علوم محیطی (Sep 2023)
Comparison of remotely sensing and meteorological data-derived drought indices in mid-western Iran
Abstract
Introduction: Drought, as one of the major natural hazards, affects the environment, society, agriculture and economy. Several indices have been developed for drought quantification based on ground data and remote sensing. Traditional drought quantification methods are based on meteorological data and conventional criteria and are usually not available in near real- time. On the other hand, data based on remote sensing are continuously available and can be used to detect several aspects and characteristics of drought. The purpose of this research is to investigate and compare different indices derived from remote sensing and meteorological data for local scale drought monitoring (eastern part of Kurdistan Province).Material and methods: Seven drought indices were compared, including Vegetation Condition Index (VCI), Vegetation Drought Index (VDI), Vegetation Health Index (VHI), Vegetation Supply Water Index (VSWI), Normalized Difference Vegetation Index (NDVI), Temperature Condition Index (TCI) and Standardized Precipitation Index (SPI). Remote sensing indicators are derived from MODIS data. The meteorological index SPI is obtained by combining the data of rain gauge stations and gridded precipitation data. The digital maps of the seven drought indicators have been prepared for the period of 2002-2021 with the same time interval (16-days). To analyze the characteristics of each drought index, a comparative method including the selection of specific periods of drought and spatial drought identification characteristics has been used. The comparison of drought indicators was done May, which is the growing season. Finally, Pearson's correlation analysis was used to evaluate the behavioral similarity of the indicators.Results and discussion: The spatial comparative analysis between the drought indicators showed that all the indicators had certain adaptations in the distribution of the regional scale of drought, especially those derived from similar data sets. Meanwhile, the difference in local scale distribution was found among different groups of indicators. The results showed that the general trend of the VSWI had a better compliance with the standardized precipitation index. Based on the correlation analysis, it was proved that the VSWI can be a better reflection of the amount of rainfall and the severity of drought due to the lack of rainfall. In addition, the land surface temperature (LST) contributes more to the VSWI results than the reflectance information. A two-period (32-day) delay of the indicators indicating the state of vegetation is a good indicator of the meteorological drought conditions in the study area. The absence and lack of rainfall in at least five periods (80 days) earlier can had a serious effect on the state of vegetation in the existing conditions. Plain and mountainside areas located in the central, eastern and south-eastern parts of the study area were more sensitive to drought conditions than other parts due to the dominance of grain farming, especially rainfed farming.Conclusion: While remote sensing drought indicators have many advantages in the analysis of drought in real-time, meteorological drought indicators are still the priority for drought monitoring. This is due to the dependence of hydrological and agricultural systems on meteorological conditions. Mainly, these hydrological and agricultural systems in different regions respond to meteorological fluctuations with different time delays. Understanding these complex relationships between meteorological, hydrological and agricultural systems can be useful in early preparation programs against drought and its management.
Keywords
