Comparison of two fuzzy methods to determine the optimum soil depth in land suitability evaluation for wheat

Background and objectives: Land suitability evaluation as part of a sustainable agricultural ‎system, leads land utilization to a direction, which can preserve land quality in a desirable ‎level while ensuring the highest income. In the traditional studies of land suitability ‎evaluation, the land units fit between two classes can only view one of the primary classes are ‎defined for land suitability. Therefore, the method is not able to show the continuous reality ‎of land and soil spatial variability. The development of modern methods (such as fuzzy logic) ‎has solved this problem to a great extent. Therefore, the main objective of this research was to ‎determine the optimum depth of soil for quantitative land suitability evaluation for wheat ‎production using the fuzzy sets theory and comparison of two different fuzzification methods ‎of land suitability classes in the Shahediyeh region, Yazd. ‎
Materials and methods: After crop harvesting and taking soil samples from three depths of 0 ‎to 30, 30 to 60 and 60 to 100 cm in 90 observation points, and from genetic horizons of ‎representative pedon, which excavated in the region, the parameters needed for land suitability ‎evaluation of wheat were measured. Then, land suitability classes were calculated using the ‎square root method and were compared in three depths from 0 to 30, 0 to 60 and 0 to 100 cm. ‎Fuzzy calculations used for this study were conducted in two different ways. In the first ‎method, fuzzification of land index calculated for wheat was conducted as trapezoidal fuzzy ‎intervals. In the second method, firstly the variables used to calculate the land suitability ‎classes for wheat, including climatic parameters, and soil and topographic characteristics, were ‎defined as fuzzy numbers, and then the land index was calculated as a triangular fuzzy ‎number. To obtain the numerical values of the result of applied fuzzy method (defuzzification ‎process), the centroid method was used. Finally, to show the different land suitability classes, ‎the resulting numbers were converted to degrees of fuzzy membership. To compare the results ‎of two fuzzy methods used for different depths, the similarity index that presented by Sarma, ‎was used.‎
Results: The compatibility percentage between the results of representative pedon and ‎observation points was remarkable (more than 60 %) only for 0 to 100 cm depth. In addition, ‎the highest compatibility percentage of land suitability class was related to comparison of 0 to ‎‎60 and 0 to 100 cm depths in each of the two fuzzy methods. This fact shows that 0 to 60 cm ‎depth could be a relevant alternative for the optimal depth to evaluate land suitability for ‎wheat in the Shahediyeh region. The results also showed that there was no difference between ‎two fuzzy methods for all land suitability classes and at all the studied depths. ‎
Conclusion: Overall, considering the depth of 0 to 30 and/or 0 to 60 cm in the traditional ‎studies of land suitability evaluation for wheat is not sufficient compatible with the results of ‎representative pedon. Therefore, if the study is not intended to offer detailed information on ‎the type of limitations for civil actions, use of the results of representative pedon to a depth of ‎‎100 cm based on the traditional method is sufficient and it does not need to fuzzy ‎calculations.‎