Modeling from black to greenish: Carbon fluxes estimations in the first-year post-fire using geodatabases and remote sensing in a forest ecosystem in Portugal.

Despite extensive research on fire emissions, the contribution of post-fire CO2 emissions during ecosystem regeneration remains poorly understood. The lack of direct measurements of post-fire Net Ecosystem Exchange (NEE) limits the number of opportunities available to validate model outputs. This study aims to address this gap by evaluating the ability of DAYCENT biogeochemical model to estimate daily NEE for the first year following a wildfire, using online geodatabases and remote sensing data as the only input for the model, by comparing the results with eddy covariance (EC) observations. The model was tested in a pine forest in Portugal, equipped with an EC tower installed one month after a severe wildfire in 2017. To simulate the post-fire vegetation five different simulation were run, using different vegetation type. The analysis revealed that the savanna system simulation, composed by Loblolly pine (LOBL) and temperate C3 grasses (TMC3), aligned most closely with observed post-fire NEE, showing an R² of 0.37. However, the simulation tended to produce slightly higher emissions than observed, with an average difference of +0.76 g C m⁻² day⁻¹. This discrepancy suggests the real-world ecosystem captures more CO2 than the model predicts. The study highlights the potential of using remote sensing and online geodatabases for scalable and efficient post-fire C flux assessments, which is increasingly critical in the context of climate change and its impact on fire regimes. The findings represent a significant advancement in understanding and quantifying post-fire C dynamics.