Spatio-temporal variation of meteorological drought and its relation with temperature and vegetation condition indices using remote sensing and satellite imagery in Marvdasht city

Introduction
Drought is considered a complex hazard, whose severity depends on the climate and weather conditions of each region. In fact, drought is caused by dry and unusual weather conditions, among other things, lead to a change in vegetation characteristics. Since this dangerous phenomenon is caused by the lack of rainfall for a long period of time, it slowly and gradually leads to a natural disaster and conquers the environment compared to other environmental hazards. Therefore, it is not noticed and taken less seriously by people and authorities. Undoubtedly, the occurrence of drought and as a result the crisis of reduction and shortage of water resources is one of the main and most important risks of the natural environment that humans have faced since the past. Therefore, it can be stated that the possibility of this natural phenomenon also exists in humid areas. Drought causes unfortunate and in some cases irreparable damage to human life as well as the natural ecosystem which is very different from other natural events such as floods, storms, and earthquakes. So that it has caused wide and big problems in the economic, social, political, and cultural fields. Therefore, the impacts it causes are not only structural and the damage it causes affects different areas. Drought is one of the destructive phenomena of the natural environment that affects a significant number of countries and causes problems. Simultaneous droughts with the period of vegetation growth cause environmental ecosystem consumption, which results in biological compounds such as land surface, soil amount, and plant growth rate, that we need proper management and planning in order to deal with this phenomenon.
 
Materials and Methods
In the present study, drought in Marvdasht city was analyzed using remote sensing technology and satellite imagery as a time series. For this purpose, during the statistical period of 20 years (2000-2019), out of 460 satellite imagery of land surface temperature (LST) and vegetation cover (NDVI) were used in conjunction with the MODIS sensor of Terra satellite, from which to estimate the temperature condition index (TCI) and vegetation condition index (VCI) was used. The optimal index indicating the state of drought from satellite imagery, the SPI was also used. In this way, using the rainfall data recorded by synoptic and rain gauge stations in the study area, the SPI was calculated using MATLAB software for the period of 3, 6, and 12 months. One of the other goals pursued in this study is to determine the basic and optimal index, indicating the state of drought in the study area, which is based on TCI and VCI satellite drought indices. Thus, after calculating the SPI and its intervals for each station, the points of each ground station were placed on the maps produced from TCI and VCI satellite indicators. Then, by taking the numerical value of the corresponding points for each of the years of the investigated period, the obtained values were entered into SPSS 22 software and the amount of correlation coefficients between SPI and its intervals with TCI and VCI values was calculated.
 
Results and Discussion
According to the drought maps extracted from the TCI, the highest amount of land area with very severe drought in 2016 was 118.90 km², and then in 2018 with 112.25 km², and in 2017 with an amount of 101.66 km² has happened. On the other hand, the least amount of extreme drought area in the first place is related to 2006 with an area of 46.10 km², and then 2002 with an area of 48.21 km². In terms of the severe drought category, 2009 with an area of 433.71 km² experienced the largest area and 2007 with an area of 45.78 km² experienced the lowest amount of drought. According to the maps obtained from the TCI, a very severe drought situation is observed in the southern and southwestern parts, especially in 2016 and 2018. It is also consistent with the results of the different intervals of the SPI in 2016 and 2018. They are considered as the years in which the highest amount of drought occurred. In addition, the year 2013 has less drought than other years in all three ranges of the SPI. From the comparison of the average SPI values for the studied years with the values obtained from the two indices, TCI and VCI obtained from satellite imagery, the highest amount of correlation coefficient between TCI and six-month SPI was equal to 0.65, which indicates that the TCI satellite index is the optimal index to indicate the drought situation in Marvdasht city.
 
Conclusion
According to the maps obtained from the NDVI, the studied area has normal and semi-dense vegetation density, which is scattered in all its different areas, so it has more density in the central and northwestern parts. The results of the VCI for the studied area, in none of the years, has not faced very severe drought. In terms of medium aridity, they have experienced the highest amount of drought in 2010 with an area of 62.98 km² and after that in 2019 with an area of 50.04 km². In contrast, the lowest drought in this layer was in 2002 with an area of 5.09 km². According to the maps showing the drought condition of VCI, the studied area has almost the same distribution pattern in all areas and except for a small part of the southern part of the area which has a medium drought condition, the other areas have a mild drought condition and are not dry in most areas.