Support Vector Machines Based-Modeling of Land Suitability Analysis for Rainfed Agriculture

Abstract

Soil evaluation plays important role in the sustainable agriculture development. Based on the value of several soil and environment indicators, the agricultural land evaluation methodology is applied to land mapping units in order to compute the suitability index. This index characterizes these land-mapping units. However, there are different methodologies which have been reviewed for land capability and suitability evaluation. In the present study, the potential use of support vector machines (SVMs) algorithm was evaluated for land suitability analysis for rainfed wheat based on FAO land evaluation frameworks (FAO, 1976, 1983, 1985) and the proposed method by Sys et al. (1991). The study area was divided into thirteen land units (with thirty two representative soil profiles) and ten land characteristics including climatic (precipitation, temperature), topographic (relief and slope) and soil-related (texture, CaCO3, OC, coarse fragment, pH, gypsum) parameters were considered to be relevant to rainfed wheat. In this study economic factors have been excluded and moderate management has been assumed. The data points were divided by randomization technique and 80% data was selected to train the model and the remaining 20% was used to test the model. The Root Mean Square Error (RMSE) and coefficient of determination (R2) were used as evaluation criteria. The results showed that the corresponding values for RMSE and R2 between the measured and predicted land suitability indices using the SVMs model were 3.72 and 0.84 respectively. Moreover, the most important limiting factors for rainfed wheat cultivation are climatic and topographic conditions, and 84.38% of total lands are classified as S2 class (moderately suitable) while the remaining 15.62% are classified as S3 class (marginally suitable). It appears that SVMs approach could be a suitable alternative to performance of land suitability scenarios.