Long-term dynamics of remote sensing indicators to monitor the dynamism of ecosystems in arid and semi-arid areas: contributions to sustainable resource management

Drought is expected to increase in water bodies due to climate change. Monitoring long-term changes in wetlands is crucial for identifying fluctuations and conserving biodiversity. In this study, we assessed the long-term variability of remote sensing indicators in 25 watershed areas in Algeria known for their significant biodiversity. We employed two statistical methods, namely linear regression and the Mann–Kendall (MK) test, to capture long-term fluctuations by integrating data from various sources, including Modis and Landsat satellite data. A time-series dataset spanning 22 years was developed, consisting of the following indicators: normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), normalized difference water index (NDWI), normalized difference moisture index (NDMI), and land surface temperature (LST). We evaluated the relationships between these variables. The results indicated that NDVI exhibited a stronger temporal response compared to EVI, NDWI, and NDMI. Additionally, negative associations between NDVI and LST confirmed the impact of drought and plant stress on vegetation in the study areas (R2 = 0.109–R2 = 0.5701). The NDMI results pointed to water stress in the water bodies, showing a significant decreasing trend. The results from the MK trend analysis underscored the importance of NDVI and highlighted its strong association with EVI, NDWI, and NDMI. Understanding the dynamics of vegetation and water stress has become crucial for ecosystem forecasts.