Satellite-derived NDVI and climate projections for sustainable pasture biomass management: a case study in the Madrid region, Spain

Grasslands are among the world's most vital ice-free ecosystems, providing critical services such as carbon sequestration, biodiversity support, and pollination. However, these systems face increasing pressure from climate change, characterized by rising temperatures and altered precipitation patterns. Predictive models that integrate historical data with future climate scenarios are essential for anticipating shifts in biomass dynamics and developing adaptive management strategies to preserve the ecological and economic functions of grasslands. This thesis, developed in collaboration with the CEIGRAM Research Center in Madrid (central Spain), evaluates the efficacy of satellite-derived vegetation indices, specifically the Normalized Difference Vegetation Index (NDVI), as reliable proxy for pasture biomass production under climate change. It addresses a key limitation in current agricultural insurance models, which often rely on long-term NDVI averages that may not accurately capture real-time forage availability in a changing climate. Through a combined analysis of satellite imagery and field measurements, this research performs a statistical comparison between NDVI trends and directly measured biomass growth. The study is grounded in a mathematical model that integrates soil properties and climatic data to simulate pasture biomass. Field campaigns provide crucial data for model calibration, ensuring that spectral indices robustly reflect actual on-the-ground biomass. The research was carried out across three areas of the Community of Madrid, Spain, each chosen for their distinctive climatic and edaphic features; these locations encompass a gradient of environmental conditions, ranging from humid mountainous zones to semi-arid agricultural landscapes. The findings aim to determine if incorporating direct biomass estimation can enhance the precision of pasture monitoring systems; as shown in the chapter 4 a correlation between NDVI and pasture biomass confirms that satellite data can be effectively and scientifically employed as a reliable indicator for monitoring the status, productivity, and health of pastoral ecosystems in a continuous and large-scale manner. This research contributes to optimizing remote sensing tools for sustainable rangeland management, with potential implications for improving agricultural insurance frameworks, informing policy, and supporting climate adaptation strategies.