Validating the Carnegie-Ames-Stanford Approach for remote sensing of perennial grass net primary production

Summary

This study improved the Carnegie-Ames-Stanford Approach (CASA) model for estimating grassland net primary productivity by incorporating environmental constraints derived from multi-year Danish field experiments. By integrating maximum air temperature, vapour pressure deficit, cloudiness, and water stress into the model, the researchers reduced prediction error by 8–34% at the seasonal scale and 9% at the daily scale. The work establishes practical RUEmax values (1.9–3.1 g C MJ⁻¹) for common perennial grass species and demonstrates the feasibility of using handheld and UAV-based multispectral reflectance for grassland productivity monitoring.

UK applicability

The methodology and improved CASA model are directly applicable to UK grassland systems, particularly for pasture management and biomass monitoring, though UK climatic conditions and soil types (especially in wetter regions) may require recalibration of environmental constraint parameters. The optimal temperature (16.5 °C) and stress factors identified in Denmark provide a reference framework that could be validated across UK growing conditions.

Key measures

Net primary productivity (NPP); radiation use efficiency (RUEmax); intercepted photosynthetically active radiation (Ipar); normalised root mean square error (nRMSE); daily and seasonal NPP estimates; shoot and root biomass; CO₂ flux measurements

Outcomes reported

The study validated and improved the CASA model for estimating grassland net primary productivity using multispectral remote sensing at field and UAV scales, incorporating environmental constraints such as temperature and water stress. Seasonal radiation use efficiency values were derived for ryegrass, grass-legume mixtures, tall fescue, and festulolium under Danish sandy loam conditions.

Topic tags