Available online at www.sciencedirect.com SCIENCE

European Journal of Agronomy

DIRECT*

Europ. J. Agronomy 20 (2003) 53-62

ELSEVIER

www.elsevier.com/locate/eja

Balances of the major nutrients N, P and K at the farm and field level and some possibilities to improve comparisons between actual and estimated crop yields Angelija Bučienėa,*, Alfonsas Švedasb,1, Šarūnas Antanaitisb,2 a

Klaipėda University, Minijos 153, LT-5804, Klaipėda, Lithuania Instituto al. 1, LT-5051, Akademija, Dotnuva, Kėdainiaidistrict, Akademija, Lithuania

bLithuanianInstituteofAgrieulture,

Abstract

Since 1994 Lithuania is engaged in BEAROP—the international study ofagriculturally derived runoffofnutrients from the countries round the Baltic Sea, proposed by Swedish scientists. One of the demonstration watersheds chosen for this project, river Graisupis watershed, is located in the agricultural plains region of Middle Lithuania. Crop yield and nutrient balance data obtained for V. Liutkevičius Demonstration farm at the scale offarm and field are discussed. It was evident after calculations of the flow and balance of N, P and K to observe a discrepancy between actual and estimated values ofcrop yield. This is a common problem in Lithuania where an unbalanced fertilisation system still exists on many farms, when farmers choose the fertiliser rates, too little attention is paid to the soil analysis and soil nutrient balance. In order to make the fertilisation system more rational, a balanced fertilisation computer model was created at Lithuanian Institute of Agriculture (LIA) and used for that purpose at the Demonstration farm. A drained system of 7.4 ha divided into separate fields, growing a mixture of ley at the Demonstration farm was chosen for the nutrient balance calculation in 1997 and 1998. Calculations ofmajor nutrient flows at the field level have shown that the N balance in this field was positive and P almost in balance, however, the potassium balance was negative. It may be rational to utilise such a local organic fertiliser as urine in this field. The regression equation was developed that relates N surplus with nitrate loss and could be used more widely provided data were available on drainage discharge, total N content in soil, humus content, fertilisation rate. Soil testing at the Demonstration farm should be extended to the areas examined more than 10 years ago. More attention has to be given to the integrated plant protection and soil cultivation measures in order to obtain the planned yields. 2003 Elsevier B.V. All rights reserved. Keywords: Nitrogen; Phosphorus; Potassium; Nutrient balance; Nitrate leaching; Real and estimated yield

1. Introduction *Corresponding author.Tel.:+370-46-398-661;fax:+3706-398-652. E-mail addresses: [email protected] (A. Bučienė). [email protected] (Š. Antanaitis). 1 Deceased. 2 Tel.: +370-347-37-193; fax: +370-57-37-096.

The agricultural sector in

Lithuania has a

substantial impact on the environment. Leakage and atmospher ic emissions of nitrogen (N) compounds contribute to the general eutrophication of

1161-0301/03/$ - see front matter 2003 Elsevier B.V. All rights reserved. doi: 10.1016/S1161 -0301 (03 )00073-X

54

A. Bučienė et ai. I Europ. J. Agronomy 20 (2003) 53-62

the Baltic Sea. The water quality in rivers and lakes is affected by the discharge of N and phosphorus (P). Further more groundwater is heavily polluted by nitrate, which can cause health problems if used as drinking water. The environmental monitoring on different levels, field, farm, watershed, region or national—is one of the tools to get objective data on these environmental issues. Since 1994 Lithuania has been engaged in BEAROP an international project studying agricultural runoff from the countries round the Baltic Sea, proposed by Swedish scientists and financed by the Swedish government. The project focuses on transfer of good agricultural practice, knowledge and technology, in the broad sense of farm management, monitoring and legislation in order to reduce pollution associated with runoff and ammonia emission from agriculture in Lithuania, Latvia, Estonia, Poland and the Kaliningrad region of Russia. Two demonstration watersheds were selected for the monitoring purposes in Lithuania: Vardas watershed in the Eastern hilly region and Graisupis watershed in the agricultural plains region of Middle Lithuania (Bučienė et ai., 1998). In this paper we present and discuss crop yield, N loss in drainage runoff and nutrient balance data obtained at V. Liutkevičius Demonstration farm in the Graisupis watershed over the period 1997-1998. Discrepancies between real and estimated yield data as well as appropriate measures to reduce them are discussed. The Demonstration farms are important for education purposes, because the results obtained at the farms are more easily evident and understandable for farmers. Thus calculations of nutrient balances are important not only for monitoring purposes, but also for practitioners. With information about nutrient storage in the soil they can better plan their crop nutrition needs and save money by utilising manure and urine and only where is necessary add mineral fertilisers as supplementary nutrients. 2. Materials and methods

The Demonstration watershed is located in Dotnuva community, Kėdainiai district (Fig. 1).

Physical-geographical conditions in the watershed are typical for the Middle Plain of Lithuania. The Demonstration watershed includes the Graisupis upstream drainage area. The Quaternary sediment layer consists of sandy loam and light loam material. The landscape is flat, 62-65 m above sea level. The main soil type is an EndocalcariEndohypogleic Cambisol (FAO-UNESCO, 1997). The watershed is typical of an agriculturally productive area with fertile soils in Lithuania (Bearop in Lithuania, 1997). The largest part of the watershed area has been tile drained. The total demonstration watershed is 13.65 km" (1365 ha) consisting of: arable land+ley of 934 ha (68.4%); forests, 413 ha (30.3%); wetlands and swamps, 18 ha (1.3%). The watershed area consists of three large farms, Aušra agricultural company, Lipliūnai agricultural company and the Experimental farm of Lithuanian Institute of Water Management, 14 private farms and about 90 homelands. The total number of inhabitants is about 200. Restoration of land ownership has not finished yet, therefore, the number of farmers and homelands is changing. The main monitoring station was constructed about 1.3 km upstream of the outlet of the River Graisupis. In addition to monitoring ground water quality, precipitation was also measured. In order to evaluate the influence of land use to nutrient leaching, a monitoring site was constructed in 1996 on the main drainage collector that drains 7.4 ha of the cropped fields and farmstead. The farm belonging to V. Liutkevičius is a typical mixed farm with dairy cattle and cereals production. Besides his own 16.8 ha, in 1998 the farmer bought 10 ha and leased 77 ha of land, which consists of about 10 ha, grassland. Thus the total area of farmed arable land in 1998 was 86 ha and about 18 ha of grassland including pasture. The main crops grown were: winter wheat, barley, barley + peas, oats + peas and winter rye. Other crops like potatoes, maize for green fodder, sugar beet, fodder beet and vegetables are grown on a smaller area of a few ha. There are 18 milking cows (yielding 5000 kg of milk per cow per year) and 20 calves seven heifers on the farm. Nutrient leaching data at farm level was calculated using NPK-FLO computer program (Fager-

A. Bučienė et al I Europ. J. Agronomy 20 (2003) 53-62

55

Graisupis watershed

GRAI5UPI5 WATERSHED

Demonstration farm

Fig. 1. Graisupis watershed and Demonstration farm located in Kėdainiai district of Lithuania.

berg et al., 1993) using data from the appropriate crop rotation of the Valinava drainage plots at the experimental site in Lithuanian Institute of Agriculture (LIA), which is located 8 km North East from the Demonstration farm. Regression-correlation analysis was used in order to determine the dependence of mineral N (N-NO3 and N-NH4, kg ha-1) leached from the soil profile on different crops and factors like drainage water discharge, soil humus content, total N content and amounts

of mineral and organic fertilisers applied. Nutrient balances at the farm and soil status were calculated using the computer program NPK-FLO (Fagerberg et al., 1993). There was no manure storage before the BEAROP project started and this was calculated for an 8 months indoor period. The total capacity of manure storage is 328.5 m3 and animals are kept on deep straw litter in barns. Solid manure 50-60 t ha"1 was spread in late autumn 1997 for sugar

56

A. Bučienė et al. I Europ. J. Agronomy 20 (2003) 53-62

beet and vegetables (25% of all manure stored). Fifty percent of total amount was spread at a rate of 60 t ha -1 for potatoes, vegetables and fodder beet in the following spring and the rest applied before winter cereals. Manure was broadcast and ploughed in on the same day. The urine was spread on ley and pasture in early spring and early autumn. The drainage system of 7.4 ha was covered by ley, one area 2.4 ha being established in 1994 and in the rest was established in 1996. Thus in 1998 one part of field was 4 years old and the second one was a 2-year-old ley. The grass mixture consisted of timothy (Phleum pratense L.), red clover (Trifоlium pratense) and rye-grass (Loliwn perenne L.) in the 2.4 ha area and the rye-grass, timothy, red and white clover (Trifolium repens L.) in the remaining area. The first mixture consisted of less than 10% of clover and the second one was of about 25% clover. The field in 1997-1998 was fertilised with ammonium nitrate three times at a rate of 50 kg N ha-1 from early spring till the last cut (ammonium phosphate and potassium chloride were applied early in the spring at 14 kg ha -1 P and 47 kg ha -1 K). A drainage runoff data logger, constructed at the outlet of the drainage system recorded leaching from the fields. For drainage discharge measurements V-notch (a = 20°) weirs were used and water sampled were collected once a month for analysis. Analyses were conducted at the laboratory of the Water Management Institute: nitrate was determined photometrically, total N was determined by the same method following oxidation with potassium peroxodisulphate. Phosphate was analysed photometrically using ammonium molybdate and total P by the same method after oxidation using potassium peroxodisulphate. Potassium was determined by flame photometry (Bearop in Lithuania, 1997). FYM samples were chopped and analysed for dry matter content. For the NH 4 + -N analysis fresh samples were transferred to a Kjeltec apparatus. NaOH was added, and the resulting NH 3 was distilled and collected in an acid solution. NH 4 + was determined by acid-base titration. Total N and total P were analysed on dried samples, which were wet digested according to

the Kjeldahl method. Nitrogen and P in the acid digests were determined colorimetrically, using a Technicon Auto Analyser. Potassium was analysed on dried samples after wet digestion, using flame photometry. Mineral fertilisers were not analysed for these elements. Standard data on element contents was used in the balance calculations. Topsoil samples were analysed for pH after equilibration with a 1 M KC1 solution (soil:KCl solution 1:2.5 w/w). Total N was determined as described above for FYM. Mineral nitrogen (NH 4 + -N + NO 3 - N) was determined colorimetrically in 1 M KC1 extracts (soil:KCl solution 1:1 w/w). Plant available P and К were determined by the AL method according to Egnér et al. (1960). The organic matter content was determined by redox titration after digesting the samples with an acid K2Cr2O7 solution (Tyurin, 1931). Total С was analysed after dry combustion (Hereaus apparatus). Total N, P and К content in dried plant biomass samples were determined as described above for FYM. Denitrification was assumed to be 30 kg ha -1 per year for the sandy-loamy soil and biological N fixation was estimated using the computer program NPK-FLO. Ammonia losses from plant material was assumed to be 5 kg ha -1 , because the grass after cutting was left in the field for only 1 day and night, ammonia losses from animal excrement was assumed to be 50%) of the N content in the faces and urine (Fagerberg et al., 1993).

3. Results and discussion Data on N, P and К flows and nutrient balances for the farm and soil in 1998 are presented in Tables 1 and 2. Farm nutrient balance calculations show, that in 1998 there was a positive balance for N, P and K. The soil N balance was negative as compared with 1997, and there was still a need to increase mineral N fertiliser rates, because the yields and plant uptake had increased. However, the soil К balance was just positive and the small negative value for P has not yet been reflected in

A. Bučienė et ai. I Europ. J. Agronomy 20 (2003) 53 62

Table 1 Nutrient balance of the demonstration farm, Graisupis in 1998 N

P

K

1. Mineral fertilisers 4- seeds purchased 2. Fodder and animals purchased 3. Atmospheric deposition 4. Biological fixation in green matter and roots Total 1

85 9 17.6 10 122

10 2 0.5 13

70 1 2.3 73

Nutrient output {kg ha-1) 1. Plant production sold off the farm 2. Animal production sold off the farm Total 2 Input-output (Total 1 -Total 2) Utilisation rate (%) (Total 2/Total 1 x 100)

43 4 47 75 38

9 15 1 2 10 17 3 56 77 23

Nutrient input (kg

ha-l)

any yield reductions and soil P status remains rather high (Table 3). Observations on nitrate concentrations and leaching with drainage runoff from drainage plots in Valinava experimental site have shown, that the mineral N leached annually (y, kg ha -1 ) is a function of soil humus content (x1, %), total N content (x2, %), drainage discharge (x3, mm), amount of readily available nutrients in mineral fertilisers (x4, kg ha-1) and in organic fertilisers (x5, kg ha-1), using the following regressioncorrelation equation:

Table 2 Soil N, P and K balance in kg ha demonstration farm, Graisupis, 1998

-1

Soil balance

per year of the N

P

K

85 28 13 126

10 0.5 3 14

70 2.3 23 95

123 4 50 30 207 — 81

24 0.25 24.2 — 10

86 4 90 5

Nutrient input

1. Mineral fertilisers+seeds 2. Deposition+biological fixation 3. Nutrients in manure and urine Total 1 Nutrient output

1. Nutrients in harvest 2. Ammonia evaporation from plant residues 3. Nitrate leaching 4. Denitrification Total 2 Soil nett-flow (total 1 - total 2 = x)

57

y = (5.887x1 + 75.11x2 + 533.6/x3 4- 0.01189x4 + 0.04002x5-3.77)0.01x3

(1)

Thus, according to the results obtained, the intensity of fertiliser application was not among the most important factors, that influenced N leaching (see regression coefficients). The most important factor was drainage discharge quantity, which is a function of precipitation, the second was the soils total N reserve and thirdly soil enrichment with organic matter/humus. Fig. 2 shows a good agreement between measured and predicted total annual losses of inorganic-N using equation 1 (Švedas and Antanaitis, 2000). Nutrient balance calculations at the field level are presented in Fig. 3. Dry matter yield for the first silage cut was 7700 and 7300 kg ha-1 in the second cut. in total for its 7.4 ha 56 t dry matter was removed and a further estimated 2000 kg DM was grazed by the dairy cows after the final silage cut. The nutrient concentrations in drainage water from a separate ley field in 1996-1997 varied: between trace levels and 0.42 mg 1-1 for NH4-N, 3.4 and 10.4 mg 1-1 for NO3-N, 0.01 and 0.06 mg I - 1 for PO 4 -P (Bučienė et al., 1998). Data available from the Valinava experimental site of LIA show, that concentrations of K in the drainage runoff water from the drainage plots with the same crops and soils varied between 0.6 and 2.0 mg 1 -1 (Bučienė, 1999). Nutrient balance calculations have shown, that the N balance in this field was positive and N losses not associated with harvested material (leaching and gaseous) were lower than N removed by cutting and grazing. In such a sward is not necessary to apply more N fertiliser. There is the possibility to save some N mineral fertilisers by increasing the share of white clover in the field. The P balance is almost even and because the soil is rather rich in available P, it was felt that there was no real need to increase P application rate. In contrast opposed to P and N, the K balance is quite negative here and as the available K concentration in soil was also rather low (64 mg kg-1 K), more attention should be given to applications of organic fertiliser such as urine to this field.

A. Bučienė et al. I Europ. J. Agronomy 20 (2003) 53-62

58

Table 3 Agrochemical properties of top-soil (mean values) for fields on the demonstration farm, Graisupis, 1995-1997. Field number

1 2 3

PHKCI

7.3 7.4 7.5

Available (mg kg -1 ) P

K

135 132 113

105 79 69

Humus (g kg-1)

Total N (g kg -1 )

Mineral N (mg kg -1)

39.1 33.0 25.0

2.27 2.24 1.52

9.0 8.8 8.0

Nutrient flow mechanisms in different agroecosystems as yet are not fully understood (Tivy, 1987) although this is an active area of research and numerous different techniques and approaches are used. In LIA the main nutrient flow within soil-crop system has been analysed mostly by using correlation-regression analysis methods (Švedas et al., 1998; Bučienė, 1999). The recommended average, minimal and optimal fertilisation rates for different crops and soils in Lithuania are not often the most suitable compared with actual requirements because although fertiliser rates are adjusted with respect to soil texture, they do not consider any soil chemical properties. Therefore, the effectiveness of applied fertilisers is reduced, expenses for greater yield production are increased and environmental risks of contamination grow (Švedas, 1997; Švedas et al., 1998). Such an unbalanced fertilisation system still exists not only in the Kėdainiai district, but also in other

districts of Lithuania. The same was true for V. Liutkevičius farm in 1995-1997. After the nutrient flow and balance calculations have been made, discrepancies between the real and estimated yields based on the applied fertiliser rates were evident. In order to make fertilisation systems more precise, a balanced fertilisation computer model was created at LIA by A. Švedas and used for that purpose in the Demonstration farm (Švedas and Tarakan o v a s. 2 0 01 ) . T he co mp u te r mo d e l w a s developed after a compilation of empirical equations used for yield forecasts and relationships between crop yield, soil parameters and fertiliser efficiency. The model calculates N, P and K fertiliser rates depending on soil pH (KCl extracts), available P and K in the top-soil, organic matter and total nitrogen content, clay content, thickness of the top-soil, dry bulk density and targeted yield.

90 80 70 60 50 40 30 20 10 0

y measured y estimated

19951996

19961997

19971998

19981999

Hydrological years Fig. 2. Comparison of measured and calculated annual leaching of N-NO3 + N-NH4 with drainage runoff at Valinava during 19951999, according to the equation y = (5.887x1, +75.11x2 + 533.6/x3+0.01189x4 + 0.04002x5 - 3.77)0.01x3 (see text).

59

A. Bučienė et a!. I Europ. J. Agronomy 20 (2003) 53-62

300 Animal excrements

250 200

Biological fixation

150

Deposition

100

Mineral fertilisers

50 Grazing output

0 Output with grass yield

-50 -100

Leaching (N-NO3)

-150

Denitrification

-200 Ammonia emission

-250 Fig. 3. Nutrient balance and its components for N, P, K at the field level in the Demonstration farm in 1998.

not only in 1997, but also in 1995 and 1996 (Bearop in Lithuania, 1997). The main reason for this was incorrect management of the soil resource and non-rational fertilisation namely a surplus of P and shortage of K.

An agrochemical soil survey of this farm was completed in 1995 and repeated in two of the three fields situated in the 16.8 ha farmer's own land (Table 3). Data on fertilisation rates and crop yield for 1997 are shown in Table 4. The yields were low

Table 4 Crops, fertilisation and actual yield of the Demonstration farm, Graisupis, 1997 Field number

Area (ha)

Crops

Preceding crop

Manure applied (Mg ha -1 )

N Sugar beet Fodder beet

1

6

Barley

2

1

3

10

Ley Potatoes and vegetables Ley I and II year Oat-pea mixture of use

50 for preceding 12 crop 60 60 -

Yield (Mg ha -1 )

Application of mineral fertilisers (kg ha -1)

60

P

K

22

-

4.0

-

20.0

17 50

6.0

A. Bučienė et al I Europ. J. Agronomy 20 (2003) 53-62

60

Table 5 Data on estimated yield of the Demonstration farm using LIA computer model, Graisupis, 1995 1997 Crop

Yield (Mg ha-1) Without fertilisers

With average rate of fertilisers

1 field, 6 ha Winter rye Winter wheat Barley Oat Potatoes Sugar beets

3.5 4.4 3.6 3.2 23 32

5.0 6.1 4.7 4.4 37 48

2 field, 0.8 ha Winter rye Winter wheat Barley Oat Potatoes Sugar beets

3.4 4.3 3.4 3.1 22 31

4.8 5.9 4.5 4.3 35 46

3 field, 10 ha Winter rye Winter wheat Barley Oat Potatoes Sugar beets

3.6 4.5 3.7 3.3 24 33

5.2 6.2 4.9 4.6 38 49

requirement in this field. The same calculations were made for another two fields. The next step necessary is a soil analysis of the remaining land in the Demonstration farm and planning the crops, yields and appropriate fertilisation system in those fields. For the new soil testing it is necessary to analyse soils not only for pH, humus, N, P and K, but also for some micronutrients, determine the depths of topsoil and soil type and textural composition. There has been too little attention paid to the application of integrated plant protection measures in this farm. In order to obtain higher yields, it is necessary to use not only balanced fertilisation, but also integrated cultivation and plant protection measures, which will be a future development.

4. Conclusions

As indicated in Table 3, soils on the farm are productive and fertile and they could produce higher yields. These data were used to make a rational fertilisation model promising a competitive yield and minimal negative impact on the environment. There is still no established constant crop rotation in the Demonstration farm, which explains why fertilisation plans were variable for each field initially in 1997 (Table 5). In order to obtain the planned yields in the field Number 1 the optimum fertilisation rates should be as shown in Table 6 and applied in the spring. There are three types of crop rotation chosen for field Number 1 (6 ha) and the farmer could select the most suitable rotation and fertilisation type with respect to his resources. Type 1 includes only mineral fertilisers, types 2 and 3—mixed fertilisation system consisting of FYM and supplementary N and К commercial fertilisers, the crops like potatoes and sugar beet have a high N and К

Soil survey information as well as crop and yield planning needed to be extended to the newly maintained land in the Demonstration farm. In order to make more precise fertilisation advice systems for each field in the Demonstration farm, the use of the fertilisation computer model developed at LIA should be helpful. In addition to balanced fertilisation, more attention has to be given to integrated cultivation and plant protection measures. The regression equation (1) might be used for N balance calculations, if there arc sufficient data available on drainage discharge, total soil N content, humus content and fertilisation rate. For the major nutrients balances at the field level, it is necessary to pay greater attention to the more rational utilisation of organic fertiliser such as urine in this field and combine it with the same rate of commercial fertiliser in order to reduce the К deficit.

Acknowledgements We acknowledge Swedish colleagues from BEAROP project, Professor Arne Gustafson and Staffan Steineck for useful advises in analysis of

A. Bučienė et al I Europ. J. Agronomy 20 (2003) 53 - 62

61

62

A. Bučienė et al I Europ. J. Agronomy 20 (2003) 53-62

nutrient leaching data and helping with soil nutrient management methodology. Also thanks for the collaboration of the Lithuanian BEAROP team, Dr A.S. Šileika, Dr G. Kutra, Dr K. Gaigalis, Dr Z. Strusevičius. References Bearop in Lithuania, 1997. A project within the Baltic Agricultural Run-off Programme BAAP, Final report 1994-1997. Agricultural Run-off Management Land Report of Lithuania/by Lithuanian team: Šileika, A.-S., Gaigalis, K., Kutra, G., Bučienė, A., Strusevičius, Z.; Swedish team: Goran Carlson, Staffan Steineck, Arne Gustafson, Stefan Lofgren, Christine Jakobsson, Alf Gustafson), Lithuanian Institute of Water Management, Swedish Institute of Agricultural Engineering, Kėdainiai, Uppsala, 1997, p. 128. Bučienė, A., 1999. The methodological aspects of cropping systems complex research conducted on a small drainage plots. In: Agriculture, vol. 65 (in Lithuanian, with English abstract). Dotnuva-Akademija, pp. 27-47 ( i n Lithuanian, with English abstract). Bučienė, A., Šileika, S.-A., Gaigalis, K., 1998. Nutrient balance and management in r. Graisupis watershed and demonstration farm. In: Monographs in Systems Ecology, vol. 2. Klaipeda university, Klaipeda, pp. 6-11. Egnėr, H., Riehm, H., Domingo, W.R., 1960. Untersuchungen iiber die chemische Bodenanalyse als Grundlage fur die Beurteilung des Nahrstoffzustandes der Boden. II. Chemische Extraktionsmethoden zur Phosphor- und Kaliumbestimmung. Kungliga Lantbrukshogskolans Annaler 26, 199-215.

Fagerberg, В., Salomon, E., Steineck, S., 1993. The Computer Program NPK-FLO, vol. 9. Swedish University of Agricultural Sciences, Department of Crop Production ScienceInternational Publications, Uppsala, p. 47. FAO-UNESCO, 1997. Soil map of: the world revised legend with corrections and updates. Technical Paper 20, ISRIC, Wageningen. Švedas, A., 1997. Utilization problems of ecologically sensitive land. In: Scientific Social and Economic Problems of Utilization of Ecologically Sensitive and Non-productive Agricultural Land in Lithuania. Lithuanian Academy of Sciences, Lithuanian Ministry of Agriculture and Forestry, Lithuanian Agricultural Advisory Service, Vilnius, pp. 9-13 (in Lithuanian, with English abstract). Švedas, A., Antanaitis, Š., 2000. The relation between leached nitrate amount with drainage runoff and environmental factors. In: Agricultural Sciences 4, Lithuanian Academy of Sciences. Vilnius, pp. 24 31 (in Lithuanian, with English abstract). Švedas, A., Tarakanovas, P., 2001. Tręšimo Planavimas. Kompiuterio Programa "Tręšimas" (in Lithuanian). Akademija, p. 34 (in Lithuanian). Švedas, A., Dabkevičius, Z., Kadžiulis, L., Lazauskas, S., 1998. Utilization of Lithuania's climate and soil resources for agriculture. In: Ecological Sustainability of Lithuania in a Historical Perspective. Lithuanian Academy of Sciences International Centre for Scientific Culture World Laboratory Lithuanian branch. Lithuanian National Committee of the UNESCO programme "Man and Biosphere". Vilnius, pp. 72-82. Tivy, J., 1987. Nutrient cycling in agro-ecosystems. Applied Geography 7,93-113. Tyurin, I.V., 1931. A new modification of the volumetric method of determining soil organic matter by means of chromic acid. Pochvovedenie 26, 36-47.