Document Type

Thesis

Date of Award

2025

Degree Name

Master of Science (MS)

Department

Sustainability

First Advisor

Ranjeet Dr. John

Abstract

This study presents an integrated geospatial framework for understanding grassland productivity dynamics and livestock movement behavior across the Mongolian steppe. Mongolia’s grasslands constitute one of the world’s largest pastoral ecosystems, yet they face increasing vulnerability due to climate variability, overgrazing, and land degradation. To quantify vegetation change and grazing responses, this research combines multi-decadal remote sensing with high-frequency GPS-based livestock movement tracking. Using Landsat imagery from 1990 to 2023, Aboveground Biomass (AGB) and Canopy Cover (CC) were estimated through Random Forest regression models integrating key spectral vegetation indices (e.g., NDVI, OSAVI, CTVI, TNDVI, IPVI) and climatic predictors such as summer precipitation. The models achieved high predictive accuracy (AGB R² = 0.85, CC R² = 0.84), revealing a long-term decline in biomass and canopy cover across southern and western Mongolia, while the northern Meadow Steppe maintained higher productivity. Temporal trend analysis using the Mann–Kendall test and Sen’s slope indicated significant spatial heterogeneity, with 2023 anomalies highlighting both degradation hotspots and resilient zones. Complementing these landscape-scale findings, GPS collar data collected at 5–15-minute intervals from multiple small livestock herds were analyzed to assess movement speed, altitude use, and directional behavior across seasons and steppe types. Results showed distinct seasonal and spatial contrasts, with herds in the Meadow Steppe exhibiting higher speeds and greater altitude variability during spring and summer, while those in the Typical Steppe demonstrated more linear and constrained movement patterns. By integrating satellite-based vegetation monitoring with fine-scale behavioral data, this research offers a comprehensive understanding of how environmental variability shapes livestock behavior, providing actionable insights for sustainable rangeland management and adaptive grazing strategies in the context of a changing climate.

Subject Categories

Sustainability

Keywords

Aboveground biomass Grassland Livestock behaviour Remote Sensing

Number of Pages

70

Publisher

University of South Dakota

Available for download on Sunday, July 12, 2026

Share

COinS