Research on a Symbolic Regression-Based Model for Valuing China's Population-Ecosystem Service Demand

The quantification of ecosystem service demand value serves as a critical bridge connecting human well-being with ecological management. Addressing the current academic gap in valuation frameworks that precisely couple with supply classification systems and are difficult to integrate into Decision Support Systems (DSS), this study develops an ecosystem service demand analytical model. Based on ecological characteristics and administrative divisions, mainland China was divided into six management zones. Guided by Human Need Theory, a demand classification system was constructed. By integrating socio-economic big data with symbolic regression algorithms, we decoded the quantitative relationships between population scale and various demand values across regions, satisfying the requirements of DSS for rapid computation and real-time simulation. Results show that: (1) Spatial Distribution Characteristics: Within the population interval below 5 million, the demand values for various services in the Yellow River Basin Ecological Restoration Coordination Zone are higher than those in other regions under the same population base. (2) Evolutionary Patterns of Demand: The simulation curves reveal distinct environmental carrying capacity thresholds across all regions. Beyond these critical points, the marginal fulfillment costs of ecosystem services surge, driving a rapid upward trend in demand value. (3) Model Accuracy and Application: With the introduction of a time-factor correction, the average model error is controlled within 10%, and the accuracy is improved by 20%. This study establishes a classification and accounting framework that balances computational simplicity with realistic alignment, achieving multi-scale quantitative assessment of demand value and providing core algorithmic support for ecosystem service decision support systems.