Monitoring cropland daily carbon dioxide exchange at field scales with Sentinel-2 satellite imagery

This is the authors' abstract. We don't add key points for this paper.

Improving the accuracy of monitoring cropland CO2 exchange at heterogeneous spatial scales is of high importance to limit spatial and temporal uncertainty of terrestrial carbon (C) dynamic estimates. A combination of field scale eddy covariance (EC) CO2 flux data and spatially matched Sentinel-2 derived vegetation indices (VIs) was tested as an approach to upscale agricultural C flux. The ability of different VIs to estimate daily net ecosystem exchange (NEE) and gross primary productivity (GPP) based on linear regression models was assessed. Most VIs showed high (>0.9) and statistically significant (p<0.001) correlations with GPP and NEE although some VIs deviated from the seasonal pattern of CO2 exchange. In contrast, correlations between ecosystem respiration (Reco) and VIs were weak and statistically not significant, and no attempt was made to estimate Reco from VIs. Linear regression models explained generally more than 80 % and 70 % of the variability of NEE and GPP, respectively, with high variability amongst the individual VIs. The performance in estimating daily C fluxes varied amongst VIs depending on C flux component (NEE or GPP) and observation period. RMSE values ranged from 1.35 g C m-2 d-1 using the Green Normilized Difference Vegetation Index (GNDVI) for NEE to 5 g C m-2 d-1 using the Simple Ratio (SR) for GPP. This equated to an underestimated net C uptake of only 41 g C m-2 (18 %) and an overestimation of gross C uptake of 854 g C m-2 (73 %). Differences between measured and estimated C fluxes were mainly explained by the diversion of the C flux and VI signal during winter, when C uptake stayed low while VI values indicated an increased C uptake due to relatively high crop leaf area. Overall, results exhibited similar error margins as mechanistic crop models. Thus, they indicated suitability and developability of the proposed approach to monitor cropland C exchange with satellite derived VIs.