Considering the fact of the barren coastal saline farmland, the large spatial variability of soil properties, and the low benefit of extensive management, the precise division method of management zones is studied, to take differential measures to improve the level of land use of saline land. In this paper, taking farmland in Wudi County as a research area, we used grid method combined with fixed-point field sampling of land use status and laboratory analysis to obtain soil attribute data. We applied geo-statistical method to analyze the spatial variation characteristics of soil attributes in ArcGIS 10.2. Fuzzy c-means algorithm (FCM) was used to calculate the fuzzy membership degree of each sample point in MATLAB R2016a. The interpolation was used to predict the spatial distribution of the fuzzy membership degree. The partition was carried out based on the principle of maximum membership degree. The accuracy of partition results was verified by variability analysis and LSR difference significance test. The results showed that the farmland soil in Wudi County was slightly and moderately salinized with low content of available nitrogen, medium content of organic matter and available phosphorus, and high content of available potassium. The content of organic matter, available nitrogen, available phosphorus, available potassium, and salt showed a moderate variability (coefficient of variation 25.0%-52.3%). The spatial variability was relatively large to be regulated in different zones. The nugget effect values of available potassium, salt content and pH were less than 25%, which were mainly influenced by the soiltexture,groundwatersalinityandother structural factors. The nugget effect values of organic matter, available nitrogen, and available phosphorus were between 50% and 75%, which were greatly influenced by cultivation methods, fertilization and other human random factors. Therefore, the farmland in the county was divided into three zones with estimated area of 25.6 thousand hm², 17.6 thousand hm² and 32.4 thousand hm². The coefficient of variation of soil nutrients in the three zones was 23.9%-51.5%, 15.9%-50.3%, and 14.7%-33.0%, respectively. The results showed that there were significant differences among the three zones, while the variability within the zones was significantly lower than that in the non-zones. And the management zoning and the spatial distribution characteristics of soil attributes have a high degree of fitness. The results of zoning can be used as the operational units of differentiated management. The results provide a basis for the unified management within each division and differentiated management among different zones, which is helpful to improve the precision management level of coastal saline farmland.