Integrating Sentinel-2 information into a growth model for assessing Alpine grassland dynamics under climate change

In the Alpine region, as in other mountainous areas of the World, grasslands dominate the agricultural landscape, providing key ecosystem services to human societies. Ongoing climate change is already altering their seasonal growth patterns. Monitoring these responses and predicting future shifts is therefore of great importance for identifying suitable adaptation measures.

Recent studies have demonstrated the potential of remote sensing for the observation of grassland dynamics. Satellite data, however, can also be used to inform grassland growth models, which in turn are key tools for translating climate change scenarios into manageable information.

In this contribution we discuss the integration of information derived from Sentinel-2 data into a mechanistic model of grass growth that has been validated for low-altitude sites but never systematically applied to grasslands at high Alpine locations. We use satellite inferred growing season start and snow cover information to calibrate and validate the model across the region of the Swiss National Park (Grisons, Switzerland), a biodiversity-rich ecosystem encompassing dry and wet pastures, wetlands and shrubs.

The thus established model is then employed in conjunction with the national climate change scenarios for Switzerland to explore possible responses of alpine grasslands to mid-century climate change under the assumption of a business-as-usual emission pathway. In these simulations we account both for the effects of altered temperature and precipitation patterns as well as for the effects of elevated CO₂ concentrations.

Contributing to the activities of the Swiss National Centre for Climate Services, our work shows how remote sensing products coupled with mechanistic models can provide advanced predictive capabilities for developing scientific baselines needed to underpin climate change adaptation.