Assessment of climate change impacts on rainfall erosivity (R) across Türkiye using panel-data regression
Soil and Tillage Research, cilt.264, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 264
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.still.2026.107371
- Dergi Adı: Soil and Tillage Research
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, BIOSIS, Compendex, Environment Index, Geobase, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO)
- Anahtar Kelimeler: Climate change, Panel data analysis, Precipitation variability, Rainfall erosivity, RUSLE-R, Türkiye
- Ankara Üniversitesi Adresli: Evet
Özet
Soil erosion by water is a dominant land-degradation process in Türkiye, primarily driven by rainfall erosivity (R₃₀-factor), the metric that captures the energy and intensity of precipitation. Understanding how rainfall erosivity will evolve under future climate conditions is crucial for anticipating soil erosion risks and designing sustainable land management strategies. This study analyzes temporal and spatial variations in R using a panel-data regression with annual precipitation and elevation predictors for 2004–2021, and projects future values for 2025–2099 under RCP 4.5 and RCP 8.5. Long-term rainfall records from 209 meteorological stations and dynamically downscaled projections from three CMIP5 models (GFDL-ESM2M, HadGEM2-ES and MPI-ESM-MR) under two different climate scenarios were used to characterize changes in rainfall energy and intensity. Climatic indices such as the Modified Fournier Index, Precipitation Concentration Index, and Seasonality Index, together with topographic attributes, were incorporated as explanatory variables in the panel model to capture both temporal trends and regional variability. Candidate panel specifications with log-transformed continuous variables were compared for the historical period, and a parsimonious projection model based on annual precipitation, elevation, and regional dummy variables was retained for future simulations. Statistical inference was based on Driscoll–Kraay robust standard errors to account for heteroskedasticity, serial correlation, and cross-sectional dependence. Although annual precipitation was identified as a statistically significant predictor of rainfall erosivity, future sub-daily rainfall intensity projections were not available and therefore could not be explicitly represented in the modeling framework. Consequently, annual precipitation was adopted as the principal climatic predictor, and the resulting projections should be interpreted as regional-scale annual erosivity scenario assessments rather than event-scale or high-precision station-level projections. Historical fit assessment against DEMMS-derived R₃₀ values (R² = 0.596) indicates moderate predictive performance for national-scale applications. Results indicate notable regional contrasts: rainfall erosivity is projected to increase markedly in Eastern Anatolia but decline moderately in Southeastern Anatolia and along the Black Sea coast. Findings underscore the urgency of region-specific adaptive soil and water conservation strategies integrated into Türkiye’s sustainable land-management and climate-adaptation frameworks. The approach demonstrates that empirical-statistical modeling offers a practical and data-efficient basis for erosion-risk assessment in data-limited contexts.