Assessing Rodent-Induced Ecological Disturbance in Natural Grasslands Using Multi-Source Spatial Data

Area of Science:

• Ecology
• Remote Sensing
• Environmental Monitoring

Background:

• High rodent densities degrade grasslands via foraging and burrowing.
• Traditional ground surveys are insufficient for large-scale, dynamic rodent damage assessment.
• Ecological imbalance results from intense small mammal activity.

Purpose of the Study:

• To evaluate rodent damage severity in alpine meadows and typical steppe.
• To develop an integrated framework combining ground, UAV, and satellite data for damage assessment.
• To model and map rodent damage at a regional scale.

Main Methods:

• Extracted ecological parameters (biomass, vegetation cover, burrow density, etc.) from 36 plots per grassland type.
• Constructed a plot-scale Rodent Damage Index (RDI).
• Linked RDI with satellite-derived Remote Sensing Ecological Index (RSEI) using separate linear regression models for each grassland type.

Main Results:

• The alpine meadow model showed superior fit and predictive performance (LOOCV R2 = 0.729) compared to the typical steppe model (LOOCV R2 = 0.478).
• Model performance significantly differs between grassland types.
• Integrated multi-source, cross-scale spatial data proved effective for rodent damage assessment.

Conclusions:

• Integrating ground, UAV, and satellite data offers an effective approach for assessing rodent damage.
• Rodent damage assessment requires a grassland-type-specific approach for accuracy.
• Findings support targeted rodent damage management planning.