A multi-temporal scale framework for comprehensive quantification and attribution of anthropogenic impacts on runoff

Area of Science:

• Hydrology and Water Resource Management
• Environmental Science
• Geospatial Analysis

Background:

• Runoff variation attribution is crucial for water resource conservation and management.
• Previous studies often used single-scale approaches and overlooked multicollinearity.
• Attribution analysis is limited by single-factor assessments.

Purpose of the Study:

• To establish a methodological framework for runoff reconstruction, quantification, and contribution rate analysis.
• To identify optimal temporal scales and explanatory variables for runoff modeling.
• To integrate multi-source data for a comprehensive attribution of anthropogenic drivers.

Main Methods:

• Aggregation of hydro-meteorological variables across multiple time scales.
• Spearman correlation and variance inflation factor (VIF) analysis for variable optimization.
• Runoff reconstruction using Random Forest Regression Model (RFRM) and Soil and Water Assessment Tool (SWAT).
• Integration of remote sensing and statistical data for anthropogenic driver identification.

Main Results:

• Precipitation identified as the dominant meteorological driver (Spearman's p=0.90 at seasonal scale).
• Monthly and bimonthly scales determined as optimal for modeling.
• RFRM and SWAT showed complementary strengths, with SWAT excelling in high-flow simulation.
• Multi-source data integration revealed the dual role of human activities on runoff.

Conclusions:

• The developed framework offers methodological guidance for temporal scale and model selection in runoff attribution.
• The approach provides a transferable method for quantitative attribution of runoff variation.
• Findings support improved water resource conservation and management strategies.