Soil quality indices of paddy soils in Guilan province of northern Iran: Spatial variability and their influential parameters

•Four components explained 70% of the variability in the overall soil quality index.•Decrease in paddy soils quality was partly due to decrease in clay, COLE, P and K.•Decreasing clay content and COLE reduce available water for rice growth.•Fuzzy sets with non-linear scoring function provided more accurate quality indices.•Majority of area had low quality and severe limitation and need better management. Soil quality (SQ) is an important issue in soil science, ecology, agronomy, and environmental sciences that has been received increased attention due to its importance in the sustainability of ecosystem and soil management. Biological and physicochemical soil traits, as sensitive variables to changes in soil functions, mainly used as major parameters in SQ assessment. Soil quality index (SQI) is the most suitable index to evaluate the quality of soils. This study aimed to 1) evaluate the quality of paddy-field soils using three approaches (integrate quality index, IQI; physical index, PI; Fuzzy) to develop a credible SQI for rice cultivation and 2) model the spatial variability of the SQIs. 120 soil samples were taken from rice cropping land within the study area (Shaft and Fouman counties, Guilan province, and north of IR Iran). Fifteen soil variables were considered to calculate SQI of the paddy soils. Principal component analysis (PCA) approach applied to choose the minimum data set (MDS). Both non-linear and linear scoring procedures used to compute the SQI. Four principal components (PC) explained nearly 70% of the overall SQI variability. Fuzzy method clearly investigated the main limiting soil factors in paddy fields and the GIS-based SQ maps could be useful for decision-makers. The correlation coefficients (r) between the fuzzy results by non-linear scoring and rice yields were relatively good (0.70). Whereas, the results of Gupta model and yield had very low correlation (r = 0.24). Application of fuzzy sets with non-linear scoring function provided a more accurate assessment of SQ as compared with that obtained in other approaches. Majority of the study area had low SQ and the most severe limitation (SQI