Delineation of site specific nutrient management zones for a paddy cultivated area based on soil fertility using fuzzy clustering

Delineation of soil nutrient management provides basis for variable fertilization management in precision agriculture. The present fertilizer recommendations for paddy fields in Iran are typically uniform for large regions. This results in over-application of fertilizers in areas with high nutrient levels and under-application of those, in areas with low nutrient levels. A 24,000-ha paddy field was selected to define management zones (MZs) for more precise soil nutrient management. Soil samples were taken from 303 locations. Soil samples were analyzed for clay, pH, cation exchange capacity (CEC), organic carbon (OC), total nitrogen (TN), available phosphorous (AP), and available potassium (AK). After analyzing the spatial variability of these properties with geostatistical technique, they were used to delineate soil fertility management zones using principal component analysis (PCA), and fuzzy cluster algorithm. Results indicated that the optimum number of MZs for soil fertility was four. Analysis of variance indicated that there is heterogeneity in soil fertility status among different MZs. The mean values of soil nutrients in each zone can be used as a reference for variable-rate fertilization. Management zone 4 represents the most AK deficiency where soil K-supplying capacity was low. Furthermore, these results indicated that cluster analysis by reducing within-zone variability provided an opportunity for farmers to adopt site-specific nutrient management.

► Available K (AK) is a constraint to rice yield in Iran, while it is spatially variable.
► For-site specific K management, fuzzy cluster analysis divided the region in to four zones.
► For each zone, the AK is presented that is a basis for K management.
► For fertilizer variable rate technology yield-based system should be developed.