---

This study is aimed at identifying of depopulated rural settlements over a quarter century since 1365 (1986) by 1390 (2011) in Isfahan province and representing a geographically interpretation of the phenomena, using statistical methods and spatial analysis in the geographic information system environment. To do this, at first, layers of the spatial distribution of settled villages of 1365 and identified depopulated ones over the two and a half decades period of the study have based on the detailed reports of public censuses’ results of Iran statistical center identified and prepared in ArcGIS environment. Afterwards, the rural settlements exodus analyzed based on the prepared, corrected and classified natural parameter layers of altitude, slope, aspect, temperature, land types and distance from wells and springs, spatially. Definition and calculation of two spatio-statistical indicatorsdad been the basis of analysis; “the settled villages density at the beginning year of 1365” and “relative percent of depopulated villages over the study period with respect to the total beginning settled ones”. The indicators calculated for each of the layers’ zones using the “zonal mean statistic” function from the spatial analyst tool of ArcGIS and analyzed statistically by correlation and Chi-Square tests. Results revealed that there is a significant relationship between exoduses and natural factors of slope, land types, altitude and distance from wells respectively. While that was not true for aspect parameter. The influence knowhow of natural factors on rural exodus, helps the landuse planning projects to more consider them and ensure the more stability of rural settlements agains the more exodus in the future.

---

The main purpose of this research is to monitor the groundwater level using remote sensing science and satellite images and its relationship with land use. For this purpose, the relevant images were taken first and the necessary pre-processing was applied on each of them. Then the images were modelled and classified. In order to study land use changes, the land use classification map was extracted for 2002 and 2018 years using the object-oriented classification method and then to study land use changes, the land use change map was extracted for a period of 16 years. The highest rate of change is related to the use of rangeland to rainfed agriculture, rangeland to irrigated agriculture, forest to irrigated agriculture and rainfed agriculture to irrigated agriculture. Also, among the modified land uses, forest land use has the lowest increase. After extracting the land use change map in order to select the best interpolation model from different models, all models were evaluated and the Kriging method was more accurate than other methods, which among the different modes of the kriging method Also, K-Bessel model for 2002 and Circular model for 2018 have the highest accuracy. The results of groundwater survey showed that the highest and lowest average water level in 2002 belongs to rainfed agricultural use and water use and in 2018 rainfed agricultural use has the highest average water level and forest use has the lowest average water level.