Spatiotemporal drought analysis and future risk assessment using multi-index remote sensing approach and hybrid trend-based prediction modeling


Polat A. B., Alumert E., Akçay Ö.

ENVIRONMENTAL MONITORING AND ASSESSMENT AN INTERNATIONAL JOURNAL DEVOTED TO PROGRESS IN THE USE OF MONITORING DATA IN ASSESSING ENVIRONMENTAL RISKS TO MAN AND THE ENVIRONMENT, cilt.1, sa.198, ss.1-37, 2026 (SCI-Expanded, Scopus) identifier

Özet

This study presents a comprehensive spatiotemporal drought assessment for Çanakkale province, Türkiye, utilizing multi-index remote sensing approaches over a 20-year period (2005–2024) coupled with predictive risk modeling for 2025–2027. Four key environmental parameters were derived through the Google Earth Engine platform: Normalized Difference Vegetation Index (NDVI), Land Surface Temperature (LST), Standardized Precipitation Index (SPI), and Soil Moisture Condition Index (SMCI). Multiple satellite data sources were integrated, including Landsat 7 ETM + , MODIS/MOD11A1, CHIRPS precipitation dataset, and TerraClimate hydrological data. The retrospective analysis revealed significant climatic variability characterized by inter-annual LST fluctuations, progressive NDVI enhancement toward 2024, and pronounced negative trends in both SPI and SMCI indices during recent years. Particularly, SMCI reached − 1.14 in 2023, indicating severe soil moisture deficit. Spatial heterogeneity was evident across the province, with differential vegetation dynamics and precipitation patterns between coastal and interior regions. A Principal Component Analysis-based integrated drought index was developed, explaining 68.7% of total variance, providing comprehensive drought characterization beyond univariate approaches. A hybrid trend-based forecasting framework incorporating seasonal decomposition, climatological constraints, and stochastic variability was implemented. Model validation demonstrated robust performance for LST (R2 = 0.85) and NDVI (R2 = 0.88), while SPI and SMCI exhibited challenges inherent to normalized indices with small-magnitude variations. Prospective projections indicate systematic elevation in composite drought risk from 2.58 (2025) to 2.67 (2026–2027), representing a 3.5% increase and persistent moderate-to-high drought vulnerability. These findings provide critical insights for regional water resource management, agricultural planning, and climate adaptation strategies in Mediterranean ecosystems facing intensifying drought pressures.