Modelling vegetation phenology in alpine grasslands of North-East Italy integrating microclimatic and meteorological Drivers: A 23-Years Time-Series analysis

Grasslands in mountainous regions often exhibit unique variations in the amplitude and timing of their growing seasons in response to microclimatic and meteorological variability. This study investigates the environmental drivers of vegetation phenology in the Alpine grasslands of North-East Italy using a 23-year MODIS NDVI time series (2001-2023), harmonic regression, and Generalized Additive Models (GAMs). Results highlight a strong negative correlation between altitude and NDVI metrics, with higher elevations exhibiting lower vegetation cover, greater seasonal variability, and shorter growing seasons. Slope similarly constrains vegetation growth, while aspect influences microclimatic conditions but shows weaker overall effects. Temperature and solar radiation positively influence NDVI, enhancing vegetation vigor and extending the growing season, while precipitation plays a more limited role, with limited variability across the region. Wind exposure shortens the growing season, reducing NDVI values, while snowfall significantly delays and shortens vegetation growth, increasing seasonal variability. Soil properties, particularly water-holding capacity, pH, and texture, strongly influence vegetation dynamics, with silt-rich soils favoring longer and more stable growing periods. The developed GAMs demonstrated high predictive accuracy (R2 = 0.70-0.91), effectively estimating phenological transitions, including the start and end of the vegetative season, with an accuracy of approximately 11 days. These findings enhance our understanding of the complex interactions shaping alpine pasture ecosystems and provide valuable tools for sustainable management, conservation, and climate adaptation strategies.