Uzaktan algılama teknikleri kullanılarak Afganistan'da iklim değişikliğinin etkilerinin incelenmesi

Afghanistan has not been immune from significant changes in climate parameters affected by climate change in recent decades. The main purpose of this study is to use remote sensing methods to investigate the effect of climate change in Afghanistan. In this study, data from various sensors on the Google Earth Engine platform between 2000 and 2023 were used to investigate the climate parameters such as Normalized Difference Vegetation Index (NDVI), Vegetation Condition Index (VCI), and Temperature Condition Index (TCI). , vegetation health index (VHI), land surface temperature (LST), air temperature (Ta), precipitation, soil moisture, actual evapotranspiration (AET), potential evapotranspiration (PET), wind speed, and forest loss. These climate parameters were examined on a monthly, seasonal, and annual basis for each of the eight areas that make up Afghanistan. The analysis of noteworthy trends was conducted using the Mann-Kendall test and homogeneity test, while Pearson's correlation coefficient was utilised to investigate the associations among the parameters. In addition, the ARIMA model was used to predict trends for various climate parameters, with the exception of the forest loss parameter, for the years 2024 to 2035. R2, RMSE, and MAE metrics were applied to assess the model's predictions, with good outcomes. The results showed that NDVI increased while LST decreased in most regions, except in the north, where NDVI declined despite its overall high values. Precipitation trends varied, but soil moisture decreased consistently across all regions due to insufficient rainfall and stony soils. The western region experienced the highest wind speeds, impacting vegetation and soil moisture. Agricultural drought was most severe in the north and west, while deforestation declined from 2000 to 2023, except for a rise in 2023. Strong correlations among climatic parameters confirmed the reliability of the data. Lastly, seasonal maps were created for these parameters in the following years: 2000, 2012, and 2023.