonon, B. J., Wachsmann, N. G. and Norton R. M. 2007: The
adoption of precision agriculture in an Australian broadacre cropping system—
Challenges and opportunities. Field Crops Research 104: 68–76.
21. Koch, B., Khosla, R., Frasier, W. M., Westfall, D. G. and Inman D. 2004:
Economic Feasibility of Variable-Rate Nitrogen Application Utilizing Site-Specific
Management Zones. Agron. J. 96:1572–1580.
22. Le Bars, M., Allaya, M. and Le Grusse, P. 2005: Manuel d’utilisation d’Olympe.
CHIAM. p:6
23. Lewis D. R., McGechan M. B. 2002: A Review of Field Scale Phosphorus
Dynamics Models. Biosystems Engineering 82 (4), 359–380
24. Link, J., Graeff, S., Batchelor W. D. and Claupein W. 2006: Evaluating the
economic and environmental impact of environmental compensation payment
policy under uniform and variable-rate nitrogen management. Agricultural
Systems 91: 135–153.
25. Liu J-L., Liao W-H., Zhang Z-X., Zhang H-T., Wang X-J. and Meng N. 2007:
Effect of phosphate Fertilizer and Manure on Crop Yield, Soil P Accumulation,
and the Environmental Risk Assessment. Agricultural Sciences in China, 6 (9),
p. 1107-1114.
26. Lobell D.B. 2007: The cost of uncertainty for nitrogen fertilizer management: A
sensitivity analysis. Field Crops Research 100: 210–217.
27. Llorens J-M 2001: La méthode de raisonnement de la fertilisation PK: un outil
pour l’agriculture. COMIFER-GEMAS, 5ème Rencontres de la fertilisation
raisonnée et de l'analyse de terre, Blois, France, p.99-108.
28. Llorens JM. and Perrin O. 2009: Evolution from 2002 to 2008 of the economic
interest of precision agriculture in fertilisation management. In EFITA
Conferences 09, Wageningen Academic publishers p. 98-105.
75
29. Makowski, D., Wallach, D., and Meynard, J-M. 1999: Models of Yield, Grain
Protein, and Residual Mineral Nitrogen Responses to Applied Nitrogen for
Winter Wheat. Agronomy Journal. 91 377–385.
30. Marschner H., 1995. Mineral nutrition of higher plants, second edition. Academic
press, London.
31. Melakeberhan, H. and Avendaño, F. 2008: Spatio-temporal consideration of
soil conditions and site-specific management of nematodes. Precision Agric 9:
341–354.
32. Meynard, J.M., E. Justes, J.M. Machet and S. Recous. 1997: Fertilisation
azotée de cultures annuelles de plein champ. p. 183–199. In G. Lemaire and B.
Nicolardot (ed.) Gestion de l’azote dans les agrosystemes. INRA, Paris.
33. Mollier A., De Willigen P., Heinen M., Morel C., Schneider A., Pellerin S. 2008:
A two-dimensional simulation model of phosphorus uptake including crop
growth and P-response. Ecological Modelling (210) p. 453–464.
34. Olsen, S.R., Sommers, L.E. 1982: Phosphorus, Methods of Soil Analysis. Part
2. Chemical and Microbiological Properties. American Society of Agronomy Soil
Science Society of America, Madison, Wisconsin, USA, pp. 403–430.
35. Piraux F. 2005: Variabilité de l’hétérogénéité : une approche pluri-régionale.
Colloque « Fertilisation et société »,15 et 16 novembre, Blois, France.
36. Rejesus, R. M. and Hornbaker R. H. 1999: Economic and environmental
evaluation of alternative pollution-reducing nitrogen management practices in
central Illinois. Agriculture, Ecosystems and Environment 75: 41–53.
37. Robert, P. C. 2002: Precision agriculture: a challenge for crop nutrition
management. Plant and Soil 247: 143–149.
38. ROSACE, 2005: Farming system observation network for advices and studies.
39. Teyssier, D. 2000: Index des prix et des normes agricoles 2006-2007. 19ème
édition, Synthèse Agricole, Editions Lavoisier, Paris, 199 p.
40. Sartori, L., Basso, B., Bertocco, M., Oliviero, G. 2005: Energy Use and
Economic Evaluation of a Three Year Crop Rotation for Conservation and
Organic Farming in NE Italy. Biosystems Engineering 91: (2) 245–256.
41. Swinton SM, Lowenberg-DeBoer J. 1998: Evaluating the profitability of sitespecific farming, Journal of precision agriculture 11 (4), 439-446.
76
Chapter 7
ECOLOGICAL ASPECTS OF SOIL
COMPACTION BY WHEELS
OF AGRICULTURAL VEHICLES
Jerzy BULIŃSKI
INTRODUCTION
Problem of excessive soil compaction concerns all agricultural sectors dealing
with field production, including typical agricultural crops, vegetables, horticultural
production or fruit culture. It can be observed also in human activities in forestry,
therefore, in every place where the root system of cultivated plants, bushes or trees
can not properly develop or function as a result of soil stress caused by the pressure
of wheels of implements or tractors, or where it is mechanically damaged. However,
the most noticeable effects of soil compaction can be found in agricultural field
production, connected with human activity and application of technical means of
production.
Cause of this situation is well known and has been described in references for
many years. It is believed that soil compaction occurs when the stress values created
by wheel of vehicles running over the field exceed a natural soil strength [Van den
Akker 1994]. Majority of published works concerning compaction [Botta et al. 2004,
Alakukku et al. 2003, Güclü et al. 2002, Way et al. 2005] point out that the main
factors leading to excessive compaction of agricultural soils are changes in
technologies of mechanized field operations. These changes result from economic
limitations which force application of high-capacity outfits driven by powerful tractors
[Abu-Hamdeh 2003]. According to Alakukku et al. [2003] the sale of brand new
tractors of power above 44 kW increased in Germany more than twice over 6 years.
Similar trend can be found in Poland where during last 9 years the number of tractors
used in agriculture increased by over 20%, while average tractor power increased by
over 25% [GUS 2009]. There is a strict connection between weight and power of
tractors in Podlaskie province [Marczuk 2006] and an increase in tractor power
requirement for driving machines usually results in the increased field pressure
loading.
77
Figure 1. Dependence between weight and power of tractors in farms of Podlaskie
province [Marczuk 2006]
Trend towards application of heavier machines and tractors can be found in
entire Europe. The weight of sugar beet combine harvesters produced currently can
reach during work at full load even 60 Mg with axle pressure over 10 Mg [Nosalewicz
2005]. Importance of the problem was emphasized in 1980 when there were started
serious international studies on the impact of high axle pressure vehicle wheels, and
International Working Group for Soil Compaction by Vehicles of High Axle Loads
(within ISTRO) was set up. Initially, this team involved scientists from Denmark,
Finland, Norway, Sweden, Netherlands, Canada and USA, and afterwards scientists
of 35 European scientific institutes and universities were involved (including Poland.
Hitherto investigations on soil compaction [Soane et al. 1995, Güclü et al. 2002,
Halvorson et al. 2003, Meyer et al. 1996] point out clearly that the created wheel
pressures lead to very unfavourable changes in physical, chemical and biological
properties, both in topsoil and subsoil layers; they influence intensity of mineral and
organic substances, thus, the soil fertility and potential yielding ability. It is also
known that restoration of the initial state of highly compacted deeper layers of the soil
profile is very difficult.
ASSESSMENT OF SOIL COMPACTION
Soil density and soil compaction are the commonly used indices [Ronai and
Shmulevich 1995; USDA-NRCS, 1996; Lipiec 2002]. Various measuring devices are
used in field investigations to determine these parameters. Unification of applied
methods is important with respect to reliability of obtained results of investigations,
correctness of conclusions and possibility of comparisons between results achieved
in various scientific institutions. Under Polish conditions, procedures of soil properties
measurements can be found in PN-88/B-04481 Standard. However, in many cases
there are used the methods and devices which make difficult interpretation or
comparison of results. Samples in investigations on soil density are usually taken by
open pit method and by horizontal driving in the separate cylinders into investigated
layer (Fig. 2a) – often differing in volume or construction (e.g. edge for cutting the
sample). This measurement can be executed with the use of Eijkelkamp probe
vertically driven in soil and equipped with a set of cylinders, placed one above
another in the probe’s pipe (Fig. 2b). Different measurement methods can lead to
different findings (Fig. 2c).
78
b/
a/
c/
Figure 2. Effect of measurement method for soil density on obtained values
[Powałka and Buliński 2003]
Soil compaction is usually measured with the use of cone penetrometers
driven in soil manually or mechanically using electric or hydraulic drive. The ASAE
Standard [1993] commonly used in ground investigations recommends application of
probes with a cone end of apex angle 30º, driven in soil at speed 30 mm/s. The
Standard allows for application of two types of cones – smaller one of diameter 12.83
mm and cross section at the base 130 mm2 (0.505 in.) and bigger one of diameter
20.27 mm and cross section 323 mm2 (0.798 in.). Differentiation of the cone size is
determined by measurement conditions: bigger cone for light soils, smaller cone for
heavy soils. Other cone dimensions can also be found in investigation methodology.
In soil bin investigations carried out by Buliński and Majewski [1998] the soil
compaction was measured with the use of two cone probes: St1 – cone diameter -
79
11.40 mm, apex angle – 600, and St2 – cone diameter -20.27 mm, apex angle – 300 .
Three speeds of probe driving in soil amounted to 0.6 - 1,2 - 1.8 m min-1 and five soil
compaction levels (expressed by soil volumetric density) were equal to 1.65,-1.67,1.69,-1.77,-1.81 g cm-3. The cone dimensions St2 and highest speed of driving in soil
corresponded to ASAE S313.2 Standard in the range of measurement execution and
equipment used. Statistical analysis on the results obtained under same conditions of
measurements proved significance of differences In soil compaction measured with
the use of both probes. Comparison between probe indications is presented in Fig. 3;
it can be seen how important is observing the standard recommendations and
determination of precise measurement conditions.
900 St [kPa]
2
800
700
600
500
400
300
200
100
0
St1 [kPa]
0
200
400
600
800
1000
1200
1400
Figure 3. Values of soil compaction measured with probes St1 i St2
[Buliński and Majewski 1998]
In field investigations one can find penetrometers acting dynamically (impact
probes) which determine the soil compaction on the basis of weight’s impact number
(descending on the probe’s shank from a defined height), needed to drive the cone to
a given depth, e.g. 0.15 m. This measurement method is quite different and its results
can not be related to the results obtained by application of static probes.
EFFECT OF COMPACTION ON CHANGES IN SOIL PHYSICAL PROPERTIES
One of the factors which change considerably the physical properties of soil in
wheel track zone is multiple running of outfit wheels over the same track. Horn et al.
[2003] found that as a result of this running the soil is deformed both horizontally and
down its profile (Fig. 4).
80
Figure 4. Effect of multiple running of tractor on soil deformation [Horn et al. 2003]
As it is evident from investigations the soil density gradually increased in entire
profile of arable layer up to 10th run. According to researchers the soil compaction
can be initially connected with decreasing space between the soil aggregates and
separate aggregates are durable. As stress increases further, the layer’s
homogeneity in the profile along with damage of its structure and changes in
configuration of particles. Under pressure of wheels the part of pores elongated in the
layer under wheel track reduce their dimensions, loose continuity and change
configuration from vertical to parallel in relation to the field surface [Pagliai 1987,
Słowińska-Jurkiewicz and Domżał 1991]; this makes difficult a vertical penetration of
plant roots through the layers created. In the compacted layers there were found
[Larson et al. 1994, Lowery and Schuler 1994] the limited gas exchange, decreased
water permeability and its availability for plants, deterioration of draining system
functioning, worse development of root system and poorer uniformity of plant growth.
Increase in soil density is accompanied by a decrease in biggest pores [Buliński,
Niemczyk 2007], which are taken mainly by the roots of cultivated plants, enable soil
aeration and adjust its hydraulic conductivity. Kuht et al. [2003], investigating
changes in nitrogen content in plants during germination and shooting phases in
variably compacted soils, found no significant differences in the content of this
component in plants growing under higher moisture content. An increase in soil
compaction above 1.58 g cm-3 led to a clear decrease in nitrogen content in the
plants’ dry mass during germination. It has been pointed out since many years
[Trzecki 1969, Lipiec 2002] that maximal ability for vertical overcoming soil resistance
by plant roots does not exceed 3 MPa, while at compaction bigger than 2 MPa the
root development is limited [Copas et al. 2009]. Differences in percentage distribution
of roots in particular layers of not compacted and compacted soil profile (Fig. 5) can
differ substantially [Lipiec et al. 2003].
81
Figure 5. Dependence between root distribution of spring barley and soil density in
arable layer [Lipiec i inni 2003]
In the case of root plants such as sugar beet or carrot these difficulties can
lead to deterioration of their technological value.
Investigations of Powałka [2005] carried out with application of a camera
recording image of field surface in the zone of wheel tracking proved that the front
tractor wheel moving over freshly ploughed field creates a soil wedge (Fig. 6a), and
the soil is pushed aside creating in a superficial layer the linear fractures along
planes of biggest stresses (Fig. 6b). According to the author, this phenomenon
concerns mainly the narrow front wheels of light tractors, when side shifting of soil
includes almost entire strap aside the track of width 300 mm. When tyres of front
tractor wheels are wide, width of a side zone of superficial deformations does not
exceed 100 mm. Referring these results to agricultural practice one can find that
passage of tractor over the freshly ploughed field leaves behind the zone of changed
soil properties in a strap 09-10 m wide along the track axis. Compacting effect of
agricultural vehicle wheels is not limited to superficial layers of field.
Figure 6. Passage of wheel on loosened soil layer: a/ - longitudinal section of field
profile, b/ - soil fractures and deformations along the length and breadth of passage
[Powałka 2005]
The substantially compacted zones, reported in many investigations, reached
beneath the range of annual tillage operations, especially if performed at substantial
soil moisture content. Passages of 6-row sugar beet combine harvester of total mass
82
35 t resulted in strong soil compaction and decreased water conductivity in a layer
0.3-0.5 m by almost 90% [Arvidsson 2001], while under conditions of bigger moisture
content the changes in soil compaction were found even at 0.7 m, and an increased
compaction (penetration resistance) was found on light (sand) and heavy (clay) soils.
It is believed that soil resistance to stress decreases with an increase in water
content [Horn, Lebert 1994]; it should be noted that every soil has its specific
moisture content, when pressing causes the biggest compaction. This moisture
ranges 14% for clayish sands and sandy clays to 26% for dusts and heavy clays
[Buliński 1998] which means that soil reaction to pressures often can vary, even
during one day when it rains [Sommer and Brunotte, 2003]. In determination of soil
susceptibility to compaction PD (Packing Density) index is proposed, which includes
soil volumetric density, structure, organic matter and clay particle contents. The index
is calculated with equation: PD = Db + 0.009×C [Van Den Akker and Hoogland 2009],
where PD - packing density in g.cm-3, Db – volumetric density in g cm-3, and C - clay
content (weight %). There are three ranges of PD parameter: low < 1.4, medium 1.4
– 1.75, high > 1.75. The higher PD parameter value, the lower soil susceptibility to
compaction. Since soi compaction results from the wheel impacts, Prinker and
Grečenko [2009] proposed numerical index CC (Compaction Capacity) for
determination of the risk of harmful soil compaction by a loaded wheel.
An external effect of soil compaction is a rut created within the wheel passage;
his leads to a transverse irregularity of field surface. Depending on soil type, its
moisture content during passage and wheel pressure, the rut depth can vary
substantially. It is evident from the Author’s own investigations that a single passage
of medium power tractor (Ursus 4512) over the ploughed clayish sand of moisture
content 10% created the ruts of depth 12 cm, and after eight passages the rut bottom
was situated 16 cm beneath field (Fig. 7).
Figure 7. Cross section of field surface within tractor passage: Y0 – shape of field
surface prior to passage, Y1 – after one passage, Y8 – after eight passages of tractor
Source: author’s investigations
The soil uplift marked on diagram aside the track, referred to field level prior to
passages (marked with broken line (Y0)), confirms the previously cited investigations
[Powałka 2005]. Index of transverse field irregularity after 8 passages increased over
2.5-fold in relation to a state priori to passages; it is also an important factor
83
influencing operational conditions of the machines and operation quality of their
working units.
Changes in operational conditions of agricultural implements and machines on
compacted soils are also connected to an increase in working resistance leading to
increased costs of operations, bigger fuel consumption and bigger frictional wear of
machines’ elements. It is evident from investigations [Buliński 1998] that substantial
increase in ploughing resistance occurred particularly in root crop cultivation
technologies, as a result of soil compaction by heavy combine harvesters and
trailers. After potato harvest the mean soil compaction increased by over 70%, dry
soil density by over 100% when compared with the period prior to harvest; this
resulted in an increase in ploughing resistance by about 25%.
EFFECT OF COMPACTION ON CROP YIELD
An important and measurable effect of excessive soil compaction is a
decrease in crop yield connected with deterioration of their growth and development
conditions. Figures reported in numerous references vary substantially depending on
crop type, soil conditions, climate and especially moisture content. Ablas et al. [1994]
reported that soil compaction caused a 38% yield decrease of maize for silage on
light soils at low precipitations, Jorajuria et al. [1997] found a 74% yield decrease of
grass within the passage track and 18% decrease in adjacent areas, Botta et al.
[2004] reported a 9 - 38% soybean yield decrease depending on passage intensity,
while according to Radford et al. [2000] the soil compaction decreased wheat
germination by over 20%. As a result of soil compaction the potato yield decreased
by 21-35%, sugar beet yield by 8.7% , wheat by12.9 – 13.6% (depending on date of
compacting), maize by 7.5% [Soil compaction 2001]. It can be seen that connected
reference data are very different, but they underline importance of the soil type. It
was also found that under various soil conditions the same loading of field with
machine passages can change the water infiltration intensity, soil fauna size, while
the cereal grain yield and straw yield on clay loam soil can be decreased by several
per cents; however, on silt loam soil it is not valid [Blanco-Canquia and Lal 2008].
Besides, application of mounted machines instead of trailed ones for fertilizing and
plant protection enabled to decrease a mean soil compaction by 9% and dry soil
density by 10%, due to a machine mass transfer on wider rear wheels of the tractor
and due to decrease in total wheel pressures [Buliński 1998]. [Gemtos et al. 1999]
maintain that evaluating the effect of soil compaction on crop yield one should
consider soil chemical properties and also statistical criteria of changes: significance,
range, differentiation and stability and also economic aspects.
STABILITY OF SOIL COMPACTION EFFECTS
Unfavourable effects of excessive soil compaction by vehicle wheels can be
found also in subsequent years. Hakansson [2007] reported that passages of outfits
of axle loading 10 t resulted in an increase of soil density at depth 50 cm and it was
visible after 8 years. According to Grogan [2003] passages of machines of axle
loading 20 Mg lead to a deep soil compaction and it can be found over 11 years and
even longer [Alakukku 2000], although these changes are most visible during 2-3
years after compacting and a decrease in crop yield can occur in the next year and
amount to 30%. Under favourable soil and climatic conditions the compaction effects
may not be obvious, but can occur in moisture shortage periods or after large
84
precipitation deteriorating the air-water balance; it can limit the range of root
penetration and possibility for nutrients uptake and also deteriorate the root
metabolism.
METHODS FOR REDUCING
VEHICLES’ WHEELS
SOIL
COMPACTION
BY
AGRICULTURAL
Suggestions on reducing soil compaction by agricultural vehicles’ wheels
reported most often in references can be generalized as the following principles.
Reducing field area compacted by wheels – this need results from the fact that
in typical field operation technologies such as cereal or potato cultivation over half of
field area at harvest time is covered by tracks of passages [Buliński 1998]. In some
places the wheel tracks overlap more than 20 times resulting in deep and substantial
changes in soil structure. In places of double overlapping the aeration porosity at
depth up to 150 mm can approach 15%, which is considered as limit value from the
viewpoint of proper conditions for plant development [Powałka 2005]. Special frame
devices reported in references [Raper 2005] astriding field straps of width equal to
several meters and moving along constant paths with mounted implements and
machines are still mainly within scientific investigations due to specific construction
and limitations. Traffic paths set up during sowing do not consider the outfit passages
executed prior to sowing, i.e when impact of wheels on a wet and loosened soil can
lead to substantial and unfavourable changes in soil properties. Some investigations
[Buliński 1998, Marczuk 2006] point out that problem can be solved by application of
pre-sowing paths set up at secondary tillage operations with the use of satellite
aiding systems of positioning accuracy equal to several centimeters. In that system
the area free of compaction after cereal harvest amounted to almost 64%. Great
importance can also have execution of ploughing with „on land” system and
application of multifunctional machine sets performing several operations during one
passage; however, in small farms it can be difficult with respect to required
equipment.
Reducing axle load of vehicles should concern mainly the heavy harvesting
machines like potato combine harvesters of total weight 50 Mg and single axle load
of 10 Mg. With respect to severe effects of strong soil compaction there is
recommended in Sweden to limit axle load of agricultural vehicles to 6 Mg [Danfors,
1974]. Activities connected with optimization of outfit mass distribution can be also
rational, since no critical states are created. The former Sowiet Union GOST
Standards determined values of pressures with respect to admissible stress values at
depth 0.5 m depending on soil type and moisture content at passage [Rusanov
1994]. The International Working Group (by ISTRO) dealing with problem of soil
compacting by vehicles of big axle loads assumed a load level of 50 KN as „high”.
Reducing tyre inflation pressure can be achieved by application of special lowpressure wide-profile tyres working in tandem or twin system, as well as by reducing
inflation pressure in typical tyres. Although it is rarely used in practice, it enables to
increase a contact area of tyre with ground by 3 - 19% [Powałka 2005]. Similarly, an
increase in tyre width enabled to decrease specific pressure values by almost 20%,
and soil compaction and density within the rut by about 10% [Buliński 1998]. Some
manufacturers give various inflation pressure values for transport and field
operations. Another solution often cited in references and used to decrease the
traction system pressure on soil and soil compaction is a tracklayer mechanism. New
materials and modern production technologies enable to obtain a quality of tracks
85
comparable to tyres, with better traction parameters which are significant during field
operations. As a result, the offer of heavy tractors and combines manufacturers more
and more often includes the products equipped with tracklayer traction systems.
Reducing operations on wet soils is difficult, since spring field operations are
mostly executed under such conditions. On wet soil the range of wheel compaction is
bigger. It is recommended to delay operations by several days, if possible. Equally
important is modification of plant cultivation systems by reduction of ploughing and
application of shallow soil loosing operations, if possible.
SUMMARY
The mentioned above methods aimed at reducing harmful compaction do not
exhaust all possibilities. This activity calls for a system approach with consideration to
farmer’s needs and requirements of environment connected to plant protection.
Therefore, there is the need of a broad knowledge on all factors influencing the soil
reaction to stress under specific conditions. It will to analyze the risk of creating
conditions leading to soil degradation. There is also important determination of
technologies for mechanized field operations particularly hazardous from the
viewpoint of soil compaction, and also discussion on possible technical solutions in
agricultural vehicles to reduce an impact of traction mechanisms on the soil.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
Abu-Hamdeh Nidal H. 2003: Soil compaction and root distribution for okra as
affected by tillage and vehicle parameters. Soil & Tillage Research 74 25–35.
Alakukku L, Weisskopf P., Chamen W.C.T., F.G.J. Tijink ,Van der Linden J.P.,
Pires S., Sommer C., Spoor G. 2003: Prevention strategies for field trafficinduced subsoil compaction: a review Part 1. Machine/soil interactions. Soil &
Tillage Research 73 145–160
Alakukku, L. 2000: Response of Annual Crops to Subsoil Compaction in a
Field Experiment on Clay Soil Lasting 17 Years. Advances in GeoEcology 32.
Catena Verlag, Reiskirchen, Germany, pp. 205–208.
Alblas, J., Wanink, F., Van den Akker, J.J.H., Van der Werf, H.M.G. 1994:
Impact of traffic-induced compaction of sandy soils on the yield of silage maize
in The Netherlands. Soil & Tillage Research. Vol. 29, 157–165.
Arvidsson J. 2001: Subsoil compaction by heavy sugarbeet harvesters in
southern Sweden. I. Soil physical properties and crop yield in six field
experiments. Soil & Tillage Research
60 67-78.
Arvidsson J., Trautner A., van den Akker J.J.H. 2001: Schjønning Per. Subsoil
compaction caused by heavy sugarbeet harvesters in southern Sweden. II.
Soil displacement during wheeling and model computations of compaction.
Soil & Tillage Research 60 78-89.
ASAE STANDARDS: 1994 Soil Cone Penetrometer. ASAE S312.3. Standards
Engineering Practices Data. American Society of Agricultural Engineers St.
Joseph. Mi. USA, p.687.
Blanco-Canquia H., Lal R. 2008: Axle-Load Impacts on Hydraulic Properties
and Corn Yield in No-Till Clay and Silt Loam. Agronomy Journal. 100:16731680
Botta G.F., Jorajuria D., Balbuena R., Rosatto H. 2004: Mechanical and
cropping behavior of direct drilled soil under different traffic intensities: effect
86
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
cropping behavior of direct drilled soil under different traffic intensities: effect
on soybean (Glycine max L.) yields. Soil & Tillage Research 78 53–58
Buliński J, Majewski Z. 1998: Ocena rozbieżności wyników pomiaru zwięzłości
gleby w różnych warunkach wykonywania badań. Problemy Inżynierii
Rolniczej, nr 2, 5-12.
Buliński J. 1998: Zagęszczenie gleby w różnych technologiach i związane z
tym opory orki. Rozprawy naukowe i monografie. Wyd. SGGW Warszawa. s.
140.
Buliński J., Niemczyk H. 2007: Effect of moisture content in soil on its
compaction caused by multiple running of agricultural vehicles. Annals of
Warsaw Agricultural University. Agriculture (Agricultural Engineering) No 50
Warsaw 3 – 8.
Copasa M. E., Bussana A. J., Driliasa M. J., Wolkowski R. P. 2009: Potato
Yield and Quality Response to Subsoil Tillage and Compaction. Agronomy
Journal 101:82-90
Danfors, B. 1974: Packning i alven. (Compaction of the Subsoil). Swedish
Institute of Agricultural Engineering, Uppsala, Sweden. Rep. S 24, 91 pp. (in
Swedish with English summary).
Gemtos T. A.;. Chouliaras N; Marakis St. 1999: Vinasse Rate, Time of
Application and Compaction Effect on Soil Properties and Durum Wheat Crop.
J. Agric. Engng Res. 73, 283 – 296.
Grogan J. 2003: Is Soil Compaction Limiting Yields?. Technical Bulletin. Issue
67: February.
Güclü Yavuzcan H., Vatandas M.,. Gürhan R. 2002: Soil strength as affected
by tillage system and wheel traffic in wheat –corn rotation in central Anatolia.
Journal of Terramechanics 39 23–34.
GUS – Środki produkcji w rolnictwie w roku gospodarczym 2008/2009. Wyd.
GUS. Warszawa 2009. s. 20.
Håkansson I. 2007: Swedish experiments on subsoil compaction by vehicles
with high axle load. Soil Use and Managemen Volume 1 Issue 4, Jan. 2007,
113–116.
Halvorson Jonathan J., Gatto Lawrence W.,. McCool D.K. 2003: Overwinter
changes to near-surface bulk density, penetration resistance and infiltration
rates in compacted soil. Journal of Terramechanics 40 1–24
Horn R, Way T., Rostek J. 2003: Effect of repeated tractor wheeling on
stress/strain properties and consequences on physical properties in structured
arable soils. Soil & Tillage Research 73 101–106
Horn R., Lebert M. 1994: Soil compactibility and compressibility. In Soane BD.,
van Ouwerkerk C. Soil Compaction in Crop production. Elsevier, Amsterdam.
45-70.
Jorajuria D., Draghi L., Aragon A. 1997: The effect of vehicle weight on the
distribution of compaction with depth and the yield of Lolium/Trifolium Soil and
Tillage Research Volume 41, Issues 1-2, March 1997, Pages 1-12
Kuht J., Reintam E., Loogus H., Nugis E. 2003: Changes of nitrogen
assimilation and intracellular fluid pH in plants of barley depending on bulk
density of compacted soils. Agronomy Research 1 (2), 139-144,
Larson W.E., Eynard A., Hadas A., Lipiec J., Soane B.D.; Ouwerkerk C
van.1994: Soil compaction in crop production. Developments in Agricultural
Engineering 11. 597-625;
Lipiec J. 2002: Wpływ stanu zagęszczenia gleby na wzrost i funkcjonowanie
87
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
roślin. Acta Agrophysica., 63, 49 – 62.
Lowery B.,Schuler R.T. 1994: Duration and effect of compaction on soil and
plant growth in Wisconsin. Soil & Tillage Research Vol 29, No 2-3, 205-210.
Marczuk T. 2006: Możliwości zmniejszenia ugniatania gleby kołami pojazdów
rolniczych przez dobór agregatu ciągnik – maszyna. Dysertacja IBMER
Warszawa 2006 r.
Meyer K., Joergensen R.G., Meyer B. 1996: The effects of reduced tillage on
microbial biomass C and P In sandy less soils. Applied Soil Ecology. 5 71-79.
Nosalewicz A. 2005: Ocena wpływu przejazdów ciężkich maszyn rolniczych
na zagęszczenie gleby przy użyciu modelu SOCOMO. Acta Agrophysica, 6(3),
753-759
Pagliai M. 1987: Micromorphometric and micromorphological investigations on
the effect of compaction by pressures and deformation resulting from tillage
and wheel traffic. Soil compaction and regeneration. Proc. of workshop on soil
compaction, consequences and structural regeneration processes, Avignon,
France. 37-38
Powałka M, Buliński J. 2003: Metody pomiaru parametrów gleby w warunkach
polowych. Przegląd Techniki Rolniczej i Leśnej 10/2003, str 14-16
Powałka M. 2005: Wpływ nacisków kół ciągników rolniczych na zagęszczenie
gleby w warstwie ornej. Dysertacja WIP. SGGW Warszawa 2005. s.84.
Prikner P., Grečenko A. 2009: Compaction capacity rating of off-road tires.
Járművekés Mobilgépek,II. Évf. No.1. 259-268
Radford B. J., Bridge B. J., Davis R. J., McGarry D., Pillai U. P., Rickman J. F.,
Walsh P. A., Yule D. F. 2000: Changes in the properties of a Vertisol and
responses of wheat after compaction with harvester traffic. Soil & Tillage
Research, Vol. 54, Issues 3-4, April 2000, 155-170.
Raper R.L. 2005: Agricultural traffic impacts on soil. Journal of
Terramechanics 42: 259–280
Ronai D., Shmulevich I. 1995: Comparative analysis of some soil compaction
measurement techniques. International Agrophysics, Vol. 9, nr. 3, s.227-223.
Rusanov, V.A. 1994: USSR standards for agricultural mobile machinery:
permissible influences on soils and methods to estimate contact pressure and
stress at a depth of 0.5 m. Soil & Tillage Research. 29, 249–252
Słowińska-Jurkiewicz A.,Domżał H. 1991: The structure of the cultivated
horizon of soil compacted by the wheels of agricultural tractors. Soil & Tillage
Research., Vol 19, No 2-3, 215-226.
Soane B.D., van Ouwerkerk C. 1995: Implications of soil compaction in crop
production for the quality of the environment. Soil & Tillage Research., Vol.35,
nr 1-2, s. 5-22,
Soil Compaction: Causes, Effects, and Control. University of Minnesota. Paper
FO-03115. 2001
Sommer C., Brunotte J. 2003: Lösungsansätze zum Problembereich
Bodenschadverdichtung in der Pflazenproduktion. Landnutzung und
Landentwicklung. Vol. 44, nr 5, s. 220-228.
Trzecki S. 1969: Próba wyznaczenia granicznych oporów gleby I zawartości
powietrza dla początkowego wzrostu korzeni roślin uprawnych (badania
wazonowe). Postępy Nauk Rolniczych. 6. 71 – 78.
USDA-NRCS 1996: Soil Quality Resource Concerns: Compaction. USDANRCS
Soil
Quality
Inst.
Ames.
IA
http://
www.statlab.iastate.edu/survey/SQ1/sqihome.shtml
88
45.
46.
47.
48.
Van den Akker, J.J.H, Hoogland T. 2009: Comparison of Risk Assessment
Methods to Determine the Subsoil Compaction Risk of Agricultural Soils in The
Netherlands ISTRO 18th Triennial Conference Proceedings, June 15-19,
Izmir-TURKEY T4 - 034 – 1-9
Van den Akker, J.J.H. 2009: Prevention of subsoil compaction by tuning the
wheel load to the bearing capacity of the subsoil. Proceedings of the 13th
International Conference of ISTRO Aalborg, Denmark, 537-542.
Way T.R,. Erbach D.C., Bailey A.C, Burt E.C,. Johnson C.E. 2005: Soil
displacement beneath an agricultural tractor drive tire. Journal of
Terramechanics 42 35–46.
PN-88/B-04481 – Grunty budowlane. Badania Laboratoryjne
89
Chapter 8
ANALYSIS OF THE WORKING
RESISTANCE OF THE WORKING UNITS
USED IN SIMPLIFIED TILLAGE
Krzysztof PIECZARKA
INTRODUCTION
Success in the modern agricultural market can be achieved only by enterprises
that are able to offer good products at a competitive, low price [Michałek 1997]. One
important element of a financial strategy leading to lower prices is the reduction of
operating costs for agricultural equipment, and the reduction of losses associated
with their use. This reduction in operating costs is an urgent challenge for the whole
of the Polish agriculture industry, and also considering the issue of environmental
degradation, primarily to the soil and air.
Modern agriculture is characterized by extensive human interference in plant
genetics, broad introduction of chemicalization, and maximum reduction in manual
labour in cultivation by mechanization and automation. More efficient modern
technologies allow the connection of working units to tractors with greater and greater
widths, power and weight. These types of tractor and working units are characterized
by higher yields, but on the other hand may damage soil structure [Bekker 1977,
Błaszkiewicz 1995, Buliński 2000, Dawidowski 2000].
In these conditions of a global crisis, continuing increases in the prices of the
means of production (e.g. fertilizers, liquid fuels, etc.) and the constant level of
purchase prices of products are decreasing the profitability of agricultural production.
Financial solvency of farms can be maintained through a reduction in production
costs, for example through the elimination of various tillage operations, and
consequently lower energy consumption.
This simplification in tillage has been discussed for a dozen years or so. Test
have been performed to examine the assumptions that the simplified method is
competitive when compared to traditional methods of tillage, and that it will gradually
displace them and result in the introduction of new alternative solutions. The results
of these studies indicate that in addition to a reduction in the time of a tractor's work
compared with traditional methods, a decrease in energy consumption and labourconsumption can also be obtained [Elster 1991].
The simplified method may also lead to a reduction in emissions and better
use of the tractor's engine, in compliance with the recommended agronomic time
90
limits, which is important especially in large scale farms. Simplified tillage is gaining
more and more supporters because of the minimal input of energy. Traditional tillage
is the most energy-intensive; it can be responsible for up to 40% of the total energy
consumption. A large number of tractor runs (5-15) causes the compaction of up to
200% of the cultivation area. The high energy consumption compared to simplified
tillage was for example shown by Heyland et al. [1997] and Roszkowski [1980].
Ploughing by only 0.01 m deeper increased fuel consumption by about 0.5 dm3/ha
[Gonet 1991]. Several experiments confirm that the introduction of simplifications
does not reduce the yield of maize and oats.
Simplified tillage also has several significant drawbacks. These include the
adverse accumulation of phosphorus and potassium compounds, and decreased
respiration capacity and pH of the soil [Pabin et al. 2000]. Additionally, some
researchers observed no differences in yield between the traditional and simplified
methods, for example Biskupski et al. [2000] who used a rigid tine cultivator as the
basic tool for tillage for spring barley and peas on medium and heavy soils. Simplified
methods even resulted in lower yields for winter triticale. The observed advantage of
the simplified method was lower fuel consumption during tillage for spring barley and
broad bean.
Simplified cultivation technology is only possible with the use of specialized
machines which are offered in a varied range and often with very different
construction solutions. The aim of this paper is to examine their operating resistance,
not previously discussed in specialist literature.
AIM OF THE STUDY
The aim of this study was to analyze the operating resistance of selected
working units used in simplified tillage. The study was conducted with a rigid tine
cultivator, a disc harrow and a seed drill for direct sowing. The operating depths and
speeds of the units were measured to determine their effect on resistance.
MATERIAL
The study was carried out with three units used in simplified tillage
technologies. The first unit was a rigid tine cultivator used for no-till loosening of the
soil immediately after harvest or after the initial cultivation, performing post-harvest or
pre-sowing cultivation. The analyzed cultivator is manufactured by the Czech
company Farmet. Technical and operational parameters of the cultivator are
presented in Table 1.
Table 1. Technical specifications of the Farmet cultivator
Parameters
Value
Unit
Number of tines
7
[pcs]
operating width
3
[m]
operating depth
0.08-0.15
[m]
minimum power requirement
100/136
[kW/KM]
surface efficiency
2.4-3.6
[ha*h-1]
operating speed
2.2-3.3
[m*s-1]
weight
1140
[kg]
91
Another unit analyzed was the disc harrow DC Classic 4500 by Czech
company Strom. Technical and operational parameters of the unit are summarized in
Table 2. This unit is equipped with discs on a central frame and operating sections
arranged in the shape of the letter "X". The unit has a rear bar which prevents the
formation of ruts in the field. Operating angles are controlled mechanically with pins.
Lateral, vertical and angular adjustment of dispersion zones are available. Its side
discs are smaller in diameter, 0.56 m at the first section and 0.61 m at the second
section. Solid construction, large weight of the unit, and the weight on individual discs
provides excellent soil penetration even at low moisture levels and in soils for which
plowing is not recommended.
Table 2. Technical specifications of the disc harrow Dowlands Classic DC 4500
manufactured by Strom
Parameters
Value
Unit
operating width
4.5
[m]
operating depth
0-0.15
[m]
Max. weight per disc
121
[kg]
Number of discs
40
[pcs]
Total weight
4460
[kg]
Power requirement
100/136
[kW/KM]
The cutting of the stubble and the depth of soil loosening can easily be
adjusted by setting the angle of the operating sections, in accordance to soil
conditions.
The third examined machine is a versatile tilling-sowing Ultima unit
manufactured by Köckerling. The unit can be used on the arable layer, mulch sowing
or direct sowing. The operating elements of the unit are spring tine hoes with a high
load rating (up to 200 kg), capable of band-sowing and with hydraulically adjustable
operating depth and shock absorption. The unit is also equipped with an STS shaft
with narrow self-cleaning rings that fill with soil. Depth of sowing is regulated by
individual wheels, leading each hoe independently, ensuring a good fit to the
irregularities of surfaces. The unit also has spring levellers before the STS shaft, and
a harrow as the last element of the unit. Technical and operational parameters of the
unit are summarized in Table 3
Table 3. Technical specifications of the ULTIMA unit by Köckerling
Parameters
Value
Unit
operating width
4
[m]
weight of sowing submin. 200
[kg]
units
number of sowing sub23
[pcs]
units
minimum power
125/170
[kW/KM]
requirement
STS shaft
ø 0.53
[m]
weight
5500
[kg]
harrow
ø 0.013
[m]
92
METHODS
Field studies were conducted in August 2009 in Chwałów, Lower Silesia,
Wrocław County, Mietków commune, on a private farm. Measurements were made
on the stubble after the harvest of winter oilseed rape. The tests were performed on
soil classified as a very good rye complex, valuation class IIIa. Granulometric
composition of the soil was determined in accordance with PN-R-04032; soil was
classified as light clay. Soil moisture (12.5%) was determined in accordance with
standard ISO11461:2001.
Physical and mechanical properties of the soil were described as maximum
shear stress and soil compactness. Maximum shear stress was determined by an
Eijkelkamp field shear vane borer. Measurements were made at the following soil
profiles: 0.03-0.07m, 0.08-0.12m, and 0.13-0.17m. Soil compaction was measured
using a conical penetrator with electronic registration of penetration resistance and
depth. The cone used had a base area 0.0001 m2 and 60° side angle. The rate of
penetration was 0.03 m*s-1 in the range of depth 0-0.2m. The simultaneous recording
of penetration resistance and depth was performed by a CompactDAQ station
manufactured by National Instruments, connected to a computer recording analog
signals at 200 Hz sampling rate on each channel.
Measurements of the working resistance of the examined units were
performed using two tractors: basic (CASE MX170) and auxiliary (Zetor 16145).
Tractors were coupled using a specially designed tow bar, in which an electronic
force transducer was mounted (Tecsis, measurement range 0-100 kN). Diagram with
measurements is shown in Figure 1.
Figure 1. Method of measuring the resistance of an operating unit: 1- tested unit, 2auxiliary tractor, 3-tow bar with a force transducer, 4 - basic tractor
Analog signals from the force transducer were recorded in a fashion similar to
the measurements of soil compactness. During the test, three speeds were selected:
1.7, 2.2 and 2.8 m*s-1, and three values of operating depth: 0.05, 0.1 and 0.15 m.
The only exception was the ULTIMA unit, for which research was carried out only for
93
a depth of 0.05m. The rolling resistance of the Zetor 16145 tractor was also
measured and subtracted from the obtained results.
RESULTS AND ANALYSIS
The compaction of soil describes the binding of individual soil particles with
one another, which results in soil resistance when loosened. This parameter also
directly affects the resistance of operating machines. Figure 4 shows the relative
compaction of the experimental plot. The diagram presented in Figure 2 shows that
the compaction of the 0-0.04 m layer is fixed at around 0.06 MPa. at the depth range
of 0.04-0.14m compaction increased proportionally with depth from 0.61 MPa to 2.51
MPa. The measured values of compaction therefore suggest that increasing the
operating depth will manifest in increased operating resistance to the analyzed units.
Table 4 summarizes values of maximum shear stress for established soil
profiles.
soil compactness [MPa]
3
2,5
2
1,5
1
0,5
0,18-0,2
0,16-0,18
0,14-0,16
0,12-0,14
0,1-0,12
0,08-0,1
0,06-0,08
0,04-0,06
0,2-0,04
0-0,02
0
depth of penetration [m]
Figure 2. The values of relative compaction for each soil depth range
Table 4. Summary of maximum shear stress
0.03-0.07 m
0.08-0.12 m
maximum shear stress
19.5
41.5
[kPa]
0.13-0.17 m
43.5
The data presented in Table 4 shows that stress measured at 0.08-0.12m and
0.13-0.17m is twice that at 0.03-0.07m.
Figure 3 shows resistance of the Farmet cultivator at the studied speeds and
depths, indicating that increasing the depth of work for established speeds always
increases the operating resistance of this unit. The largest increase in resistance
values, by 5280 N (61%), was observed for a depth increase from 0.05 to 0.1m at a
speed of 1.7 m/s. The smallest increase in resistance, by 2370 N (15%), occurred for
an increase in depth from 0.1 to 0.15 m at a speed of 2.2 m/s. An increase in the
operating speed of the unit also caused an increase in resistance at the established
94
operating depths. In this respect, the largest increase in resistance, by 2040 N (23%),
was observed for the speed increase from 1.7 to 2.2 m/s and at a depth of 0.05m.
The smallest increase in the value of resistance was 30 N (0.2%), which occurred for
the increase in speed from 2.2 to 2.8 m/s at a depth of 0.1 m.
Figure 4 presents the resistance on the DC 4500 disc harrow at the
established speeds and depths.
working resistance [N]
25000
20000
15000
0,05 m
0,1 m
0,15 m
10000
5000
0
1,7
2,2
2,8
speed of the unit [m/s]
Figure 3. Draft resistance on the Farmet cultivator at the established speeds
and depths
30000
working resistance [N]
25000
20000
0,05 m
0,1 m
15000
0,15 m
10000
5000
0
1,7
2,2
2,8
speed of the unit [m/s]
Figure 4. Resistance on the disc harrow DC 4500 at the established speeds
and depths
The diagram shown in Figure 4 indicates that increasing the operating depth at
the established speeds always increases the resistance of the operating unit. The
largest increase in resistance, by 6060 N (42%), was measured at the increase of
depth from 0.1 to 0.15m at a speed of 1.7 m/s. The smallest increase in resistance,
by 3650 N (26%), occurred for the increase in depth from 0.05 to 0.1 m at a speed of
2.2 m/s. An increase in speed of the unit also caused an increase in resistance at the
established operating depths. The greatest increase in resistance, by 1670 N (14%),
was measured for the speed increase from 1.7 to 2.2 m/s at a depth of 0.05m. The
95
smallest increase in the value of resistance, by 420 N (3%), was observed for the
increase in speed from 2.2 to 2.8 m/s at a depth of 0.05m.
Figure 5 shows resistance on the ULTIMA unit at the established speeds at
0.05m.
working resistance [N]
25000
20000
15000
10000
5000
0
1,7
2,2
2,8
speed of the unit [m/s]
Figure 5. Resistance of the ULTIMA unit at the established speeds at 0.05m
The diagram shown in Figure 5 shows that at a depth of 0.05m, resistance at
the established speeds did not differ significantly, which was confirmed by a
statistical test. Increasing speed from 1.7 to 2.2 m/s increased resistance by 1050 N
(6%).
Due to the fact that the analyzed units had various operating widths, the
measured resistance values were divided by the width of the operating machines and
then compared with one another. Figure 6 shows the resistance values per meter of
the analyzed units, at an operating depth of 0.05 m.
6000
opór roboczy [N]
5000
4000
3000
2000
1000
0
1,7
2,2
2,8
prędkość agregatu [m/s]
kultywator Farmet
brona talerzowa
agregat Ultima
Figure 6. Resistance of the analyzed units per meter of operating width
and 0.05 m depth
96
The diagram in Figure 6 shows that at the established speeds, the highest
resistance per meter of operating width was measured for the Ultima unit and the
smallest for the disc harrow. At 1.7 m/s, differences in the resistance between these
units was 1910N (71% increase). For 2.2 and 2.8 m/s, the difference was about 1700
N (55% increase). However, given that the Ultima unit can perform cultivation of soil
and sowing in one drive, the related energy saving is probably significant. Comparing
the resistance per meter between the Farmet cultivator and the disc harrow at 1.7
m/s, the difference was 8%. For 2.2 and 2.8 m/s, the resistance difference was about
17%. Thus, the use of the Fermet cultivator entails higher energy expenditure than
the disc harrow.
In order to determine the effect of independent variables on the dependent
variables, the obtained results were subjected to multivariate analysis of variance.
Calculations were performed using Statistica 8.0. Results of the analysis are
summarized in Table 5.
Table 5. Results of multivariate analysis of variance
type of unit
operating speed
resistance
0.37
0.0493*
operating depth
0.0002*
* has a significant effect at α=0.05
The analysis shows that the operating resistance is significantly influenced by
the speed and depth of work, at a significance level α = 0.05. The type of unit had no
effect on the operating resistance.
CONCLUSIONS
1.
2.
3.
4.
For the analyzed units, increasing operating speed and depth of work increased
the working resistance, as confirmed by statistical analysis.
An increase in the speed of the Farmet unit from 1.7 to 2.2 m/s, at 0.05m,
caused a significant increase in resistance (by 23%). An increase in speed from
2.2 to 2.8 m/s had no significant effect on resistance. In the case of the disc
harrow, the greatest difference (by 14%) was measured for the speed increase
from 1.7 to 2.2 m/s at a depth of 0.05 m. In contrast, the Ultima showed no
significant differences in relation to differences in resistance for operating
speed.
An increase in operating depth of the Farmet unit from 0.05 to 0.1m, at a speed
of 1.7m/s, caused the greatest increase in its resistance by 60%. For the disc
harrow, the greatest increase in resistance (increase by 41%) was observed for
an increase in depth from 0.1 to 0.15m at a speed of 1.7 m/s.
Resistance per meter of operating width, at a depth of 0.05 m, was the highest
for Ultima and the smallest for the disc harrow.
REFERENCES
1.
Bekker M.G. 1977: Evaluation of tires in road locomotion. Zesz. Probl. Post.
Nauk Roln. 183, 125-131.
2.
Biskupski A., Kaus A., Włodek S., Pabin J. 2000: Wpływ uproszczeń uprawy roli
na plonowanie roślin, czasochłonność uprawy i zużycie paliwa. Inżynieria
Rolnicza 6, 85-90.
97
3.
Błaszkiewicz Z. 1995: ,Badania wpływu rolniczych opon prowadzących na
zagęszczenie i właściwości wodno-powietrzne gleby lekkiej. Probl. Inż. Roln. 4,
6, 5-13.
4.
Buliński J. 2000: Wpływ prędkości ruchu i rodzaju agregatu ciągnikowego na
zagęszczanie gleby kołami. Inżynieria Rolnicza 6, 111-117.
5.
Dawidowski J.B., 2000. Zagęszczanie gleby kołami pojazdów rolniczych.
Inżynieria Rolnicza 6, 23-30.
6.
Elster M. 1991: Conservation of soil and water by using a new tillage system for
row crops. Crovet crop for clean water. Soil and Water Conservation Society,
Jackson, Tennessee.
7.
Gonet Z. 1991: Metoda i niektóre wyniki badań energochłonności systemów
uprawy roli. Fragmenta Agronomica 2, 7-18.
8.
Heyland K., Solansky S., 1997. Berichte über Landwirtschaft. Neue Folge 195,
s. 15-30.
9.
Michałek R. 1997: Inżynieria rolnicza w procesie transformacji polskiego
rolnictwa do Unii Europejskiej. Inżynieria Rolnicza 1, 13-20.
10. Pabin J., Włodek S., Biskupski A., Runowska-Hryńczuk B., Kaus A., 2000.
Ocena właściwości fizycznych gleby i plonowania roślin przy stosowaniu
uproszczeń uprawowych. Inżynieria Rolnicza 6, 213-219.
11. Roszkowski A. 1980: Ocena efektywności energetycznej i ekonomicznej
produkcji kukurydzy i jęczmienia. IBMER, Warszawa.
98
Chapter 9
THE INFLUENCE OF FERTIGATION
ON AGRO-PHYSICAL PROPERTIES
OF ARABLE PLANTS
Ján JOBBÁGY, Ján SIMONÍK, Marek KLIMKIEWICZ
INTRODUCTION
Fertigation was compared to broadcast application of solid NPK fertilizer with
cabbage, carrot and onion. In the broadcast application, P and K were given as a
single application in spring and N was split according to the existing
recommendations. Growth and nutrient uptake were monitored by monthly
samplings. In 1999, natural rainfall was low, and irrigation was applied according to
the tensiometer measurements. Cabbage yields averaged to over 90 t ha-1 in both
years. Carrot yielded according to the samplings close to 90 t ha-1. Onion yielded 4050 t ha-1 [Salo et all, 2002].
In the years of 2004-2006 research on influence of surface and subsurface
drip irrigation on yield and quality of carrot grown on ridges and on the flat ground
was investigated. Nitrogen fertilizers (100 kg.ha-1) were applied in two rates. The first
rate was preplant applied; the second was used by fertigation. Surface and
subsurface drip irrigation significantly increased the yield of carrot. Drip irrigation at
both cultivation methods increased pH and decreased the salt concentration of the
soil [Dyśko, Kaniszewski, 2007].
The optimal nutrition of plants suppose the supply of root zone with nutrients
especially during their intensive uptake because the deficit of nutrients or their
temporal deficiency, may cause decreasing of expected crop. Therefore, the
technologies of fertilization begin ever more asserting with focus to consistent feeding
of nutrients to root zone of arable crops, during vegetation. Its efficiency increase with
it, which show with stabilization and crops increasing. The possibility of water using
as a carry medium of nutrients, so called fertigation is the ideal solution of this
problem (AGREF, 2007).
The fertigation and feeding of nutrients and fertilizers by irrigation water
utilizing the basis of micro-irrigations, especially of drop irrigation. It is precise
distribution of irrigation water; it is in precise amount and time, directly to localized
root zone. Distribute also the dissolved nutrients and fertilizers at the same degree of
uniformity distribution of irrigation water [Hríbik, 2009].
99
Effects are noteworthy:
•
highest degree of utilization reflect in economical and ecological results,
•
better communication between need and prompt accessibility of nutrients,
obtained with possibilities of divided multiple application of nutrients,
•
multiple synergy effect of current irrigation and supplies of nutrients to amount
and quality of production [Hríbik, 2009].
The aim of chapter is to research the effect of fertigation to properties of
selected plant.
MATERIAL AND METHODS
The irrigation with fertigation using was applied on allocated field. The grown
plant was carrot. Part of field was without fertigation application. The basic
fertilization - 500 kg.ha-1 fertilizer NPK was used in the whole field. The irrigation rate
25 mm was apply on the field, with parallel application of fluid saltpetre calcic,
amount 40 to 45 kg.ha-1. The same irrigation rate was applied on the second field
without fertilization. The carrot is cultivating on long planar fields, whereby steep land
is very slow or zero. The grown species of carrot was MAESTRO (Nederland) with
recommended length of vegetation term 120 days.
The cultivation works were not realized during vegetation, done was one
chemical sprinkling with preparation AFALON, ration 2 dm3.ha-1.
The carrot is vegetable with middle requests for soil water. Its moisture need
during vegetation is 520-620 mm. Its successful growing request to fill up the natural
rainfalls with irrigation water in range 120-180 mm, at average year, in our condition.
The belt irrigator CIPA INTERNATIONAL VILLA MINOZZO ITALIA, model
600GX completed with fertilize equipment was used (Fig. 1, 2).
Figure 1. Mobile irrigator VILLA MINOZZO 600GX ITALY
The fertilize equipment APH use the power water, which by distributing valve,
work cylinders and force arm start hydraulic cylinders. The membrane pump sucks
and injects fertilizer solution to the pipes of irrigation equipment.
The amount of transported fertilizer is adjustable in range 1 to 400 dm3.h-1.
The equability of fertilizer distribution depends on uniformity of irrigation water
distribution on side of irrigator.
100
Conic
Ku =
φh − φk
,
l
(1)
where: φ h – diameter of root head, m, φk – diameter of root end,
m, l – root
length, m
Coefficient of convergence (λ1) is expressed with ratio of diameters for head ( φ h ) and
bottom part ( φk ) of root.
Coefficient of slenderness (λ) is expressed with ratio of length ( l ) and with head
diameter ( φ h ) of root.
Figure 2. APH Fertilization device
RESULTS AND DISCUSSION
For determination of advantages and disadvantages of fertigation there was
needed to take samples from row length 1.42 m conform for area 1 m2, from parcel
with fertilization and than thereafter of etalon.
The following operations were performed:
•
weighing of all roots, kg,
•
separation of toppings,
•
weighing of particular roots, kg,
•
determination of length of particular roots, m,
•
determination of head and bottom diameters of roots, m.
The 102 pieces of carrots were gathered on point of supply (with fertilization).
The length of roots was ranged 71-271 mm with value of variation coefficient 23.20%.
The average value of roots length was 169.06 mm. The measured values were
ranged 10 to 241.30 g, for evaluation of single roots weight. The variation coefficient
value was 60.10% for median 78.51 g. Determination of conic was got out from
length and diameters measuring. the values ranged froom 0.021 to 0.135. The
average value was 0.069 with value of variation coefficient 32.19%.
The average length varied in dependence on: kinds, penetrometric resistance
of soil, soil structure, nutrient uptake during vegetation, locality of growing and the
method of growing, and using of irrigation or without its using –using of fertigation or
without its using. Every individual needs for its growth not only nutrients, water,
sunlight but also sufficient place for supple growth.
101
25
120,00%
100,00%
Frequency, pc
20
80,00%
15
60,00%
10
40,00%
5
20,00%
0
,00%
90,80
110,61 130,41 150,21 170,01 189,82 209,62 229,42 249,23 269,03 288,83
Další
Roots length, mm
Početnosť
Freq.
Kumul. %
Cumul. %
Frequency, pc
Figure 3. Root lengths – with fertilizing
35
120,00%
30
100,00%
25
80,00%
20
60,00%
15
40,00%
10
20,00%
5
0
,00%
32,90
55,80
78,71 101,61 124,51 147,41 170,31 193,22 216,12 239,02 261,92
Roots weight, g
Další
Početnosť
Freq.
Kumul. %
Cumul. %
Frequency, pc
Figure 4 Root weights – with fertilizing
30
120,00%
25
100,00%
20
80,00%
15
60,00%
10
40,00%
5
20,00%
0
,00%
0,03
0,04
0,05
0,07
0,08
0,09
0,10
0,11
0,12
Conical, m m
Figure 5. Conical – without fertilizing
102
0,13
0,14
Další
Freq.
Početnosť
Cumul.
Kumul. %
%
Table 1. Basic statistics of measurements parameter (with fertilizing)
Parameter
Length, mm
Weight, g
Conical, Mean
169,06
78,51
0,069
Median
171,00
69,60
0,066
Modus
178,00
26,20
0,051
Standard deviation
39,23
47,18
0,022
Minimum
71,00
10,00
0,021
Maximum
271,00
241,30
0,135
Number of samples
102
102
102
Coefficient of
23,20
60,10
32,19
variability
The most important indicator of production is biological crop. The carrots
samples were separated of toppings and after than were weighted. It results from
weight measuring that for assigned measuring was biological crop of carrots 80.08
t.ha-1.
WITHOUT FERTIGATION
Gathered was from 1 m2 73 pc of carrots by sampling on researched parcel
without ferigation. The odds are that from results of pieces number, that the number
decreased about 29 pc. The measured lengths were ranged from 40 to 256 mm. The
average value was 174.74 mm with value of variation coefficient 26.47%. For used
fertigation was its value lower. The results of single roots weight measuring were
ranged by evaluation of crop at interval from 5.5 to 350.2 g. The average value was
97.69 g with variation coefficient value 65.21%. The value of variation coefficient
returns to be higher.
120,00%
25
100,00%
Frequency, pc
20
80,00%
15
60,00%
10
40,00%
5
20,00%
0,00%
0
65,28
90,56
115,84
141,12
166,40
191,69
216,97
242,25
Root length, mm
267,53
Další
Početnosť
Freq.
Kumul.
Cumul.
%%
Figure 6. Root lengths – without fertilizing
Also crop results from measuring of the roots weight, which was 71.31 t.ha-1.
The average value of conic was 0.075 with value of variation coefficient 49.76%. Its
values ranged at interval 0.007 to 0.227. We will find by comparison of results with
103
results of measuring at fertigation that the average value of conic was lower for
fertigation.
From measured values of the root crop growth results that the average
biological crop varies in dependence on kinds, localities in which is grown and the
soil conditions. Results from summary of growths characteristic that biological crop is
for ours grower conditions good. In the year 2000 was in Kováčovce the average
biological crop of carrots Cartágo 29.6 t.ha-1 (min. 24.6 and max. 41.8 t.ha-1), in the
year 2001 on the fields PD Bušince was the average biological crop 43 t.ha-1 (min.38
and max.55 t.ha-1). The soil on the SELEKT Bučany [2001] was processed after
ploughing with the reploughing only to knolls at autumn and than at spring before
seeding was regenerated shape of knolls with the reploughing. The average
biological crop of carrot was 64.8 t.ha-1 (min. 38 and max. 82 t.ha-1) [Poničan
and kol., 2004].
In this chapter, we were deal with determination of coefficients of the
slenderness (λ) and the convergence (λ1) of the toots, too. The coefficient of
slenderness (λ) presents the relative relation of the average value of the roots length
(ℓ) to the root head diameter ( φh ). It is the relative value which higher value presents
the slimmer shape of root and on the contrary. The coefficient of convergence (λ1)
presents the relative relation of the root head diameter ( φh ) to its bottom root
diameter ( φk ). It is the relative value, which value near to one present the cylindrical
shape of root and for values higher as one are conical shapes of roots. The
measured and calculated values were introduced in Tables 4 and 5.
120,00%
25
100,00%
Frequency, pc
30
20
80,00%
15
60,00%
10
40,00%
5
20,00%
0
0,00%
45,84
86,19
126,53
166,88
207,22
247,56
287,91
328,25
Root weights, g
Figure 7. Root weights – without fertilizing
104
368,60
Další
Freq.
Početnosť
Cumul.
Kumul.%
%
40
120,00%
Frequency, pc
35
100,00%
30
80,00%
25
20
60,00%
15
40,00%
10
20,00%
5
0
,00%
0,03
0,06
0,08
0,11
0,14
0,16
0,19
0,21
0,24
Další
Početnosť
Freq.
Conical, -
Kumul.%%
Cumul.
Figure 8. Conical – without fertilizing
Table 3. Basic statistics of measurements parameter (without fertilizing)
Parameter
Length, mm
Weight, g
Conical, - (4)
Mean
174,74
97,68
0,075
Median
176,00
85,30
0,067
Modus
176,00
103,40
#N/A
Standard deviation
46,25
63,70
0,037
Minimum
40
5,50
0,007
Maximum
256
350,20
0,227
Number of sample
73
73
73
Coefficient of
26,47
65,21
49,76
variability
From obtained results it is possible to state that for fertigation using was the
value of slenderness coefficient higher – that is slimness shape of root. The values
for evaluation of slenderness coefficient vary few, is possible to state that for
fertilization approach to cylindrical shape of root shape. This shape is advantageous
from view point of roots manipulation by cleaning, sorting and packing.
Table 4. Root slimness
Irrigation methods
Irrigation with
fertilizing
Irrigation without
fertilizing
Measuring data, mm
Average of root length
Average of root head
Average of root length
Average of root head
105
169,06
28,03
174,4
30,22
Slenderness coefficient
(λ)
6,03
5,77
Table 5. Root convergence
Irrigation methods
Irrigation with
fertilizing
Irrigation without
fertilizing
Measuring data, mm
Average of root head
Average of root end
Average of root head
Average of root end
28,03
16,59
30,22
17,65
Slenderness coefficient
(λ1)
1,69
1,71
The slimmest shape of the root had a carrot growned on the fields in Bušince,
year 2001 (kind Sugarsnax). The roots of this kind have the slenderness coefficient
(λ) 8.67. The roots of carrot (kind Bolero) growned in Bučany (year 2001) have at
least slender shape, which value of slenderness λ = 5.03 [Poničan a kol., 2004].
From view point of the root shape (cylindrical) approach species Bolero grown
on the fields in Bučany (year 2001), where the value of root shaping λ=1.25 [Poničan
a kol., 2004].
From view point of evaluation of agro-physical properties of roots definite with
slenderness and convergence of the root results that upon this properties influence
especially the species properties and also method of over seeding preparation of soil
[Poničan a kol., 2004].
Salo et. all [2002] researched the efficiency of fertigation applied on field for
growing of cabbage, carrot and onion.
CONCLUSION
The aim of attempt was to determinate the advantages of fertigation. The
grown agricultural plant was carrot – species Maestro. It results that by fertigation is
increasing of crop. The increase of crop was 8.77 t.ha-1 (80.08 – 71.31). It is visible
that for fertigation using the input costing increasing (fertilizes price, fertilizing
equipment, installation etc.). The results of carrot crop out of fertigation using ranged
50 - 60 t.ha-1. It is to consider by every farmer if sufficed the accruals of crops by
growing with fertigation to covering of increased costs. The average crops
accompanied by company Paulis – Gardening services are without using of
fertigation 53 t.ha-1 and with using fertigation 81 t.ha-1. The increase of crop is
28 t.ha-1.
REFERENCES
1. AGREF KOMÁRNO, 2007: Rukopis prednášky.
2. Dyśko J. – Kaniszewski S. 2007: Effect of drip irrigation, N-fertigation and
cultivation methods on the yield and quality of carrot. In: Vegetable Crops
Research Bulletin, Poland, 2007 vol. 67, 25-33, DOI: 10.2478/v10032-0070027-3
3. Hríbik, J. 2009: Synergické účinky zavlažovania špeciálnych plodín. Dostupné
na intrnete: http://www.zahradaweb.cz/projekt/clanek.asp?pid=2&cid=2014
4. Poničan, J. – Jech, J. – Angelovič, M. – Žitňák, M. 2004: Vplyv predsejbovej
prípravy pôdy na jej vlastnosti ovplyvňujúce pestovanie koreňovej zeleniny
v podmienkach precízneho poľnohospodárstva. In: Acta technologica
agriculturae, č.1 , ročník 7, 2004, s.7-10, ISSN 1335-2555
106
5.
6.
Poničan, J. – Jech, J. – Angelovič, M. – Žitňák, M. 2004: Vplyv vlastností pôdy
na agrofyzikálne vlastnosti koreňovej zeleniny. In: Acta technologica
agriculturae, č.4 , ročník 7, 2004, s.94-98, ISSN 1335-2555
Salo, T., Suojala, T. And Kallela, M. 2002: The effect of fertigation on yield and
nutriet uptake of cabbage, carrot and onion. Acta Hort. (ISHS) 571:235-241,
http://www.actahort.org/books/571/571_28.htm
107
Chapter 10
ECOLOGICAL ASPECTS OF ENERGETIC
PLANTS HARVESTING
Aleksander LISOWSKI, Adam STRUŻYK, Jacek KLONOWSKI, Tomasz
NOWAKOWSKI, Jarosław CHLEBOWSKI, Michał SYPUŁA, Stanisław GACH
INTRODUCTION
Poland’s obligations to reduce air pollution can be filled up by using the
appropriate amount of biomass to decrease emission of greenhouse gasses. It will be
possible at considerable increase in the area of long-term plantations of energetic
plants. Intensive agricultural production technologies could contribute to
environmental degradation. Therefore, great attention should be directed to any
environmentally friendly measures, including those related to reasonable cost, new
techniques and harvesting of these plants. Harvest technologies of the plants for
energy purposes have passed through different stages of the evolutionary
development. In the initial period harvesting of energetic plants applied the
techniques used in agriculture or forestry, depending on plant species. Successively
with acquired experience working units or the complete machines were modified, and
a mature stage of development of this market segment offered specialized machines
with modern solutions of working and mobile systems. The current market offers the
machines that do not provide further opportunities to develop and improve
techniques and technology of energetic plants harvesting. Although the potential use
of modern harvest technology is high, in practice, there are still used very different
techniques of mowing and processing of biomass. It concerns especially Polish
conditions, where the cultivation of energetic plants is very dispersed and areas of
single plantations are extremely diversified and range from few ares up to few
hundred hectares.
Depending on the accepted harvest technology there are used different
machines, which can be equipped with units with installed mechanisms, of specific
functional properties, adapted for the plant species and the harvest cycle.
TECHNIQUES OF HARVESTING ENERGETIC PLANTS
Analysing the environmental risks associated with plants harvest technologies
for energy purpose to reduce the negative impact on the environment it is possible to
108
allocate technical risks involved with applying tools and agricultural machines and
technological hazards associated with the use of particular harvesting methods.
Choice of plant cutting elements is a critical decision in the phase of the
machines design for the harvest of energetic plants. It is being taken back especially
to bushes and trees. While cutting the willow shoots, a working unit has the task of
the simultaneous cutting of few shoots of different diameters. All shoots, growing
from the snags, strive for vertical settings, also the ones which are offshoots. Cutting
these offshoots requires the greater energy than vertical shoots, since they have the
greater horizontal cutting surface.
Diversifying cross sections of shoots among rows of plants and along rows
makes harvest difficult. It is easier to cut smaller shoots, but it is possible to scroll
them by moving parts of the cutting mechanism. It would apply the element taking
stems down right above the cutting plain. In order to reduce the negative effect for
growing again of new shoots they should be cut in the first year after seating, at
about 50 mm level above the soil surface [Szczukowski et al., 2006], and in
consecutive 100-150 mm. The result is a better separation of shoots, minimized
damage to snag, reduced propagation of diseases and improved yields [Stuart, 1994;
Wilkinson et al., 2007].
Cutting mechanisms, which are installed on machines to harvest of energetic
plants can be divided into unsupported and supported ones. In units cutting shrubs
and trees there can be used disc saws, chain saws, disc knife, plates and flail knives,
which is one of the unsupported units. Support cutting is used in latter units and
rotational knives are common in adapters of tractor forage harvesters to harvest
maize for silage.
Choice of harvest technology of energetic plants depends on many factors.
Whole stalks can be used on the short, niche market and at schools where they
should be always in the excellent condition without breakdown barks. Whole plants
can also be harvested in two stages. This allows the drive units to use less power
than a single-stage during the harvest [Lechasseur and Savoie, 2005]. As suggested
by Hilton et al. [2005] smaller machines should be more useful in technologies using
natural drying. The harvest of plants in the grounded form is being recommended on
large plantations and where there is a well-developed market, with the possibility of
the biomass use for co-burning in the industrial coal combustion [Gera et al., 2002].
The plant material ground down in the forage harvester working unit can constitute
the raw material for further processing in the production of briquettes or pellets. It is
necessary to grind down further the chips or chaff, obtained from the forage harvester
[Bitra et al., 2008]. The particles are too large to produce durable briquettes or
pellets.
The harvest of energetic plants takes place in late autumn or winter, during the
dormant vegetation. Unfortunately, during this period in most regions there remains
bad weather or high soil moisture in autumn or early spring, making it marshy, or is
too much snow, which not only covers the surface fields, but they among clumps of
plants or on stems and leaves of grass. This is not the only characteristic of our
Central European area. Even worse conditions occur in Scandinavian countries,
Great Britain or Canada [Spinelli, 2001]. Using caterpillar suspension systems or
wide tires is necessary in such conditions. Depending on the applied technology and
the cutting mechanism, the snow can also be taken by biomass and find its way to
the cutting unit, and further with chips to the load-carrier of transport means [Pellerin
et al., 1999; Boyd et al., 2000]. Snow mixed with chips increases their moisture, and
consequently leads to more rapid digestion of biomass and its loss [Boyd et al.,
109
2000]. The date of harvest and its periodicity depends on plant species and the raw
material harvested.
The selection of the harvest technology of energetic plants depends on their
species. Quick-growing shrubs and trees, which can include shrubby willow, poplar
and Acacia Robinia, it is possible to harvest one or two stages in the annual cycles or
more in the multi [Lisowski, 2006; Szczukowski et al., 2006; Faber et al., 2009].
Shrubs such as Rosa multiflora and Acacia Robinia and perennials (Virginia mallow,
Jerusalem artichoke) and multiannual (Miscanthus, switchgrass, spartina prairie,
Cymbopogon Gerard, reed canary grass) it is possible also to harvest one or two
stages, but only in a one-year cycle (Table 1).
The one-stage harvest consists of simultaneous cutting plants and breakingup, mostly using a forage harvester, both tractor driven and self-propelled. The
logical consequence is therefore that it is also harvest of plants in the form of chips
(willow, poplar, shrubs) or chaff (grasses, perennials).
A harvest in two-stage comprises two independent phases, which are
performed by separate machines. In the first phase the plants are cut, and in the
other they are broken-up. Highlighting these phases due to wrong forms of raw
material or product that we receive. Between these phases may occur several
operations, depending on the plant species and forms of a cut material. In the case of
short rotation crops, in two-stages dominates a harvest as a whole plants and after
drying of material, its grinding in stationary conditions. Grasses and perennials, after
cutting, are collected by balers or pick-up trailers. Generally, the harvest of grasses
and perennials plants can be done by machines commonly used to harvest green
fodder, hay or straw.
Willow in Polish climatic conditions is best harvested from mid-November to
mid-March, but weather conditions do not always allow for termination of harvest in
this period. At that time the plant material moisture content is around 45-55%,
depending on the variety, region and weather conditions. In good weather and soil
conditions and well fertilization, some good variety of annual short rotation plants are
achieving yields 20 tDM·ha-1. Typically, the yield of these plants is between 812 tDM·ha-1.
Harvest of willows in subsequent years can be done through various
techniques. The appropriate cut-off already a mature shoot from snag is a very
important element of its viability. Shoots should be cut at the height of 5-10 cm above
the soil to have visible two and ultimately one dormant bud. Cutoff must be done in a
typically sharp tool such as scissors and high-speed saw. Work of scissors secateurs
is very inefficient and this technique can be used in particularly small plantations
harvested in the one-year cycle [Dubas and Tomczyk, 2005]. With very spreading
shoots an additional person is sometimes needed, apart from operator, to incline and
receive of cut stems. On the basis of tests Kwaśniewski et al. [2006, 2008] found that
when cutting with a chain saw, better performance (0,024 ha·h-1) is achieved by
employment of two helpers, than one (0,016 ha·h-1). The work performance achieved
under these conditions was significantly lower than in mechanized harvesting
(0,7 ha·h-1), but the study of authors suggests that the time predisposed to the
harvesting of plants is quite long. However, in respect to the risks associated with this
technique of harvesting and for safety reasons it cannot be recommended. In
addition, when planning a harvest one should take into account the possibility of
adverse weather conditions, such as the autumn and spring sloppy, residual snow
cover, which may affect the performance of manual harvesting. Real disposable time
to harvest willow and other energetic plants can be much shorter than the
110
theoretically predisposed to mid-October to mid-March. With rising labor costs
harvesting by hand will be displaced by mechanical harvesting. Application of forage
harvesters
or
special
machines,
provides
performance
of
3-4 ha/day [Dreszer et al., 2003].
Table 1. Harvest technologies for energetic crops and machinery used
Method of harvest
Short rotation crops (poplar, Grasses (Miscanthus,
willow) and shrubby (Rose switchgrass, spartina prairie,
multiflora, Acacia Robinia) reed canary grass,
Cymbopogon Gerard) and
perennials (Virginia mallow,
Jerusalem artichoke)
Cycle:
Mounted, pulled or selfMounted, pulled or selfone-year
propelled forage harvesters
propelled forage harvesters
with adapter for maize harvest, with a adapter to maize
non-row dependent head
harvest, row or non-row
Onepreferably (Kemper, Krone).
dependent head.
stage
Cycle:
Tractor or self-propelled forage
harvesters with special
2-5-year
adapter.
Transport
Tractor volume trailers or trucks
Cutting; one Secateurs, combustion
Rotary mowers (with the
year cycle
strimmers, tractor mowers with conditioner for grasses),
circular saws, special cutting tractors driven or selfmachines and pilling, bundling propelled.
or pressing.
Cutting; 2-5- As for the one-year cycle of
year cycle
willow, plus combustion saw.
Cutting; 5Poplar: forest machinery,
harvesters group (≤ Ø15 cm)
Two- 30-year
harvesters (> Ø15 cm).
stages cycle
Bailing /
Round balers, large-size
balers, pick-up trailers.
harvest
Loading and Manual, front or grab loaders, Front or grab loaders for bales.
unloading
forwarders or skidders.
Transport
Tractor trailer or cars, trailers Tractor trailers or trucks, selfforest.
loading trailers.
Wood chippers, chippers
Straw cutters, knife grinders.
Shredding
device.
Source: authors’ elaboration
Among the energetic plants willow ligneous can be harvested by using the
most diverse technologies. One-year willow can be harvested using standard
machines, which are used to harvest corn. If the harvest takes place in 1 -, 2 - or 3year cycles willows can be harvested in the form of whole stems or broken-up by
forage harvesters equipped with a special adapter. With longer harvest rotation
cycles, plant shoots are thicker and have more uniform quality wood with less content
and thin bark of young plant parts. Harvest of plants with larger diameter shoots,
however, requires the use of machines more durable and energy units more
powerful. In the case of harvesting the whole plants there are also needed transport
111
means with longer load-carriers and more schedule a larger width of returns on
plantation fields. According to Scandinavian experience harvest of willows in 3 to 5year-old cycles is most justified economically [Mola-Yudego and Pelkonen, 2008],
and poplar and 5-7 years (even 20-30) [Molas, 2008]. In the third year willow shoots
reach the height of 6-8 m and a diameter of 80-100 mm. Willow tree with a diameter
greater than 100 mm should be harvested using techniques used in forestry. In the
three-year harvest cycle, yield can be very high and ranges 60-120 t·ha-1 of wet
material (with an average moisture content around 46%).
One-year willow shoots can also be harvested wholly in bulk or in bundles or
bales, using a special round balers equipped with the cutting disc and the beater
breaking-up units [Lavoie et al., 2008] or only beater cut-breaking-up unit [Schroeder
et al., 2008].
With a harvest of one-stage-related disadvantages: a very high initial cost and
the resulting high cost of the purchase price depreciation, need to start and run
plantations in accordance with the requirements of combine harvesting, the need to
establish large integrated plantations, which may reduce the transport costs of
combine harvester on long-distance, chipping out the possibility of wet weight of the
storage environment, hence the extracted biomass must be burned immediately or
subjected to drying, which considerably increases the price per unit of energy
obtained [Pasyniuk, 2007]. The use of heavy machinery poses a significant threat to
soil compacting, which contributes to the destruction of the structure and subsequent
soil erosion. Needs to be tackled by using tracked systems or driving low-pressure
tires.
Past experience and the specifics of Polish agriculture indicate that
established and future plantations are scattered and range in size from a few to 20
hectares. This implies the implementation of the machines, which will be available to
owners of small plantations, allowing for timely and proper carrying out of all quality
agrotechnical operations, while a relatively low utilization rate does not affect
significantly the costs of sustainable production of solid biofuels.
Conclusions from the studies carried out so far are varied and it is difficult to
clearly identify and to recommend the use of a specific harvest technology. Similarly,
the harvest of maize and other plants recommended for different technologies vary
depending on plantation size, the existing machines in the farm, holding of another
production, possession of transport means and other technical facilities. In general,
the one-stage technology is recommended for large plantations, and two-stage for
the small ones. The experience of countries that grow energetic willow (Australia,
Brazil, Denmark, Finland, Ireland, Canada, Germany, Norway, Sweden, USA, United
Kingdom, Italy) are varied and recommendations with respect to harvesting
technologies of energetic plants depend on many factors, which should be
considered in Polish climatic conditions.
In harvesting of grass and perennial plants there can be proposed the same
techniques as for green and low- and high-stems plants (standing or strawy materials
from swaths) (Table 1). Cutting units of harvesting machines should be set to the
lowest cutting height, in order to avoid losses, but not too low, to avoid picking-up soil
or polluting strawy material. Some machines of Claas and Deutz-Fahr companies are
specially adapted to cutting plants and their bindings. To increase the economic
effects the plant cutting, bailing and logistics still must be developed. Loss of material
during the harvesting of plants may reach even 50-60% (reed canary grass), due to
low specific density, especially when the material is broken-up into chaff. Due to low
specific density the transport of material of certain plants is economically justified only
112
on a small distance, less than 80 km. Rectangular bales are more adventageous than
the round ones and more suitable for long-distance transport.
WILLOW HARVESTING MACHINES
Among the willow harvesting machines there are two basic types of solutions:
cutting biomass with breaking-up into chips or harvesting the whole plants (Fig. 1).
One-stage harvest may be performed using the modified machines which had been
originally used for harvesting other plants, such as maize or sugar cane. Modification
focuses on the adapter for cutting plants.
Harvest of whole shoots in the annual cycles is performed with the use of
machines such as reaper, bundler and a specially designed machine connected to a
tractor.
In the 90s of twentieth century, when cultivation of short rotation crops was
started on a commercial basis, there was no suitable machines for the harvest. After
the first attempts to use the existing machines, which were designed to harvest other
plants, the work began on the modification of machines or prototypes that meet the
user requirements. Scandinavian companies have been pioneers and also Denmark,
the Netherlands, Great Britain, Italy, Germany, Australia and Canada.
In 1993, the Swedish company Salix Maskiner AB, part of Salixphere made
two prototypes of machines. One of them was a pulled machine - Bundler model (Fig.
1e), equipped with a chain saw and binding units. During cutting off the bundle by
circular saw the tractor-machine aggregate has to stop, thereby reducing
performance of harvest. The second machine was a chopper of Bender 1 mounted
on the linkage of the tractor, equipped with a reversible system (Fig. 1d). The
machine underwent further modifications, and currently available version is Bender
6WG. A characteristic feature of the machines offered by this company is cutting of
shoots by a chain saw, carried out on four rollers. Application of chain saw reduces
weight of machines and simplifies the transmission system. Positive feature of the
chain saw is also the fact that during the harvest of willows in winter conditions with
covering snow, the amount of its inclusion in the harvested biomass, compared to
other cut mechanisms, is minimal. Critical remarks on the Bender machine presented
the U.S. experience [Abrahamson et al., 2008].
At first independently Nordnic Biomass company from Denmark, and then in
cooperation with the Border Biofuels Ltd company from Great Britain produced few
generations of machines for harvesting willow in the whole stems form. At first they
were simple machines attached to the tractor. For the second and third generation
self-propelled machines were qualified, respectively HE ALL - Rounder and Mantis.
With the newest machine of fourth generation, offered by Nordic Biomass, is
attached, Stemster TR model from 2006. Both the Mantis self-propelled machine,
and attached Stemster TR are characterized by a high performance from similar
under the technical consideration and functional Empire 2000, designed by
Segerslätt.
113
a
b
c
d
e
f
g
h
i
Figure. 1. Willow harvesting machines: a - Claas Jaguar harvester with HS2 adapter,
b - New Holland harvester with CRL adapter, c - Krone Big X forage harvester with
HTM Woodcut 1500 adapter, d - Salix Maskiner mounted Bender 6WG forage
harvester with a chain saw, e - Bundler's prototype, f - New Holland prototype cutbaling machine, g - KWE-7 prototype pulled mower of Metaltech Ltd (PIMR project),
h, i - prototype semi-mounted harvester's SIPMA SA (an original concept, the
implementation of the project PBZ-MNiSW-1/3/2006)
Source: Elaborated on the basis of [Spinelli, 1999; Lechasseur and Savoie, 2005; Lavoie et al., 2007;
Pasyniuk, 2007; Lisowski, 2009]
In the international scale the most field self-propelled forage harvester became
commonly used with adapters for the harvest of willow and the modernised Austoft
7700 combine harvester for sugar cane harvesting. The German Claas company
offered the first HS1 adapter with two circular saws in Claas Jaguar 695 forage
harvester. The newest HS2 adapter from 2006 works with the Claas Jaguar 900
forage harvester (Fig. 1a). In 1998, the company Coppice Resources Ltd (CRL) from
the UK, in cooperation with Claas, developed a slightly different adapter (Fig. 1b).
The latest versions of the CRL adapters are adapted for forage harvesters of Claas,
Case New Holland, John Deere, Krone companies, and are more widespread than
the HS2.
The HS2 adapter is equipped with two circular saws with teeth cutting from
sintered carbides. The material is broken-up by a standard drum unit of Jaguar
forage harvester cutting into chips of the required length up to 60 mm. The maximum
diameter of willow stalk, which can be cut and broken-up is 70 mm. Depending on
conditions, the type of field, and the maximum diameter of stalks the peak
performance of the machine amounts to 70 t·h-1. With the HS2 it is possible to
harvest two twin rows of plants between which should be kept a distance 0.75 m, and
between next pairs of rows - 1.5 m. These requirements sometimes make the direct
114
application of the Jaguar combine harvester with HS2 adapter impossible in harvest
of willow; therefore the plantation was put on excluding restrictions resulting from the
construction of the machine meant for the more late harvest of plants.
Works on adapting forage harvesters or mowers to the harvest of willow were
carried out also by other companies. In 2006 Krone equipped the mower BIG X V8
with the Woodcut 750 adapter with two circular saws of diameter of 750 mm each,
and in 2007 Woodcut 1500 (Fig.1c) with a circular saw of diameter 1500 mm.
Another machine used in willow harvest is Austoft 7700, which in the basic
version had been designed to harvest the sugar cane. Its modification was to use
circular saws of diameter 600 mm, instead of knife disc with 5 knives and removing
the unit cutting green tops of the sugar cane and redesign of the discharge spout. It
is self-propelled machine weighing about 12.5 t, equipped with a caterpillar
suspension system, which reduces soil compaction. The performance of harvester
Austoft 7700 in willow harvest is similar to Claas Jaguar [Pellerin et al., 1999]. In
other studies, Spinelli and Kofman [1996] found some problems with obstructing cut
material, uneven height of cutting, damage snag and poor quality of wood chips,
which related mainly to the great unevenness of the distribution of large particle size.
Two-stage harvest with the possibility of natural drying can be carried out
using a special cut-baling machine, which cuts down, breaks-up and rolls material
into a cylindrical bale [Lavoie et al., 2007]. Modified baler (Fig. 1f) perhaps to be also
available in the technology of the harvest much cheaper than breaking-up by forage
harvester, therefore can be more available on small farms or in common use on a
smaller scale. The machine can be used to harvest willow, annual or perennial plants
and grasses [Feler et al., 1999]. In 2007 in the cut-baling machine there was
introduced another modification, consisted of dismantling four circular saws and
installing a new cutting unit with a beater rotor [Schroeder et al., 2008]. Experience
shows that in case of the willow harvest with beater unit the speed of shooting of
stems can be smaller [Stuart, 1994].
Other models of machines, specially designed to willow harvest, or modified
often by users themselves, either finished functioning on tests of the prototype or
were used in the limited range, without their dissemination. Similar activities can be
observed on the Polish market of willow harvesting machines. There is therefore a
characteristic, initial phase of development of these machines. Apart from previously
mentioned manual tools, to willow harvest there are used adapted machines,
designed by the user's own structures and simple machines made by small
companies. These machines are cheap and characterized by a very simple
construction, but do not meet even minimal safety requirements and, therefore,
cannot be recommended.
The professionally fabricated prototypes of willow harvesting machines can
now include three models. In 2008 PIMR workers in Poznan developed and
constructed a machine for mowing and automatic tying of stems using the wicker
basket and KWE-7 pulled mower (Fig. 1g), working in the same way as the model of
Roland's Stemster Nordic Biomass. In 2009, on the basis of concept of workers of
the Department of Agricultural and Forest Machinery of WULS in Warsaw, the SIPMA
SA company developed technical documentation, and made a semi-mounted
harvester prototype (Fig. 1h, i).
KWE-7 energetic willow mower, made by the Mechanical Unit of Metaltech Ltd
in Mirosławiec, is designed to cut the whole stems of energetic willow and cut off
stems laying on the load-carrier of the machine. To cut stems of plants at a height of
15 cm above the ground there are used two circular saws. Cut stalks are picked up
115
by two V-shaped conveyors, which also perform the role of distributors. Stalks
assembled in the centre area of the distributor are transported by the next pair of
driving conveyor to the load-carrier, equipped with the special system of chain
conveyors. Chain conveyors, moving perpendicularly to the direction of machine,
gather the cut stalks on the left side of load-carrier. After filling the load-carrier, its
contents can be discharged anywhere on the field or substituted trailer. It is being
reached after the get out of plumb of left load-carrier side by hydraulic cylinders and
running the chain conveyor with drivers of triangular shape. The machine weighing
8430 kg has a robust design, supported by a tandem system, but requires a tractor
with power at least 105 kW. Working units of the machines are controlled by
hydraulic system independently. Functioning mechanisms and working units are
controlled by an electronic control and steering is held from the cab of the tractor
operator. The maximum diameter of harvested willow stems can amount to 70 mm
and the height to 7 m. The length of the load-carrier is 4.5 m, capacity 18 m3 and the
loading of about 7 t.
Tractor forage harvester (Fig. 1h, i), constructed under the project PBZMNiSW-1/3/2006 was designed after a detailed analysis of specially constructed
machines to harvest of energetic plants or modified machines, which primary purpose
was the harvesting of high-stems plants. The conclusions of this analysis and the
assumptions adopted by the authors [Nowakowski et al., 2008] were the basis for
designing of our own construction of the machine, to which was applied for a patent
no P 385 536 [Lisowski et al., 2008]. To transport position the machine is assembled
behind the tractor to reduce the width and to meet requirements of traffic regulations.
In this position the machine is mounted on the linkage of the tractor (Fig. 1i). In the
working position the part of the machine is resting on the outer ground wheel, and the
second wheel is raised to avoid out rigidity of system.
On the part of the frame attached to a tractor there are put: the tank with
hydraulic oil, bevel gear transmission transmitting the drive from the PTO of tractor to
the drum with knives and the hydraulic pump powering, through the electro hydraulic
distributor, the electronic motors and hydraulic cylinders. In the second part of the
frame, supported on wheels, there is a drum cutting unit. Cut stalks are directed by
discharge spout to the load-carrier of transport means. The throw direction of material
is set through turning of the discharge spout by hydraulic motor, and its range - by
proper lowering of the deflector installed on the end of the discharge spout.
The fed cutting material is compacted and held while cutting by feedcompacted toothed rollers. The force of rollers is adjusted by changing preload
springs. To the body of the rolls unit there are pivotally suspended exchangeable cut
adapters. The adapters are unweighted by screw springs of pre-regulated tension.
The height of cutting plants is adjusted continuously by the two plunger cylinders
which are symmetrically arranged under the unit.
The cutting adapter with driving and pick-up-feeding units is designed in two
versions (Fig. 2). In one of the solutions in the feeding unit there were applied
rotational fingers drivers (Fig. 2a), and the second - vertical worm rollers (Fig. 2b). In
both cases, the drive circular saws and rolls with driver fingers or rolling of worms
hydraulic motors were applied. Shafts of circular saws are powered by high-speed
hydraulic motors with a maximum speed around 4000 rpm, what at 0.5 m diameter of
the disc allows for cutting of the peripheral speed of 100 m·s-1. Such value of the
speed results from the need of obtaining a smooth cut surface and reduces the risk of
tearing bast under bark, which may contribute to the putrefactive processes of snags.
Therefore, in the willow harvesting machines the peripheral speed of circular saws
116
are applied 63-118 m·s-1 [Lechasseur and Savoie 2005; Lisowski, 2009]. Rolls with
the driver elements are driven by the low speed hydraulic motor by a gear reducing
the speed to 38 rpm. At the external diameter of drivers 0.5 m, the ends of fingers are
reaching the peripheral speed of 1 m·s-1. The ratio of peripheral speed to take
material to the cutting plants is very large and is 1:100. These parameters can be
obtained from the peripheral speed at nominal speeds of hydraulic motors. Since the
speed tractor-machine unit can be changed depending on field conditions, and
therefore use a hydraulic motor, electro-controlled distributor, will refine the optimum
speed, so that peripheral speed is at least greater than the speed of motion of the
tractor. The surplus of the speed of 10-30%, determined by kinematic factor, should
provide the proper picking up cut stems of plants.
a
b
Figure 2. Cutting adapters with: a) divers fingers, b) spiral rolls
Source: authors' elaboration
Application of tilting element, hydraulically controlled, allows for the plants
cutting by circular saws when plants are bent. The accumulated energy in the
inclined stem will be used after its felling to direct part of the butt-end stem to the next
feed-compacted rolls unit.
Other working units of the machines are similar to those found in forage
harvesters to harvest of low- and high-stalks plants to the silage.
Assuming that the machine will be equipped with row independent unit, with
adjustable spacing dividers, it can cut single rows spaced 0,7-0,8 m [Nowakowski et
al., 2008] or a twin rows spaced 0.75 m. The ground speed of tractor-machine
aggregate will depend on field conditions, but does not exceed the recommended
maximum 8 km·h-1 [Lechasseur and Savoie, 2005]. Cutting plant height of 100 mm
on the surface corresponds to the scope of existing solutions (50-100 mm)
[Szczukowski et al., 2006]. The basic machine is equipped with a breaking-up drum
unit, allowing for cutting the stalks of diameter 70 mm into chips of dimension 2060 mm (according to the standard PN-91/D-95009).
Preliminary examination of the machine led to determination of total power
input from the tractor PTO through the machine at idle, which amounted to 19.14 kW.
The greatest requirement for power have circular saws (9.93 kW) and hydraulic pump
(4.48 kW), which represents 76% of total capacity idle. Power consumption of each of
the other units (active dividers, driver fingers, rolls unit, breaking-up drum) ranged
from 1.03 to 1.35 kW.
The analysis of knowledge about the technology and machines used to
harvest of energetic plants that are used in various solutions and launched in the 90s
twentieth century boom in energetic plants in the Scandinavian countries, extend to
Poland. Moreover, despite of many studies and tests there are still unresolved
questions about the economic and ecological profitability of the cultivation of
117
energetic plants and there is no indication of optimum conditions for biomass
cultivation, harvesting and processing. Frequently cited examples are the results of
natural experiments, and there is lack of special model studies. As an example of a
model approach to analyze a harvest of energetic plants can serve the experience of
Sokhansanj and Fenton [2006]. To calculate the cost of harvesting and transportation
of biomass, they applied a mathematical model IBSAL - Integrated Biomass Supply
Analysis and Logistics developed by Sokhansanj [2006]. The IBSAL model contains
various algorithms, including harvesting, processing, milling, storage and transport.
Model input data include: information about local weather, the net average yield of
biomass, timing of harvest, losses of dry matter, plant moisture content during
harvest, operating parameters, equipment, machinery costs in $·h-1.
For this reason this research aimed at development of a mathematical model
and an algorithm for calculation and simulation of complex research; the analysis that
will allow for selection of the best harvest technologies for energetic plants was
formulated.
SIMULATION MODEL OF THE PLANTS HARVESTED FOR ENERGY PURPOSES
The selection of machines found in plant production technologies should be
guided by specific requirements, which include, among the others, issues concerning
compacting of soil, resistance of machines and slippage occurred. Accomplishment
of the requirements specified above will allow for performing the determined work
with less fuel consumption, and lower emission of exhaust gas into the atmosphere.
Choice of technology outside the one- and two-stage harvest is also often
conditioned with machines possessed in the household, in order to reduce
investment. It also follows from other requirements for the machines to harvest of
plants with one, two and three years. Proposed simulation model takes into account
both current and new technologies of the harvest, namely:
•
harvest of broken-up plants with tractor forage harvester - Z1 technology,
•
harvest of broken-up plants with propelled forage harvester - Z1a technology,
•
harvest of plants with cut-baling machine - Z2 technology,
•
harvest of whole stems with cutting machine - Z3 technology,
•
harvest of whole stems with cut-piling machine - Z3a technology,
•
harvest in bundles (packages) with cut-packaging machine - Z4 technology.
In each of these harvest technologies, recognized as basic, there are options
that include various solutions of the receipt, the transport and breaking-up of plants.
In order to facilitate the comparison of methods there was assumed that in every
technology material is broken-up as final product (chaff or chips).
A mathematical algorithm that enables to perform simulation of the
expenditures incurred in technologies of the willow harvest was developed. For
calculations, the following preliminary assumptions were adopted: the plantation area
of 10 ha, the yield of annual plants of 25 t·ha-1, plants moisture content during harvest
53%, the transport distance of 1.0 kilometers, the time of working shift 8 h, the time of
the harvest dispositional to 30 days. Initial assumptions can be changed depending
on the plantation.
It was also assumed that in considered technologies the harvesting machines
(basic) would be single. Depending on the parameters of these plantations the sets
of machines will be selected, which ensure the continuous process of harvest.
Subsequently accepted sets of machines used in harvesting technology. Working
parameters, the prices of machines and tractors were described with relations.
118
Calculations carried out allowed for getting the efficiency of units. For determining the
criterion indicators the methodological studies were used [Muzalewski, 2006;
Muzalewski and Olszewski, 2000], as well as information material [Gromadzki, 2009].
The specific costs of biomass producing for energy purposes can be
characterized by many factors, among others the fuel and labour consumption. Fuel
consumption related to 1 t of dry matter, abstracted in these technologies is different.
The lowest fuel consumption was found in the Z1a technology, and the largest in the
Z3, which was respectively 7.24 and 12.68 kg·t-1DM. Fuel consumption in other
technologies is 26.3-40.1% lower than fuel consumption in Z3 technology. The fuel
consumption of a tractor in the aggregate to the harvest in relation to consumption in
the whole technology was 13.6-39%. Breaking-up of plants after harvest increases
fuel consumption by 24.4-41.5%. Also drying of broken-up plants in Z1 and Z1a
technologies increases the fuel consumption by 29.2-34.1%.
A similar trend is shown for labour consumption in individual technologies as
for fuel consumption. The smallest expenditure of labour working 0.61 rbh·t-1DM was
found in Z1a technology because of the smaller number of operations, the greatest
one in Z3 technology (6.2 rbh·t-1DM) because of operations carried out manually,
which caused decreasing of efficiency. The labour consumption in other technologies
is about 72.4-81.8% less. The labour consumption of plants harvesting in relation to
the total labour consumption of technologies represented 9.5-21.7% and was
dependent on the efficiency of harvesting machines. In all technologies the transport
amounts to 8.4-30.1% and handling - 11.5-70.6% of expenditure of labour. The high
labour consumption in Z3 technology results from the low efficiency of manually
handling work. Breaking-up of stalks after drying increases the labour consumption
by 11.5-22.1% and drying of chaff in Z1 and Z1a technologies increases it by 42.170.5%.
In the structure of chips cost productions the Z1 technology is characterized by
the smallest specific costs (119.77 zl·t-1DM or 6.45 zl·GJ-1), and the largest - Z1a
(366.12 zl·t-1DM, 19.73 zl·GJ-1). The costs of other technologies amount to 11,9615.22 zl·GJ-1 and are about 22.9-39.4% lower than in Z1a technology. The high costs
of harvest (81.2 and 71.2%) in Z1a and Z3a technologies are caused by the use of
expensive machines, in this case on too small areas. In Z3 technology the harvest
amounts to 8.9% of total costs, and handling operations up to 50.5% due to the low
productivity of manual work. Breaking-up of stalks after drying increases the cost up
to 8.6-22.4% and drying of chaff in Z1 and Z1a technologies up to 6.6-13.2%.
SUMMARY
Technologies of direct harvest of willow in chips, with the application of existing
machines on appropriately large plantations, are cheaper than a two-stage
techniques in the form of whole stems [Styles et al., 2008]. The harvest technologies
of whole stems are more flexible, due to fewer constraints of harvest conditions and
can benefit [Kofman, 2005], which result from lower costs, the possibility of natural
drying, letting for acquiring better fuel [Harders, 2002]. In Great Britain, the plants
stored in such conditions, within seven months reduced the relative moisture content
of 53-55% to 19% [Hilton, 2000]. The disadvantages of this willow harvesting method
can also include a larger number of rides of transport means, the need to use special
trailers for the transport of long stalks (up to 8 m) and the loader to load them,
providing larger width for returns and surface storage of large volumes of raw
119
material. Essentially in this harvest technology is to perform more operations and is
more laborious than the single-stage harvest technology.
Based on previous research results [Lechasseur and Savoie, 2005] it can be
concluded that the use of machines to harvest the whole stems of willow, can
significantly reduce the specific fuel consumption per ton of fresh plants. It was found
that during the operation of Empire 2000, Frobbesta and Hyd-Mech machines
harvesting the willow in whole stems form the specific fuel consumption was 0.501.00 l·t-1, and machines of Claas Jaguar 695, Bender and Austoft 7700, which
simultaneously cut down and breaking-up stalks - 0.95-1.20 l·t-1. Smaller values of
this index for machines harvesting whole plants result from the fact that during this
treatment plants are only cut and transported as a whole to the load-carrier area of
the machine or the trailer. After the storage period the breaking-up of stalks on chips
are done. Therefore, direct comparison between the requirement for power or other
energy indicators of harvesting machines that perform different functions is a
simplification. The lack is also of explicit criteria for valuation - economical measures
with which it would be possible to measure benefits ecological, associated with the
use of specific techniques of harvest.
Scientific work financed by the Ministry of Science and Higher Education in 20072010 as a research project ordered PBZ-MNiSW-1/3/2006.
REFERENCES
1. Abrahamson L.P., Volk T.A., Priepke E., Posselius J., Aneshansley D.J., Smart
L.B. 2008: Development of a willow biomass crop harvesting system in New
York. Minneapolis, pp. 21.
2. Bitra V S., Womac A R., Chevanan N., Sokhansanj S. 2008: Comminution
properties of biomass in hammer mill and its particle size characterization.
Published by the American Society of Agricultural and Biological Engineers, St.
Joseph, Michigan, www.asabe.org.
3. Boyd J., Christersson L., Dinkelback L., 2000: Energy from willows. SAC,
http://www1.sac.ac.uk/envsci/external/willowpower/willow_s.pdf.
4. Dreszer K., Michałek R., Roszkowski A. 2003: Renewable energy - a possible
acquisition and use in agriculture. Ed. PTIR, Krakow-Lublin-Warsaw, pp. 256 (in
Polish).
5. Dubas J.W., Tomczyk A. 2005: Establishment, care and protection of willows for
energy plantations. Ed. WULS, Warsaw, 29-30 (in Polish).
6. Faber A., Kuś J. (ed.), Matyka M. 2009: Growing crops for energy. W&B
Wiesław Krzewiński, pp. 30 (in Polish).
7. Felker P., McLauchlan R., Conkey A., Brown S. 1999: Case study: development
of a swath harvester for small diameter (<10 cm) woody vegetation. Biomass
and Bioenergy, Vol. 17, 1-17.
8. Gera D., Mathur M P., Freeman M C., Robinson A. 2002: Effect of large aspect
ratio of biomass particles on carbon burnout in a utility boiler. Energy & Fuels,
16, 1523-1532.
9. Gromadzki J. 2009: Product price list of tractors and agricultural machinery. Ed.
PIMR, Poznan, pp. 160 (in Polish).
10. Harders G. 2002: Harvest, loading and transporting of willow chips from
agricultural land – a system analysis. JTI Institut för Jordbruks – och miljö
teknik. Raport JTI, pp. 294.
120
11. Kofman P.D. 2005: Establishment and harvesting of willow SRF in Denmark.
http://www.coford.ie/iopen24/pub/pub/Seminar/2005/Kofman.pdf.
12. Kwaśniewski D., Mudryk K., Wróbel M., 2006: The set of energy willow using a
chainsaw. Inżynieria Rolnicza, 13, 271-277 (in Polish).
13. Kwaśniewski D., Mudryk K., Wróbel M. 2008: Evaluation of energy willow
harvesting scythe using diesel. Inżynieria Rolnicza, 10, 159-165 (in Polish).
14. Lavoie F., D’Amours L., Savoie P. 2007: Development and field performance of a
willow cutter-shredder-baler. VDI BERICHTE, VOL 2001, 311-316.
15. Lavoie F., Savoie P., D’amours L., Joannis H. 2008: Development and field
performance of a willow cutter-shredder-baler. Applied Engineering in
Agriculture, 24(2), 165-172.
16. Lechasseur G., Savoie P. 2005: Cutting, bundling and chipping short rotation
willow. The Canadian society for engineering in agricultural, food, and biological
systems. Paper no. 05-080, pp. 12.
17. Lisowski A. 2006: Shearing and crushing energy willow. Technika Rolnicza
Ogrodnicza Leśna, 4, 8-11 (in Polish).
18. Lisowski A. 2009: Harvester to harvest energy crops. Agrotechnika, 11, 38-39
(in Polish).
19. Mola-Yudego B., Pelkonen P. 2008: The effects of policy incentives in the
adoption of willow short rotation coppice for bioenergy in Sweden. Energy
Police, 36(8), 3062-3068.
20. Molas R. 2008: Poplar and Sida - energy crops. Odnawialnezrodlaenergii.pl, pp.
2 (in Polish).
21. Muzalewski A. 2006: Operating costs of machinery. Eksploatacyjno-economic
indicators of machines and tractors used on individual farms. Ed. IBMER. no.
21, pp 32 (in Polish).
22. Muzalewski A., Olszewski T. 2000: Economic and organizational aspects of
collaborative use of agricultural machinery. Ed. IBMER, Warsaw, pp. 94.
23. Nowakowski T., Lisowski A., Strużyk A., Dołżycki A., Smaga M. 2008: The
concept of machine harvesting willow shrub. Technika Rolnicza Ogrodnicza
Leśna, 3, 2–4 (in Polish).
24. Pabis S. 1985: Methodology and methods of empirical science. PWN, Warsaw,
pp. 267 (in Polish).
25. Pasyniuk P. 2007: Problems of mechanization of cultivation and harvesting of
shrub willow Salix viminalis, Problemy Inżynierii Rolniczej, 1 (55), 145-154 (in
Polish).
26. Pellerin R.A., Aneshansley D.J., Phelps A., Abrahamson L.P. 1999: Evaluation of
tree harvesters and delivery systems for short rotation willow crop.
ASAE/CSAE-SCGR, paper no. 995055.
27. Savoie P. 2005: Cutting, bundling and chipping short rotation willow.
CSAE/SCGR Meeting, Winnipeg, Manitoba, 25-29 June, no. 05-080, pp. 11.
28. Schroeder B., Savoie P., Lavoie F., D’Amours L., Kort J. 2008: Woody Biomass
Harvest from Willow Rings. Short Rotation International Crops Conference
Bloomington, Minnesota, August 18-22, 2008.
29. Spinelli R. 2001: Report on the CRL Mk II SRC harvester.
www.treepower.org/harvesting/crl.pdf, pp. 7.
30. Spinelli R. 1999: A review of short-rotation forestry harvesting in Europe.
http://www.woodycrops.org/paducah/spinelli.html.
31. Spinelli R., Kofman P. 1996: A Review of Short-Rotation Forestry Harvesting in
Europe, First Conference of the Short Rotation Woody Crops Operations
121
Working Group, Paducah, KY, September 23-25.
32. Stuart W.B. 1994: Mechanization of short rotation, intensive culture wood crops.
http://www.woodycrops.orga/mechconf/stuart.html, pp. 131.
33. Styles D., Thorne F., Jones M.B. 2008: Energy crops in Ireland: An economic
comparison of willow and Miscanthus production with conventional farming
systems. Biomass and Bioenergy, 32(5), 407-421.
34. Szczukowski S., Tworkowski J. 2003: Production of perennial energy crops in
the region of Warmia and Mazury - current state and prospects. Postępy Nauk
Rolniczych, 3 (in Polish).
35. Szczukowski S., Tworkowski J., Stolarski M. 2006: Willow. Ed. Plantpress,
Krakow, pp. 46 (in Polish).
36. Wilkinson J.M., Evans E.J., Bilsborrow P.E., Wright C., Hewison W.O., Pilbeam
D.J. 2007: Yield of willow cultivars at different planting densities in a commercial
short rotation coppice in the north of England. Biomass and Bioenergy, 31(7),
469-474.
122
Chapter 11
EFFECT OF SEASON ON CARBON
DIOXIDE AND AMMONIA PRODUCTION IN
BROILER HOUSING
Štefan MIHINA, Monika KNÍŽATOVÁ, Jan BROUČEK
INTRODUCTION
Modern poultry housing is designed and constructed to reduce heat loss and
improve energy efficiency, however when coupled with reduced ventilation, can result
in elevated levels of carbon dioxide, ammonia and other air contaminants, which
may adversely affect the health and productivity of flocks [Olanrewaju et al., 2008].
On the other hand Lendelová and Botto [2009] documented that pre-warming of
incoming air could decrease this negative influence of reduced ventilation.
Air quality is one of the most difficult aspects of broiler management to grasp. This is
probably because it is not possible to “see”. It is difficult to visually determine the
volume and direction of fresh air being brought into a broiler house.
During cold weather the primary function of the ventilation system is to
eliminate ammonia and moisture from the broiler house. Carr et al. [1990] relate
higher concentrations of NH3 in winter months with reduced ventilation rates to
conserve as much heat as possible. Many producers underestimate the detrimental
effect of ammonia. Ammonia levels of just 25 ppm have been found to depress
growth and increase feed conversion in broilers [Miles et al., 2004]. The housing
environment, including factors like carbon dioxide levels and oxygen levels, is known
to influence the incidence of ascites (broiler pulmonary hypertension syndrome) in
broiler chickens. The problem arises from very high metabolic rate of rapidly growing
broiler strains. Subsequently, in less well ventilated poultry house as well as at higher
altitudes, oxygen becomes a limiting factor as far as their health, welfare and
performance are concerned [Movassagh Ghazani et al., 2008]. The quality of the inhouse environment is highly dependent upon litter quality, but the quality of litter is
seldom given sufficient emphasis. In most instances, excess ammonia in broiler
house is due to wet litter and insufficient ventilation. Litter moisture may affect the
conversion rate of uric acid to ammonium nitrogen [Liu et al., 2006]. The
transformation of organic material in litter is also accompanied by the release of
carbon in the form of CO2, methane (CH4) and other organic gases [Nicks et al.,
2003]. However, the main sources of CO2 within a poultry house include fuel
123
combustion, bird respiration and ambient air content (typically 300-500 ppm)
[Olanrewaju et al., 2008]. Both, the gas furnaces and the broilers generate CO2 and
consume O2. Oxygen consumed is equal to the volumetric CO2 produced by the birds
and is assumed double the carbon dioxide produced by open-flame natural gas
furnaces [McGovern et al, 2001]. This means, the combustion of one molecule of fuel
(CH4) generates one molecule of CO2 and consumes two molecules of O2.
Management of the broiler house environment is a study in conflict between
the need for temperature and humidity control for optimal bird performance, feed
conversion, and energy conservation. Ventilation is used to remove noxious gases
including ammonia, and carbon dioxide as well as dust and moisture. With the advent
of modern nipple-type drinkers, most broiler operations are experiencing less
difficulty with interior moisture but greather concentrations of dust, NH3 and CO2.
Recent research suggests that current recommendations for minimum ventilation
should be made based on minimum accceptable CO2 and NH3 concentrations, rather
than moisture [Xin et al., 1996].
The objective of this study was to explain the potencial sources of CO2 and to
examine the effect of some litter variables on NH3 volatilisation as well as to calculate
emission rates of both problematic gases occuring at crirical levels in broiler facilities.
MATERIAL AND METHODS
A common broiler rearing facility was monitored during 6 consecutive fattening
periods specified below (Table 1). The begining and ending dates are provided for
each flock to document the time of year when each flock was kept.
Table 1. Monitoring schedule
Fattening Season
period
1
2
3
4
5
6
summer / autumn
autumn
autumn / winter
spring / summer
summer
summer / autumn II
Date
Duration
(days)
30.07. - 07.09
23.09. - 01.11
18.11. - 27.12
02.05. - 10.06
16.06. - 25.07
10.08. - 18.09
40
40
40
40
40
40
Average
number of
chickens
23,929
24,310
24,502
24,287
23,908
24,016
Housing description
Approximately 25,000 chicks yielding a stocking density of 18 – 22 birds per
sq. m. were placed in a concrete-flored commercial broiler facility with a housing area
of 1 128 m2 (94 x 12 m) and/or interior volume of 4,455 cubic meters (0.178m3/head).
The housing area was heated to nominal temperature of 31-33° C by two gas
furnaces (Table 2). Ambient temperature was reduced as the birds progressed in age
by approx. 2° C each week to ensure their comfort. The house was mechanically
ventilated with combined tunnel and cross two-sided ventilation. Six ceiling axial fans,
with maximum capacity of 12,000 m3/h of each, and four frontal fans with maximum
capacity of 35,000 m3/h assured air exchange in chicken house. Fresh air inlets were
placed on both side walls of the hall.
A breeding area was equipped with 4 nipple drinker lines and 3 tube-style pan
feeder lines that were filled automatically. Birds (Ross 308) were fed ad libitum to a
124
final market weight of around 2 kilograms. Each one of flock was kept for 40 days of
fattening period.
New straw was used for each subsequent fattening period, littered to a depth
of 5 to 10 cm (approx. 1.6 kg/m2). The age of litter corresponded with age of birds.
No additional litter material or amendments were added to the litter at any time
throughout the study.
Table 2. Gas furnaces characteristic
Model
Power output (kW)
Natural gas consumption (m3.h-1)
Ventilation rate (m3.h-1)
Heating distance (m)
Weight (kg)
GP 70
70
7.5
5,000
50
36
GP 120
120
12.5
7,000
50
64
Sampling and calculation
Carbon dioxide and ammonia concentrations were measured by infrared
analyzer (1312 Photoacoustic Multi-gas Monitor). A self-contained pump draws air
samples into the analyzer via sample tubes from five measuring points. Air samples
were taken from air stream at two ceiling fans, two frontal fans and from outdoor
environment. The sampling points were placed approx. 180 cm over the floor.
Monitoring task operated continuously with one hour sampling interval.
At the same points, air temperature was measured by thermocouple probe.
Two thermocouple probes were placed also into litter (30 mm deep), in the front part
and opposite end of house.
Emission factors were determined using the average concentration near the
house exhaust fans reduced by outdoor concentrations of monitored gases and
multiplied by the volume of air that has passed through the building. The ventilation
rate of exhausted air was based on current ventilation capacity (%) and known rate of
air flow at 100 % efficiency (212,000 m3/h).
Statistical evaluation
A statistical analysis system (SAS ver. 9.1) and descriptive statistic were used
for a research data processing. Spearman correlation was calculated for the
evaluation of relationships between gases production and observed variables of
indoor environment. The differences were declared significant when their probability
levels were below 0.05. Tukey HSD comparison method was chosen from multiple
comparison procedures performed by analytical software Statistix 9.0 to test
significant differences between means.
RESULTS AND DISCUSSION
It is generally recommended that CO2 concentrations should be kept below
3,000 ppm in broiler housing environment [Council Directive 2007/43/EC, 2007]. The
critical values of CO2 were reached in all observed periods and CO2 level was
sometimes even tree times higher than it is allowed. Particularly during the first and
the fourth quarter of periods chickens were exposed to very high levels of carbon
125
dioxide (Table 3). However, it is important to point out, that the concentrations were
not measured at the level of chickens' heads.
CO2 accumulation can occur when additional CO2 is produced by direct
heating systems (where the exhaust gases remain inside the broiler house) and
when the ventilation rate is operated at extremely low level [EC, 2000].There are two
main sources of CO2 in general. The first one is supposed to be heaters, however, as
the birds approached marked age, the CO2 source is primarily from the birds
respiration [McGovern et al., 2001]. A speculation can be accomplished if is
assumed, that a typically natural gas consists of 97.6 % methane, 1.5 % ethane,
propane, butane, 0.1 % CO2, and 0.8 % nitrogen. The gas furnaces used for heating
in chicken house (Table 2) had natural gas consumption of 7.5 and 12.5 m3/h and the
power output of 70 and 120 kW. If we burn completely 20 m3 of natural gas and
assuming that it is pure methane (CH4), it follows this equation:
CH4 + 2O2 → CO2 + 2H2O
Since 1 mole of a gas occupies 22.4 liters at STP (standard temperature and
pressure, i.e., temperature of 0° C and pressure of 101.325 kPa), 20,000 liters of CH4
contains 20,000/22.4 = 892.86 moles of CH4.
Since for each mole of methane we get one mole of carbon dioxide (see
equation above) and one mole of CO2 has a mass of approx. 44 g, so 892.86 mole of
CO2 has a mass of approx. 892.86 x 44 or 39.3 kg of CO2. That is, the complete
combustion of 20 m3/h at STP of natural gas results in the production of about 39.3
kg/h of CO2. However, both heating units were not running continuously. Thus we
can not say this amount of CO2 was produced every hour.
Table 3. CO2 and NH3 concentration range in individual quarters of fattening periods
Days of fattening
period
1. t o 10.
x
CO2 (ppm)
summer / autumn I
autumn
autumn / winter
spring / summer
summer
summer / autumn II
NH3 (ppm)
summer / autumn I
autumn
autumn / winter
spring / summer
summer
summer / autumn II
4,123
1 1. to 2 0.
mi n ma x
(n = 960)
939
x
2 1. t o 3 0.
mi n ma x
(n = 960)
9,427 2,323
x
mi n ma x
(n = 960)
993 5,272 2,459
5,767 1,395 10,067 4,394 1,172 7,141 3,694
3 1. t o 4 0.
x
mi n ma x
(n = 960)
1. t o 4 0.
x
mi n ma x
(n = 3840)
986 5,050 2,651 1,092 5,161 2,889
959 7,643 3,273 1,069 5,693 4,282
939
9,427
959 10,067
6,242 1,129
6,108
4,035
8,889 4,940 2,261 7,419 5,759 2,686 9,157 5,425 1,698 8,471 5,592 1,129 9,157
9,828 3,449 1,224 6,296 2,922 1,324 6,732 2,889 1,564 6,112 3,842 977 9,828
887 10,489 2,209 994 4,363 2,783 1,283 5,418 2,856 1,461 5,554 2,971 887 10,489
977
3,960 1,043 10,236 2,630 1,289 4,913 3,091 1,601 6,313 2,943 1,515 5,375 3,156 1,043 10,236
(n = 960)
(n = 960)
(n = 960)
(n = 960)
(n = 3840)
2.8
0.7
5.1
2,2
0.8
6.1
4,2
0.8
13.3
4.7
1.1
13.3
3.5
0.7
13.3
4.0
1.6
6.6
3,8
1.2
7.6
4,8
1.4
12.3
5.8
2.0
11.6
4.6
1.2
12.3
2.3
1.1
4.2
2,6
1.1
10.5
10,2
4.0
26.0
13.7
4.6
29.1
7.2
1.1
29.1
1.8
0.4
4.1
1,0
0.1
2.6
3,4
0.6
14.0
4.2
1.4
14.0
2.6
0.1
14.0
1.2
0.0
11.1
0,5
0.0
2.1
3,7
0.4
14.1
5.0
1.9
14.0
2.6
0.0
14.1
1.2
0.1
3.4
1.7
0.2
6.2
5.8
1.6
19.6
4.7
2.2
11.1
3.3
0.1
19.6
Carbon dioxide levels in the broiler house atmosphere tend to increase over
time with bird growth and respiration [Miles et al., 2006]. Corresponding to intensive
heating at the begining, CO2 concentration decreased from placement to midfattening and then increased slightly towards the end of periods (Figure 1). It means,
one source of CO2 (gas burning) was replaced with other one (birds respiration). This
126
effect is not very evident, since more intensive ventilation (Table 4) entered this
process and CO2 was diluted in fresh air from outdoor environment.
CO2 concentration (ppm)
8000
7000
6000
5000
4000
3000
2000
1000
0
0
5
10
15
20
25
30
35
40
Time (day)
summer / autumn I
autumn
autumn / winter
spring / summer
summer
summer / autumn II
Figure 1. Course of carbon dioxide concentrations during individual fattening periods
Table 4. Parameters of indoor environment during individual fattening periods
Days of fattening period
1. to 10.
x
3
Ventilation rate (m /h)
summer / autumn I
autumn
autumn / winter
spring / summer
summer
summer / autumn II
Air temperature (°C)
summer / autumn I
autumn
autumn / winter
spring / summer
summer
summer / autumn II
Litter temperature (°C)
summer / autumn I
autumn
autumn / winter
spring / summer
summer
summer / autumn II
sd
(n = 240)
11. to 20.
x
sd
(n = 240)
21. to 30.
31. to 40.
x
x
sd
(n = 240)
39,079 27,921 73,361 49,373 114,772 49,642
28,284 7,503 34,159 4,173 46,349 10,328
25,581 5,029 32,330 2,525 35,060 2,227
22,711 7,580 39,061 14,191 126,458 58,806
27,322 19,426 68,105 44,233 98,819 55,572
30,873 13,656 66,877 41,550 124,594 64,793
(n = 960)
29.2
29.7
32.5
29.3
29.2
28.6
1.6
1.8
3.9
2.2
3.8
2.3
(n = 480)
26.7
30.7
25.2
26.9
26.2
28.7
0.49
1.52
1.26
1.76
2.09
1.04
(n = 960)
26.5
26.1
24.7
23.5
25.2
24.9
2.0
2.1
1.6
2.3
2.1
1.7
(n = 480)
27.9
27.1
28.1
27.1
27.7
28.1
0.89
1.96
1.04
1.76
1.42
1.78
(n = 960)
25.1
21.3
21.7
24.3
23.7
24.4
1.8
1.6
1.4
2.7
1.8
2.7
(n = 480)
31.1
25.3
30.5
30.3
30.0
30.8
1.56
1.58
0.85
2.12
1.76
2.12
sd
(n = 240)
130,725
90,524
39,406
102,387
128,764
145,785
58,893
18,899
2,287
41,257
47,680
55,652
(n = 960)
24.7
23.7
19.7
22.8
23.8
23.3
1.8
2.5
1.4
2.3
1.8
2.5
(n = 480)
34.4
29.0
33.3
32.9
33.5
32.2
0.97
0.89
0.72
1.05
1.35
1.01
sd = standard deviation
The amount of CO2 produced by respiration of chickens can be explained in a
similar manner as a fuel combustion mentioned before. The amount of CO2 produced
127
by birds is proportional to the heat production of the animal (1 liter CO2 for every 24.6
kJ of total heat produced). This corresponds approximately to 1.5 l/h/kg liveweight
[EC, 2000]. The production of carbon dioxide in the experiment carried out by Para et
al. [2003] decreased with increasing weight of broilers from the mean weight of
0.25 kg/head always up to the final weight of 1.5-2.0 kg/head; the initial value
reaching 1.85 l/h/kg and the final one 1.23 l/h/kg, resp. Since 1.5 liter of CO2
corresponds to 0.06696 moles (1 mole CO2 = 22.4 liters at STP) and one mole of
CO2 has a mass of 44 g, then a chicken exhales of approx. 0.06696 x 44 or 2.946 g
CO2 /h/kg. At market age of 2 kg and the capacity of 25,000 chickens kept in this
broiler house, 50,000 x 2.946 g or 147.3 kg of CO2 is emitted per hour as a
consequence of birds respiration. If we take into consideration the first day of just
hatched chicks with liveweight of 40 g, 2.9 kg of CO2/h is produced by their
respiraton. However, it is also important to point out, that the breathing frequency
changes by age markedly.
Values of NH3 were recorded in concentration range between 0.0 and 29.1
ppm. Ammonia concentration had rising tendency in all periods (Figure 2).
Correlation with bird (litter) age was very highly significant (P<0.001) (Table 5).
Vučemilo et al. [2007] associate the increasing of air concentration of ammonia with
the increase in animal age and air humidity. He reported almost septuple higher level
of NH3 concentration between the first and the fifth week of age (litter – mixture of
wooden sawdust and shavings). In our measurement we found approximately triple
increase in ammonia concentration between the first and last quarter of fattening
period (Table 3). Since the temperature of litter was quite stable during the whole
year (Table 4) and the tested differences between fattening periods were also not
significant (Table 6), lower NH3 concentrations during summer periods must have
been caused by other factors (intensive ventilation, drier litter, and crust on its
surface). Moreover, we did not notice marked differences in temperatures of air when
comparing the summer and winter period as well (Table 6).
NH3 concentration (ppm)
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
0
5
summer/autumn I
10
autumn
15
20
25
Time (day)
autumn/winter
spring/summer
30
summer
Figure 2. Course of changes in ammonia concentrations
128
35
40
summer/autumn II
Table 5. Correlations between studied variables and production of gases
Variables
CO2 conc.
summer/autumn I
autumn
autumn/winter
spring/summer
summer
summer/autumn II
NH3 conc.
summer/autumn I
autumn
autumn/winter
spring/summer
summer
summer/autumn II
Air
temperature
Litter
temperature
Chicken
(litter) age
Ventilation
rate
0.23227 0.68612 +++
0.23450 0.64880 +++
0.19248 -0.02198 -
0.26454 0.41463 ++
-0.24237 -0.51859 +++
0.02576 0.18901 -
-0.35066 +
-0.84578 +++
-0.19743 +
-0.76916 +++
0.00563 0.06354 -
-0.51503 +++
-0.84712 +++
-0.37745 +
-0.70775 +++
-0.14816 -0.04149 -
-0.65235 +++
-0.44897 ++
-0.80878 +++
-0.09191 -0.28434 -0.71000 +++
0.69925 +++
0.00281 0.82104 +++
0.75779 +++
0.74765 +++
0.71508 +++
0.64015 +++
0.72289 +++
0.92368 +++
0.62293 +++
0.70269 +++
0.78864 +++
0.49261 ++
0.69803 +++
0.75421 +++
0.60674 +++
0.68581 +++
0.74108 +++
+ P<0.05; ++ P<0.01; +++ P<0.001; - non-significant
The comparison of concentration and emission rate according to the fattening
period (season) is given in Table 6. The highest NH3 concentrations were recorded in
fattening period 3 (autumn/winter) and 2 (autumn), and the lowest in period 4
(spring/summer). The differences were significant (P<0.001 and P<0.05
respectively). CO2 concentrations reached also the highest values in period 3
(autumn/winter). The lowest ones were recorded in period 1 (summer/autumn I)
(P<0.001). In case of NH3 and CO2 emissions, the seasonal differences were not
significant. Liang et al. [2003] reported higher emission rates in summer than in
winter because of highest ventilation capacity, even thought the lower
concentrations. A number of authors [Coufal et al., 2006; Redwine et al., 2002]
published similar seasonal changes in emissions.
The ventilation rate showed a rise in all periods. The ventilation system
operated at the highest capacity during period 6 (summer/autumn II) and at the most
reduced capacity during period 3 (autumn/winter) to maintain the suitable indoor
temperature. The difference was significant (P<0.001) (Table 6). Ammonia
concentration and amount of air exhausted through the ventilation system were in
positive correlation (P<0.001) unlike carbon dioxide (Table 5). It means, that in spite
of rising ventilation rate towards the end of the fattening period, the ammonia
concentration did not decrease but it had even slightly increasing tendency.
129
Table 6. A seasonal comparison of observed variables
Fattening period
NH3 conc. (ppm)
x
sd
Significance
(n = 3840)
1
3.5
1.8
2
4.6
1.2
3
7.2
5.3
4
2.6
1.8
5
2.6
2.3
6
3.3
2.5
NH3 emission (kg/h)
Fattening period
CO2 conc. (ppm)
F = 17.17
P = 0.0000
3:2,1,6,5,4 +++
2:5,4 +
sd
2889
1430
2
4282
1456
3
5592
979
4
3842
1761
5
2971
1535
6
3156
1311
0.16
0.14
1
225.6
119.1
2
0.13
0.10
2
258.4
110.0
3
0.16
0.14
3
280.5
66.2
4
0.13
0.16
4
288.0
199.7
5
0.15
0.17
5
233.4
151.5
6
Litter temperature (°C)
0.20
0.18
6
Air temperature (°C)
291.8
186.3
(n = 1920)
30.0
3.2
1
26.4
2.5
2
28.0
2.6
2
25.2
3.7
3
29.3
3.2
3
24.7
5.4
4
29.3
3.0
4
25.0
3.5
5
29.3
3.2
5
25.5
3.4
6
29.9
2.3
6
25.3
3.1
Ventilation rate (m /h)
3
F = 2.48
P = 0.0325
(n = 3840)
1
F = 2.69
P = 0.0219
F = 37.89
P = 0.0000
3:2,4,6,5,1 +++
2:6,5,1 +++
4:1 +++
5:4 ++
4:6 +
(n = 3840)
1
F = 1.13
P = 0.3454
Significance
(n = 3840)
1
CO2 emission (kg/h)
(n = 3840)
x
F = 1.32
P = 0.2582
(n = 960)
1
89,484
59,724
2
49,828
27,001
3
33,094
5,970
4
72,654
56,594
5
80,752
57,714
6
92,032
66,165
F = 14.76
P = 0.0000
6:2,3 +++
1:2,3 +++
5,4:3 +++
5:2 +
+ P<0.05; ++ P<0.01; +++ P<0.001; sd = standard deviation
From a seasonal point of view, CO2 emissions reached the highest values
(280 t) in flock “summer/autumn II“ (Table 7). This was attributed to increasing
ventilation rate of the building (P<0.001). Relatively high emission rates were also
determined in flocks “autumn/winter“ (269 t) and “spring/summer“ (276 t). This was
significantly affected not only by ventilation rate, but also by increasing CO2
concentrations (P<0.01, P<0.001). There was also a moderate correlation between
ventilation rate and CO2 concentration - the higher the ventilation rate, the lower the
concentration of CO2. Moreover, there was also no statistically significant correlation
between CO2 concentration and CO2 emission in three flocks (Table 5).
The NH3 emission factor ranged from 5.17 g/head to 7.81 g/head per period.
Gates et al. [2008] reported almost 3 times higher ammonia emission (17.4 g/head
for one period in fattening to life weight 2.1 kg housed on sawdust litter). Lacey et al.
[2002] emphasize that different values of emission factors published by American
and European authors are caused by different climatic conditions and differences in
average live weight of animals. He reports emission factor 19.8 g NH3/head for 49
days fattening cycle (average life weight of chickens 1.03 kg).
130
Table 7. Summaries of CO2 and NH3 emission data (kg/h)
Days of fattening period
CO2
summer / autumn I
autumn
autumn / winter
spring / summer
summer
summer / autumn II
NH3
summer / autumn I
autumn
autumn / winter
spring / summer
summer
summer / autumn II
1. to 10.
11. to 20. 21. to 30. 31. to 40.
1. to 40.
Total emission
(kg)
EF 1)
(kg/head)
EF 2)
(kg/head)
73.11
155.75
240.21
247.02
198.11
120.29
131.96
137.91
212.57
238.70
169.18
130.72
177.71
270.09
232.31
311.21
445.82
284.22
398.38
338.60
348.55
325.07
338.93
398.31
459.27
216,563
248,075
269,280
276,497
224,050
280,156
9.05
10.20
10.99
11.39
9.37
11.67
9.46
14.02
58.81
72.51
154.8
6.47 . 10-3
14.35
16.64
29.56
65.03
125.58
5.17 . 10-3
7.72
12.14
54.31
82.33
156.5
6.39 . 10-3
5.89
5.04
62.64
53.58
127.15
5.24 . 10-3
4.38
4.91
51.18
82.94
143.41
6.00 . 10-3
4.121
15.15
82.49
85.78
187.54
7.81 . 10-3
0.043
1)
Partial emission factor calculated from average number of chickens in individual periods, represents
emission of CO2 or NH3 per head and 40 days of fattening period
2)
Annual emission factor, i.e. emission converted into seven fattening periods in one productive year
CONCLUSION
From calculations performed in this study it can be concluded, that a natural
gas burning process is responsible for a substantial part of CO2 emissions during first
days of periods, and later, the respiration of animals takes dominance. The major
part of CO2 seemed to have it´s origin in a bird respiration with assumed production
of approx. 147 kg of CO2/h. The heaters could be theoretically a possible source of
approx. 39 kg each hour, if they would work continuously. This evaluation of CO2
emission sources could be complete if also CO2 releasing from litter decomposition
would be taken ito the consideration. The temperature of litter and the temperture of
air related with chicken age and ventilation rate were main parameters observed
during this experiment, to assess their influence on carbon dioxide and ammonia
production. There was a great influence of chicken age on CO2 emissions (P<0,001),
but ventilation rate could markedly affect this correlation.The amount of ammonia
emissions was increasing by chicken (litter) age (P<0,001) probably as
a consequence of both, increasing NH3 concentration and ventilation rate (P<0,001).
However, there was not found statistically significant difference in emissions of both
gases from individual fattening periods.
Resulting annual emission factors for CO2 (73.11 kg/head) and for NH3 (0.043
kg/head) were calculated for 7 periods, i.e. one production year. On the basis of
obtained results it can be concluded that from poultry facility with capacity of 25.000
broiler chickens placed onto litter, about 1,828.103 kg CO2 and 1,000 kg NH3 is
emitted yearly.
REFERENCES
1. Carr, L. E. - Wheaton, F. W. - Douglass, L. W. 1990: Empirical models to
determine ammonia concentrations from broiler chicken litter. Transactions of
the ASAE, 33, p. 1337-1342.
2. Coufal, C. D. - Chavey, C. - Niemeyer, P. R. - Carey, J. B. 2006: Nitrogen
131
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
emissions from broiler measured by mass balance over eighteen consecutive
flocks. In: Poultry Science, 85, p. 384-391. ISSN 0032-5791.
Council Directive 2007/43/EC of 28 June 2007 laying down minimum rules for
the protection of chickens kept for meat production. The Council of the
European Union, Luxemburg, 2007, p. 19-28.
EC (2000): The welfare of chickens kept for meat production (Broilers). Report
of the Scientific Committee on Animal Health and Animal Welfare, Brussels,
Belgium, European Commission, 21 March 2000, 149 p.
Gates, R. S. - Casey, K. D. - Wheeler, E. F. - Xin, H. - Pescatore, A. J. U.S.
broiler housing ammonia emissions inventory. In: Atmospheric Environment, 42,
2008, p. 3342-3350.
Lacey, R. E. - Redwine, J. S. - Parnell, C. B. 2002: Emission factor for broiler
production operations. ASAE Annual International Meeting, GIGR 15th World
Congress, Chicago, Illinois, Paper number 024212.
Lendelová, J. - Botto, Ľ. 2009: The possibilities of effective assurance of thermal
comfort in poultry housings. In: Folia Veterinaria, 53, p. 136-137.
Liang, Y. - Xin, H. - Tanaka, A. - Lee, S. H. - Li, H. - Wheeler, E. F. - Gates R. S.
- Zajaczkowsky, J. S. - Topper, P. - Casey, K. D. 2003: Ammonia emissions
from U.S. poultry houses: part II – layer houses. In: Proc. 3rd International
Conference on Air Pollution from Agricultural Operations, American Society of
Agricultural Engineers, St Joseph, MI: ASAE, p. 147-158.
Liu, Z. - Wang, L. - Beasley, D. - Oviedo, E. - Baughman, G. - Williams, M. 2006:
Effect of litter moisture content on ammonia emissions from broiler operations.
Workshop on Agricultural Air Quality, Washington, DC, USA, p. 859-860.
McGovern, R. H. - Feddes, J. J. R. - Zuidhof, M. J. - Hanson, J. A. - Robinson,
F. E. 2001: Growth performance, heart characteristics and the incidence of
ascites in broilers in response to carbon dioxide and oxygen concentrations. In:
Canadian Biosystems Engineering, 43, 4.1-4.6.
Miles, D. M. - Branton, S. L. - Lott, B. D., 2004: Atmospheric Ammonia Is
Detrimental to the Performance of Modern Commercial Broilers. In: Poultry
Science 83, 1650-1654. ISSN 0032-5791.
Miles, D. M. - Owens, P. R. - Rowe, D. E., 2006: Spatial variability of litter
gaseous flux within a commercial broiler house: ammonia, nitrous oxide, carbon
dioxide, and methane. In: Poultry Science, 85, p. 167-172. ISSN 0032-5791.
Movassagh-Ghazani, M. H. - Karami A. R. - Ahmadzadeh, A. R. Dolgharisharaf, J. 2008: Relation between incidence of ascites syndrome and
the ventilation factor in broiler chickens. In: Asian Journal of Animal Veterinary
Advances, 3, p. 334-336, ISSN 1683-9919.
Nicks, B. - Laitat, M. - Vandenheede, M. - Désiron, A. - Verhaeghe, C. - Canart,
B. 2003: Emissions of ammonia, nitroux oxide, methane, carbon dioxide and
water vapor in the raising of weaned pigs on straw-based and sawdust-based
deep litters. In: Animal Research, 52, p. 299-308.
Olanrewaju, H. A. - Dozier, W. A. - Purswell, J. L. - Branton, S. L. - Miles, D. M. Lott, B. D. - Pescatore, A. J. - Thaxton, J. P. 2008: Growth performance and
physiological variables for broiler chickens subjected to short-term elevated
carbon dioxide concentrations. In: International Journal of Poultry Science, 7, p.
738-742. ISSN 1682-8356.
Para, Ľ. - Bis-Wencel, H. - Ondrašovič, M. - Vargová, M. - Ondrašovičová, O.
2003: Production of water vapours and carbon dioxide by broilers under
experimental and operation conditions. In: Folia veterinaria, 47, 66-70, ISSN
132
0015-5748.
17. Redwine, J. S., Lacey, R. E. - Mukhtar, S. - Carey, J. B. 2002: Concentration
and emissions of ammonia and particulate matter in tunnel ventilated broiler
houses under summer conditions in Texas. Transactions of the ASAE, 45, p.
1101-1109.
18. Vučemilo, M. - Matković, K. - Vinković, B. - Jaksić, S. - Granić, K. - Mas, N.
2007: The effect of animal age on air pollutant concentration in a broiler house.
In: Czech Journal of Animal Science, 52, p. 170-174.
19. Xin, H. - Berry, I. L. - Tabler, G. T.1996: Minimum ventilation requirement and
associated energy cost for aerial ammonia control in broiler houses.
Transactions of the ASAE 39,2, p. 645-64.
133

Podobne dokumenty