Evaluation of water stress modeling methods for estimating gross primary production in Mediterranean grasslands

Mediterranean grasslands of southwest Spain and Portugal are a crucial part of the largest agroforestry ecosystem in Europe, known as dehesa in Spain. This multi-use system is recognized to be a balanced combination of environmental and economic values. Grasslands contribute to both aspects, with a high diversity of plant species, providing essential feeding resources for extensive livestock, the primary economic activity in many of these areas. Water availability is the main limiting factor for plant growth in the region, and the production of the grasslands is closely linked to its continental Mediterranean climate, resulting in significant differences in biomass production throughout the seasons and between years. Optimizing grassland management and adapting it to the increase in droughts and extreme events described by climate projections is vital for preserving a healthy and productive ecosystem. Accurate and timely information on grassland productivity is needed at appropriate spatial and temporal scales to meet management requirements.

Light use efficiency (LUE) models link plant growth to incident solar radiation to estimate gross primary production (GPP) or aboveground biomass. The advancements and availability of remote sensing technology have led to a renewed interest in this approach, resulting in extensive research and numerous applications across various land uses, particularly at global or large scales. This study aims to improve the monitoring of grassland productivity by developing a model specifically designed to estimate GPP for this water-controlled and highly variable grassland ecosystem.

Several methods for modeling water stress in LUE models have been compared using a series of CO2 exchange measurements from eddy covariance systems at three sites in southern Spain over eight years. The opportunities and limitations of the different methods are evaluated, and a proposal is presented that effectively balances operativity and accuracy for monitoring grasslands at a high spatial and temporal resolution.