| Abstrakt: |
Background and Objectives: Natural vegetation cover is a crucial component of ecosystem change models and conservation efforts. The current research focuses on examining the spatial-temporal correlation of the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI) with meteorological indicators such as precipitation (P) and the Multivariate Standardized Precipitation Index (MSPI); biological indicators such as evapotranspiration (ET) and soil moisture (SM); and soil indicators, specifically land surface temperature (LST), using satellite imagery from Google Earth Engine over the period of 2000-2021 in Hormozgan Province. Materials and Methods: To evaluate the correlation between vegetation indices and climatic and biological variables, Google Earth Engine imagery was utilized. NDVI and EVI indices, evapotranspiration, and land surface temperature were extracted from 16-day Terra satellite images, and soil moisture (0-10 cm) from Soil Maps (Open Land Map) within the time frame of 01.01.2000 to 01.01.2022, using programming functions on a monthly basis. Precipitation and the MSPI drought index were obtained from a 22-year statistical dataset of weather stations. Using the Kriging spatial analysis method in ArcGIS, a precipitation spatial map was created. The PCA method was employed to calculate the drought index in different locations using SPSS software. Anomalies of vegetation indices were extracted to better understand their temporal changes over two decades. Correlation and trend analysis of parameters were conducted using the Spearman and Kendall methods. Results: The Spearman correlation coefficient and the Kendall rank correlation coefficient between the NDVI and EVI indices were found to be significant at 0.84 and 0.67, respectively, at a 99% confidence level. The temporal distribution of vegetation indices showed that their annual and monthly variations corresponded to the amount of autumn cultivation. In the warm and dry climate of the province, most of the precipitation occurs from December to February. Consequently, the indices had their highest values during December, January, and February, coinciding with the agricultural planting and growth season. Results revealed that the spatial variation of vegetation indices was in line with regions with higher precipitation, riverbanks, agricultural lands, and orchards. A significant inverse relationship was observed between vegetation cover indices and land surface temperature, indicating an increase in vegetation cover with decreasing temperature, and vice versa. The suitable overlap of precipitation zones and the MSPI drought index with the spatial distribution of vegetation indices, land surface temperature, and soil moisture indicated the determining role of precipitation in dry and semiarid regions. Anomalies of indices indicated a weak positive correlation during the 1378-1389 period due to frequent and severe droughts. In the second decade (2011-2022), a strong and significant correlation was observed due to a decrease in drought intensity. According to the MSPI drought index, western, central, and some eastern parts of the province with the highest levels of drought and low precipitation had the lowest vegetation index values. Conclusion: The extensive area of the region has led to a greater concentration of vegetation indices in certain areas. The highest values of NDVI and EVI are found in riverbank areas and water sources, which coincide with the agricultural centers in the plains of Rudan, Minab, Shamir, Takht, and some parts of Hajjiabad, which are geographically located in the northeastern and northern regions. An increase in any of the vegetation indices indicates an increase in their extent and abundance. However, a decrease in any of these indices results from the heterogeneous relationship between vegetation and climatic, soil, and land use factors in the region. The dynamic trend of vegetation cover and its relationship with other factors should be studied in the long term to establish a pattern. This study was conducted to the best of our abilities and with the available dataset. Long-term studies can be conducted by extracting empirical variables at the regional level and considering the impact of population growth on the region. The findings of this study can contribute to future research related to vegetation cover and its relationship with other factors, as well as nature conservation and resource allocation. [ABSTRACT FROM AUTHOR] |
