Groundwater resources are one of the main sources of water supply for agricultural sector in Iran. Therefore, this study aimed to investigate the situation of groundwater salinity in Mazandaran for use in agriculture. In this study, statistical analysis of collected data, proper semivariogram model selection, cross validation of predictions and preparing probabilistic and zoning maps using geostatistical tools in the ArcGIS software, were performed. To investigate the spatial variations and preparing zoning maps of water salinity, ordinary kriging (OK) was used and the zoning maps were prepared. Spatial structure of electrical conductivity (EC) assessment showed a moderate spatial dependence in most years. Zoning and probabilistic maps of EC showed that the salinity of groundwater will be added and the most probable salinity is related the lowland areas in the eastern part of the plain. The use of this groundwater for irrigation in the long term can decrease the rice yield and faced rice production with a serious risk. The results of the Mann-Kendall and the Sen tests indicated a decreasing trend in the area of groundwater with EC higher than one dS/m in Mazandaran plain that this expressing an improvement in the quality of groundwater in the plain.