Monitoring the Effects of Drought on Vegetation Cover and Ground Water Using MODIS Satellite Images and ANN

The main aim of the study was to investigate the effects of drought on vegetation cover and ground water resources. In the present study, an available climatic data series (2001-2017) for 9 synoptic stations in Lorestan province were analyzed to detect wet and dry years using Standardized Precipitation Index (SPI). Furthermore, a long data series of Moderate Resolution Imaging Spectroradiometer (MODIS) data was analyzed by remote sensing data and the Normalized Difference Vegetation Index (NDVI) maps have been produced for the study period (2001-2017). For all data, Kolmogorov-Smirnov test and Pearson Correlation Coefficient test between SPI and NDVI were used based on the data resource and normality test. In addition, the relationship between rainfall and groundwater levels was investigated using artificial neural network (ANN). During the study period, 2008 and 2015 were selected as dry and wet years based on SPI values, respectively. The values of the NDVI in the wet year (2015) are significantly higher than the values in the dry year (2008) at a 99% confidence level. Spatial variation of SPI shows that for intensive drought conditions (2008) and wet year (2015) the northern part of Lorestan province had the highest variation in comparison with other parts of the study area. Generally, the results of the present study show that MODIS data in a mountainous area could be a key tool in detecting the effects of intensive drought on natural vegetation cover. Furthermore, ground water level showed a significant correlation with the 3-month delay of monthly precipitation.

Automated summary

The study investigated the effects of drought on vegetation cover and groundwater resources by analyzing climatic data and MODIS data. Dry and wet years were identified based on SPI values, with higher NDVI values in wet years compared to dry years. The spatial variation of SPI showed the northern part of Lorestan province had the highest variation. The study suggests that MODIS data can be a key tool in detecting the effects of intensive drought on natural vegetation cover.