Investigating the Sustainability of Vegetation Change Trends Using Remote Sensing (Case Study: Northern River Basin of Afghanistan)

Drought is a phenomenon that occurs at different times unpredictably with different intensity and has severe effects on human society and ecosystems. In this study, Moderate Resolution Imaging Spectroradiometer (MODIS) and PERSIANN Dynamic Infrared Rain Rate (PDIR Now) sensors were applied to examine the drought effects on vegetation in Northern River Basin of Afghanistan from 2001 to 2020. Therefore, MODIS data include Enhanced Vegetation Index (EVI), Vegetation Condition Index (VCI) with a time period of 16 days and a spatial resolution of 250 meters, Land Surface Temperature (LST) with spatial resolution of 1 km and period of 8 days, and monthly precipitation data with a resolution of (4*4) km. The relationship between drought and vegetation in spring was investigated using time series analysis, regression analysis and calculation of anomalies.  The results showed that the average vegetation coverage in the whole statistical period was 45.21%. In this study, vegetation area in 2001, 2008 and 2011 have reached the lowest rate (9.9%, 9.9% and 19.3%), respectively. According to VCI, 83.5%, 81.39% and 74.9% of the basin in these years are under drought conditions, respectively. Rainfall data confirm that these years have had the lowest rainfall 96.7, 133 and 117 mm, respectively. The years 2003, 2009 and 2010 with the highest vegetation in this season were recorded mainly due to the lower LST and higher rainfall then their average period. The correlation between EVI and LST is (r=-0.87; p<0.05), EVI and Precipitation (r=0.60; p<0.05). However, in Northern River Basin of Afghanistan, LST and rainfall must be considered together to determine the relationship between drought and vegetation dynamics properly.
 Extended Abstract
1-Introduction
Investigating the sustainability of vegetation change is one of the most important issue in vegetation management and control for sustainable development. Lack of vegetation on the ground is one of the main causes of soil destruction by rain and so on. Changes in vegetation area have different factors, such as the use of forest trees for fuel, livestock pressure on pastures, forest fire, short-term and long-term droughts, all of those which reduce vegetation area and vegetation loss. Drought is a natural phenomenon in all climates that has significant adverse effects on human life. Remote sensing drought indices are effective and appropriate for spatial and temporal monitoring of drought conditions. Drought intensity, impact and duration control indices include Normalized Difference Vegetation Index (NDVI), Land Surface Temperature (LST), Vegetation Condition Index (VCI), which are based on remote sensing drought indices. The study of vegetation dynamics and their relationship with drought in northern watershed of Afghanistan is important for the ecological sustainability of this region, because vegetation has a range of characteristics such as immobility, relatively high growth rate.
Increasing vegetation in an area reduces carbon dioxide, air lead, soil moisture, soil-water protection and reduces flooding, which is associated with ecological sustainability; Therefore, in the case of instability in an environment, it will face symptoms such as desertification, soil erosion, climate change and other cases that will have devastating economic, social and climatic consequences. Therefore, in this study, EVI and VCI with SPI and LST indices were used to identify drought and changes in vegetation area. Moreover, the effect of vegetation and temperature in creating dry and wet periods and changes in rainfall, as the only reasons for droughts and wet years in the northern basin, were compared. For this purpose, time series analysis of EVI, VCI and LST indices of MODIS, SPI index and precipitation data from PDIR Now data for the years 2001 to 2020 have been used.
2-Materials and Methods
Northern Watershed of Afghanistan is located between longitudes 64˚- 68˚ 22′ East and latitudes 34˚ 28′- 37˚ 23′ North, and includes provinces of Samangan, Balkh, Jawzjan, Faryab and Sarepol. Covering approximately 70,901 km², it contains four rivers including Khulm, Balkhab, Shirintagab and Sarepol. In Northern Watershed of Afghanistan (460 EVI, 920 LST) the images of MODIS product and 240 images of PDIR Now data between 2001-2020 were applied. Using the average method in Arc GIS program, the images were converted into annual and seasonal. The output of all value of seasonal and annual pixels of EVI, VCI, LST, SPI indices and PDIR precipitation data were extracted in Microsoft Excel. Analysis, classification, correlation between indicators, the amount of changes in precipitation and surface temperature, calculation of vegetation area and dry area were performed according to the indicators. Wet and dry years were identified according to the mentioned indicators and its maps were prepared in ArcGIS program.
3- Results and Discussion
In northern watershed of Afghanistan, the main vegetation growing season begins on January 17^th, April 23^rd in the region as the peak of vegetation, which is the highest vegetation area in the region with 56.4%. From the first of May, the vegetation gradually decreases and this trend continues until the end of November. The first 15 days of December have the lowest vegetation area with 0.3% in the region.
Spring vegetation, compared with annual and other seasons, showed more changes in the area of vegetation. Northern part of Afghanistan has no forest areas and is mostly rangeland, most of the area is weak vegetation, which according to the EVI, the area of all vegetation classes except dense cover did not change. Dense vegetation with significance correlation (r = 0.57, P < 0.05) showed a relative increase which is from 2013 to 2020, due to the expansion of population and urbanization in Balkh, Samangan, Jawzjan, Sarpol and Faryab provinces including the study area and has created green spaces such as planting a variety of trees, creating parks and gardens. The area of vegetation in terms of EVI in 2001, 2008, and 2011, and in 2003, 2009, and 2010, is 9.8%, 9.9%,19.3% and 65.7%, 63.7%, 63.1%, respectively. In terms of VCI, 83.5%, 81.3%, and 74.9% of the basin in these years are under drought conditions, respectively. The central parts up to the Amu Darya River in the North of the basin, which is mostly flat areas consisting of agricultural lands and rangelands, have the highest dry area in terms of VCI, because in the years of drought, the average spring temperature of the same year compared to wet years were more with 36.9 , 35.4 and 34.3 °C.
In Northern Watershed of Afghanistan, LST and precipitation must be considered together to properly determine the relationship between drought and vegetation dynamics. Moreover, rivers had an effect on decrease and increase of vegetation in region.
4- Conclusion
The results reveal that the VCI calculated from EVI, with rainfall and LST data could be useful in assessing drought in Northern watershed of Afghanistan. The EVI of spring does not increase from 0.15 to 0.5 in study area and the total area of vegetation in the statistical period. Only the EVI>= 0.5 class has increased in the study area. Regression between EVI and dry area (r = -0.973, P < 0.05) and EVI with non-dry area (r = 0.973, P < 0.05) was the same. Correlation between all EVI classes with severe and moderate drought conditions has negative and significant positive relationships have been obtained among all EVI classes with extremely wet, very wet, moderate and normal wet conditions. According to both EVI and VCI, the least vegetation coverage is found in the years 2001, 2008 and 2011, while the most vegetation coverage is found in the years 2003,2009 and 2010. In this study, LST has a negative relationship with EVI and precipitation, and EVI has a positive relationship with precipitation. During the 20-year period, precipitation did not increase until 2009, but after that year, precipitation has increased slightly by 2020. In this statistical period, LST did not change.