Assessment of Different Methods of Soil Suitability Classification for Wheat Cultivation

J Agrobiol 29(2): 47–54, 2012 Journal of DOI 10.2478/v10146-012-0008-0 ISSN 1803-4403 (printed) AGROBIOLOGY ISSN 1804-2686 (on-line) http://joa.zf.jcu.cz; http://versita.com/science/agriculture/joa

ORIGINAL ARTICLE

Assessment of different methods of soil suitability classification for wheat cultivation

Amin Sharififar

Faculty of Agriculture, Shahrood University of Technology, Shahrood, Iran

Received: 24th November 2012 Revised: 1st April 2013 Published online: 31st October 2013

Abstract This study investigated the impact of soil temperature and soil moisture on the virulence To protect soil resources and also to achieve optimal crop production, it is essential to dedicate the most suitable land to a specific land use. Achieving this goal is possible through land use planning in conjunction with land evaluation. In this study a land suitability evaluation was carried out for wheat (Triticum aestivum) cultivation, and was performed in the Bastam region located in the north east of Iran. 104 soil profiles were sampled and 11 land units were separated. In order to find out the most correct method of physical land suitability evaluation, three methods of combining soil criteria for soil index calculation for wheat production were tested. These methods are based on parametric and maximum limitation approaches, and the results of each method were compared with the observed yield. Ultimately, the maximum limitation method was found to be the best method and was used for classification of the suitability of the study area lands for wheat cultivation. The varying results of applying different ways of evaluation in this study indicate that the accuracy of the method of land evaluation adopted should be checked before using the results for any purposes.

Key words: Almagra model; Bastam; land evaluation; MicroLEIS; soil index

INTRODUCTION vital for wise planning of land use. Before making any decisions about dedicating lands for any Scientific recognition of land resources and agricultural uses, land suitability evaluations possible land exploitations, as well as interactions should be implemented. Technically each land between specific land units with a specific use, is unit should be used for an application which is suitable for that application (FAO 1978). Suitable land use planning paves the way for sustainable development.  Amin Sharififar, Faculty of Agriculture, Land evaluation makes it possible to use lands Shahrood University of Technology, Shahrood, according to their biophysical potentialities and 36199-95161, Iran limitations, in order to protect soil resources  [email protected] from degradation and at the same time to meet farmers’ demands for optimal crop production.

47 Journal of Agrobiology, 29(2): 47–54, 2012

If there is a significant correlation between varied from 1350 to 1900 m above sea level but predicted yield and observed yield, then the elevated lands are not cultivated and the altitude results of the evaluation method would be of agricultural lands does not vary significantly. accurate and can be applied for land suitability The total area surface was about 53,500 ha. Slope classification and management. Two general gradient varied from flat to 8%. The physiography methods of land evaluation have been presented of the studied land units included: Gravelly by Sys et al. (1991b). Those methods are the Alluvio-Colluvial Fans, Pidmont Plateax and maximum limitation method and parametric Alluvial Plains. According to the bioclimatic map method. The so-called Almagra model (found in of the region (FAO 1988), the study area possesses the MicroLEIS software) which is based on the an attenuated sub-desert climate. Major land maximum limitation method for land suitability uses of the study area are agricultural, pasture assessment, is applied in this study. and fallow lands. As wheat is a strategic crop in MicroLEIS DSS (Decision Support System) many countries including the study area and also software, containing the Almagra model was built most of the farmers dedicate a high surface area in the Mediterranean climate (De la Rosa et al. of the region studied to wheat cultivation each 2004) and has been recalibrated and revalidated year, this crop was selected for evaluation to be in semi-arid regions in west Asia (Shahbazi et al. tested for soil suitability evaluation. 2009). Some case studies have used the Almagra model for land suitability evaluation (Darwish et Soil sampling and analysis al. 2006, Wahba et al. 2007, Shahbazi et al. 2008, In total, 104 soil profiles were investigated and 2009, Jafarzadeh et al. 2009) but unfortunately among those, eleven representative profiles researchers in such case studies have not were selected. Therefore 11 representative land investigated their method in comparison with mapping units, taxonomically classified to the other existing methods to find out whether the family level, were separated. The procedure of method that was used is the correct one. Models taxonomic land classification was adopted from like Almagra need to be validated when they are the soil taxonomy manual of the United States used in areas other than those for which the model Department of Agriculture (USDA 2010). This was calibrated and validated. The evaluation classification is based on field surveys and methods discussed in this paper have been used in morphological descriptions such as leaching other regions but most of those researchers have evidence, the position of soil horizons and their tested only one method of evaluation and did not depth, and chemical and physical analyses compare the results of different methods for land such as: electrical conductivity, organic carbon, evaluation. Some of them have not investigated exchangeable sodium percentage, cation the correlation of predicted results with the real exchange capacity, carbonate content, texture, observed yield of the crop. The objective of this structure, etc. These analyses were carried out study is to select the most correct land evaluation using standard methods of soil analysis in the method and then determine the suitability of laboratory. Land unit separation was carried land units for wheat cultivation. out by field investigations. Some climatic data, including temperature and precipitation rates were also used in the taxonomic land MATERIALS AND METHODS classification, which also showed Aridisols and Entisols as dominant soil classes in the Study area description region. Soil moisture regimes were Aridic and This study was performed in the Bastam region Torric, and the thermal regime was Mesic. The in the Semnan province located in the north east geographical position of the land mapping units of Iran. The study sites were located between (soil families) of the region is shown in Fig. 1 coordinates 54° 39′ to 55° 20′ of east longitude and the measured site characteristics are shown and 36° 26′ to 36° 45′ of north latitude. Altitude in Table 1.

48 Journal of Agrobiology, 29(2): 47–54, 2012

Table 1. Mean values of land characteristics of the study area

Land units Texture1 Electrical Carbonate ESP2 Rooting pH Drainage Stoniness class conductivity content % depth (cm) class % (ds/m) Abr Sandy Loam 0.4 31 2.7 95 7.8 Good 15–35 Amir Clay Loam 1.7 23.7 3.1 145 7.9 Moderate <15 Bagh Loamy Sand 1.0 5.5 2.8 22 7.9 Good 30–75 Bastamy Loam 1.7 28.5 6.3 160 7.8 Moderate <15 Bayazid Loamy Sand 1 14.7 4.1 95 8.0 Good 15–35 Kharaqan Silty Clay 6.8 55.8 13.2 125 7.8 Moderate <15 Khazaneh Sandy Loam 0.5 24.0 2 40 7.9 Good 15–35 Khij Sandy Loam 1.7 36.0 1 86 7.8 Good 10–25 Mojen Silty Loam 0.8 41.5 2.6 150 8.0 Moderate <15 Qaleh Silty Clay 2.7 30.2 5.8 150 8.0 Moderate <15 Loam Qehej Sandy Loam 0.7 20 1 140 8 Good 46

1 Texture classes as well as all other parameters values are mean values of the soil profile horizons, according to the instructions in Sys et al. (1993) 2 Exchangable Sodium Percentage of soil

Evaluation procedure given to each criterion. The total score for a In this study three ways of determining the total special land unit is also given a rate of 0 to 100 by ranking of every specific land unit are applied calculation through the three methods discussed and their outcomes are compared with soil in this paper as ways of combining the criteria productivity. The three methods, explained by Sys scores. The procedure of the maximum limitation et al (1991a), are the Storie method, the square method is the the selection of the most restricting root method (Khiddir 1986) and the maximum criterion rate and considering it as the total score limitation method. The Storie method and the for a land unit. square root method can both be subsumed under The wheat yield data in each land unit were the rubric of parametric methods. The equation used as indicators of soil potentiality which in (1) and (2) show the Storie and square root turn indicates soil productivity. However, the methods respectively. yield is not dependent only on soil and land characteristics but could also be influenced by SI = (A) × (B/100) × (C/100) × … (1) managerial factors and other factors such as diseases, which have not been considered in the SI: Storie index land evaluation. However, after interviews with A, B, C: ratings of criteria the farmers of the region, managerial differences among land units of the region were seen as

I = (Rmin) × √A/100 × √B/100 ×… (2) unimportant, and therefore the soil productivity potentiality level can be distinguished using crop I: index of square root method yield.

Rmin: the minimum rated criterion A, B, … : criteria other than minimum rated Crop potential yield criterion To compare the predicted soil suitability (potentiality) for wheat growth with the actual The rate for each criterion is obtained yield, the score (called the soil index) of each after field or laboratory measurements of the land unit, having been calculated using three land properties, and the comparison of these methods, is multiplied by the crop potential yield, measurements with the crop requirements in the and the outcome is compared with the observed reference tables. After matching measurements yield in each land unit. In this study the FAO- with threshold values, a rating of 0 to 100 is AEZ method (Food and Agriculture Organization

49 Journal of Agrobiology, 29(2): 47–54, 2012

Agro-ecological Zoning project) (FAO 1996) was Where, bgm is maximum gross biomass used to calculate the wheat potential yield. In production (kg CH2O ha.hr), KLAI is maximum the AEZ model, biomass production and the crop growth ratio, Hi is harvest index, L is growth yield is calculated by using sunlight radiation period (number of days), Ct is respiration and temperature in an ideal situation in terms coefficient and Y is crop potential yield (kg ha). of water and nutritional requirements, and The potential yield is not affected by soil diseases. The following formula is the formula for characteristics and cultivation management. A calculating crop yield: detailed explanation of the calculation of crop potential yield can be found in Sys et al. (1991a) Y = 0.36 bgm.KLAI.Hi / [(1/L) + 0.25Ct] (3) and Ayoubi and Jalalian (2010).

Fig. 1. Soils taxonomic classification of the study area

50 Journal of Agrobiology, 29(2): 47–54, 2012

Almagra model shown in Table 3. The potential yield for wheat The Almagra model was applied to determine was calculated as 6709.4 kg ha, using the AEZ the suitability classes as a method of maximum method. The correctness of this method for limitation. The Almagra computer-based model, calculating potential yield has been tested in makes the evaluation task easier and more different parts of the world through the FAO agro precise, since human error is reduced using ecological zoning projects. this method. Complete information about the In the country of the study area, several studies MicroLEIS package can be found in De la Rosa have also shown the validity of the AEZ method et al. (1992, 2004). The parameters considered as for calculating potential yield: Ayoubi et al. (2002) data inputs for the Almagra model are shown in determined the potential yield of wheat, barley Table 2. and corn in Isfahan in the centre of the country as 9.08, 9.7 and 12 ton ha respectively; Bazgir et Table 2. Input parameters considered for Almagra al. (2000) also determined the potential yield of model wheat and barley as 7634 kg ha and 7487 kg ha respectively in Kermanshah in the west of the Rooting depth country. The AEZ method for calculation of the Texture class potential yield of wheat, corn and sesame was also Drainage class applied by Rostaminia (2001) who obtained 7.42, Carbonate content 9.22 and 1.44 ton ha for those crops respectively in Ilam which is in the southwest of the country Salinity of the study area. Degree of development of the soil profile The correlation of the predicted potentiality of wheat cultivation, calculated through the three methods, with real wheat yield in different RESULTS AND DISCUSSION land units is shown in Fig. 2. The yield values (Table 3) are the average of harvests over several The results of total land ranking for every years for the region registered in the Agriculture land unit by using the Storie, square root and Organization of Bastam and are used as basic maximum limitation (the Almagra model) are data.

Fig. 2. Results of correlations between the three methods of land evaluation and observed yield of wheat

51 Journal of Agrobiology, 29(2): 47–54, 2012

Table 3. Results of land units ranking by Almagra model, square root method and Storie method compared with observed yield

Land units Almagra Square root Storie Mean observed yield of wheat (kg ha) Abr S41t2 (30)3 21 15.09 2200 Amir S2t (70) 54.7 42.79 3200 Bastamy S2ta (68.75) 46 30.79 4000 Bayazid S5t (15) 15 15 – Kharaqan S3cs (57.5) 28.9 14.62 2800 Khazaneh S4t (30) 28.7 27.6 – Khij S4t (30) 24 19.5 2100 Mojen S2t (66.66) 52 40.74 3000 Qaleh S2ca (78.3) 64 52.97 3400 Qehej S4t (30) 29 28.5 2000

1 S, suitability class; S1, optimum, S2, high, S3, moderate, S4, marginal 2 Soil limitation factors; a, sodium saturation; c, carbonate content; s, salinity; t, texture 3 Values in the parenthesis are the quantified rankings

The maximum limitation method shows the Storie soil index the and observed yield (R2 = 0.72) best correlation with soil actual potentiality. To for oil palm was found by Embrechts et al. (1988). find the significance of correlations, the t-Student Different soil indices found by other researchers test was performed. The answers obtained by the means that the Storie equation gives different test were compared with the threshold values results when the scores for the characteristics at α=0.05 and α=0.1 (certainty level of 95% and vary. The maximum limitation method for land 90% respectively). The results showed that the suitability evaluation has been used by other correlation between the yields predicted using the researchers such as Biox and Zinck (2008) as a maximum limitation method, and the observed sound method. They determined the suitability wheat yields was significant at both levels of of their study area lands for alfalfa, maize, certainty. But the correlation between both the wheat and some other crops. Examples of land square root and the Storie methods with the suitability studies in agriculture can be found observed yield were not significant statistically in the literature (Kalogirou 2002, Ceballos-Silva at the two levels of certainty. Therefore, the and Lopez-Blanco 2003a, b). results of the Almagra model were used for the The higher correlation of the Almagra classification of the suitability of the region for model with the observed yield is maybe due to wheat cultivation. The results of this type of consideration of the most limiting characteristic evaluation can be used for future remediation of which controls crop production, because in soils, since the restricting factor is distinguished. the maximum limitation method there is The best site for wheat production among all the an assumption that it is the most limiting arable sites was the Bastam site and the worst characteristic that determines the productivity was the site Bayazid site (Table 4). The most of a land. The maximum limitation method is limiting factor in the Khazaneh, Abr and Bayazid used when we are not sure about the quality sites was texture and in the Kharaqan site it was and quantity of complex interactions among also soil texture as well as high carbonate content land characteristics and how to combine those and salinity. characteristics. Therefore, the maximum The Almagra model presents qualitative limitation method gives a more objective result in suitability classes for crops, but the qualitative comparison with other methods that combine the classes were quantified by referring to the software land characteristics scores. database. This is needed for a quantitative The method used to combine the land criteria comparison between Almagra and the other two to come to a final total score for a specific land methods. A significant correlation between the unit in the land evaluation process is important

52 Journal of Agrobiology, 29(2): 47–54, 2012 because different ways of calculating total land not considered, as we assumed that climate index result in different outcomes, as has been variability among land units is negligible in shown in this research. To make a correct and the study area. In this study, a soil-specific precise decision on land management, a precise suitability classification of separated land units and sound method of land evaluation should be was carried out for wheat cultivation. Usually, adopted, and a precise and sound method can be increasing agricultural land capability correlates obtained through testing the correctness of the with a decrease in soil erosion processes, and a method. positive correlation between current land use and Two general factors including soil attributes potential land capability would be necessary (De and climatic situation, affect crop growth. In la Rosa and Van Diepen 2002). the Almagra model, climatic parameters were

Table 4. Current land uses and surface area of different land units

Land units Surface area (ha) Surface % Current land use Slope % Abr 8125 15.19% Wheat and fallow lands 0–5 Amir 1525 2.85% Agricultural lands and pasture 0–2 Bagh 7250 13.56% Pasture 5–8 Bastamy 3975 7.43% Wheat, potato, perennial crops and vegetables 0–2 Bayazid 3525 6.59% Pasture 2–5 Kharaqan 3550 6.64% Wheat, barley and perennial crops 0–2 Khazaneh 9450 17.66% Pasture 5–8 Khij 7225 13.50% Pasture and annual crops (mainly wheat) 2–5 Mojen 1725 3.22% Wheat and potato cultivation 0–2 Qaleh 4350 8.13% Wheat and barley 0–2 Qehej 2800 5.23% Wheat and pasture 2–5

The evaluation of methods tested produced barley, corn and rice in Isfahan. Science and the following results: the Almagra model as an Technology 3: 105–120 (in Persian). application of maximum limitation method has Ayoubi S, Jalalian A (2010): Land evaluation the highest correlation with the observed yield (agriculture and natural resources uses). of wheat and can be considered a sound method Isfahan University of Technology press, for land suitability classification. This study Isfahan, Iran (in Persian). demonstrated that evaluation of the suitability Bazgir M, Givi J, Jalalian A (2000): Quantitative of lands using the Almagra model is accurate for and economical land suitability evaluation. In wheat cultivation in the region studied and that Abstracts of 6th Iranian soil science congress this method can be used in other similar regions. articles. Soil Science Society of Iran. Isfahan, Iran: pp. 196–197 (in Persian). Biox LR, Zinck JA (2008): Land-use planning in ACKNOWLEDGEMENT the Chaco Plain (Burruyacu, Argentina). Part 1: Evaluating land-use options to support The author would like to thank the Soil and Water cop diversification in an agricultural frontier Research Institute of the Ministry of Agriculture area using physical land evaluation. Environ and also Agriculture Organization of Bastam for Management 42: 1043–1063. cooperating in data collection for this research. Ceballos-Silva A, Lopez-Blanco J (2003a): Delineation of suitable areas for crops using a Multi-Critera Evaluation approach and land REFERENCES use/cover mapping: a case study in Central Mexico. Agr Syst 77: 117–136. Ayoubi S, Givi J, Jalalian A, Amini A (2002): Ceballos-Silva A, Lopez-Blanco J (2003b): Quantitative land evaluation for wheat, Evaluating biophysical variables to identify

53 Journal of Agrobiology, 29(2): 47–54, 2012

suitable areas for oat in Central Mexico: a Khiddir SM (1986): A statistical approach in the multi-criteria and Gis approach. Agr Ecosyst use of parametric systems applied to FAO Environ 95: 371–377. framework for land evaluation. Dissertation, Darwish KM, Wahba MM, Award F (2006): State University of Ghent, Belgium. Agriculture soil suitability of Haplo-soil for Rostaminia M (2001): Qualitative and quantitative some crops in newly reclaimed area of Egypt. land suitability evaluation in Mehran plateau. J Appl Sci Res 2: 1235–1243. MSc thesis. Isfahan University of Technology, De la Rosa D, Van Diepen C (2002): Qualitative Isfahan, Iran. and quantitative land evaluation. In Verheye Shahbazi F, De la Rosa D, Anaya-Romero M, W (ed.): Encyclopedia of Life Support System Jafarzadeh A, Sarmadian F, Neyshaboury (EOLSS-UNESCO), Section 1.5. Land use and M, Oustan S (2008): Land use planning in land cover. Eolss Publisher, Oxford. Available Ahar area (Iran) using MicroLEIS DSS. Int at: http://www.eolss.net Agrophys 22: 277–289. De la Rosa D, Moreno JA, Garcia LV, Almorza Shahbazi F, Jafarzadeh AA, Sarmadian F, J (1992): MicroLEIS: A microcomputer-based Neyshabouri MR, Ostan S, Anayan-Romero Mediterranean land evaluation information M, De la Rosa D (2009): Suitability of wheat, system. Soil Use Manage 8: 89–96. maize, sugar beet and potato using MicroLEIS De la Rosa D, Mayol F, Diaz-Pereira E, Fernandez DSS Software in Ahar Area, North-West of M, De la Rosa D, Jr. (2004): A land evaluation Iran. J Am Environ Sci 5: 45–52. decision support system (MicroLEIS DSS) Sys C, Vanranst E, Debaveye J (1991a): Land for agricultural soil protection with special evaluation, part I. Principles in land evaluation reference to the Mediterranean region. and crop production calculation. International Environ Model Soft 19: 929–942. training center for post graduate soil scientist Embrechts J, Zulkanian P, Sys C (1988): Physical and Ghent University, Ghent. land evaluation using a parametric method Sys C, Vanranst E, Debaveye J (1991b): Land application to Oil palm plantation in north- evaluation, part II. Methods in land evaluation. Sumatra, Indonesia. Soil Survey Land Eval 8: International training center for post graduate 111–122. soil scientist and Ghent University, Ghent. FAO (1978): Report on the agro-ecological zones Sys C, Van Ranst E, Debaveye J (1993): Land project. World Soil Resources Report 48. FAO, evaluation. Part 3: Crop requirements. Rome. Agricultural Publications 7, 3. General Admi- FAO (1988): Soil map of the world, revised 1988. nistration of Development Cooperation of UNESCO, Paris. Belgium, Brussels. 199 p. FAO (1996): Agro-ecological zoning guidelines. USDA (2010). Keys to soil taxonomy. United States FAO Soil Bulletin No 76. FAO, Rome. Department of Agriculture, Natural Resources Jafarzadeh A, Shahbazi F, Shahbazi M (2009). Conservation Service, Washington, DC. (ftp:// Suitability evaluation of some specific crops ftp-fc.sc.egov.usda.gov/NSSC/Soil_Taxonomy/ in Souma area (Iran) using Cervatana and keys/2010_Keys_to_Soil_Taxonomy.pdf) Almagra model. Biologia 64: 541–545. Wahba MM, Darwish KhM, Award F (2007): Kalogirou S (2002): Expert systems and GIS: Suitability of specific crops using MicroLEIS an application of land suitability evaluation. program in Sahal Baraka, Farafra Oasis, Comput Environ Urban 26: 89–112. Egypt. J Appl Sci Res 3: 531–539.