Diurnal and seasonal dynamics of regional CO2 drawdown at Harvard Forest: Integrating remote sensing and modeling perspectives

Abstract

The diel drawdown of CO₂ provides a direct measure of light use efficiency and of the capacity of terrestrial ecosystems to mitigate climate change, as these processes influence how much carbon is absorbed by vegetation through photosynthesis. This study investigates CO₂ drawdown in Northeastern U.S. forests by analyzing five years of column-averaged CO₂ measurements (XCO₂) from the EM27 instrument at the Harvard Forest and evaluating the performance of two vegetation models, including CarbonTracker and the Vegetation Photosynthesis and Respiration Model (VPRM), in representing CO₂ fluxes. EM27 XCO₂ data reveal clear seasonal patterns with peaks in late spring and lows in late summer, reflecting an annual increase of approximately 3 ppm. We model XCO₂ enhancements by convolving simulated footprints from the Stochastic Time-Inverted Lagrangian Transport (STILT) model with biological CO₂ fluxes from both vegetation models. Observed and modeled CO₂ drawdown during the daytime all peak in the warmer months. VPRM drawdown rates closely align with observed data, showing a slight underestimation of peak values (−0.17 ppm hr⁻¹ in July compared to −0.18 ppm hr⁻¹ observed) in the average annual variation of daily XCO₂ slopes. In contrast, CarbonTracker simulates weaker CO₂ drawdown. The study highlights stable interannual variability in CO₂ drawdown, with no indication of saturation in the ecosystems' CO₂ drawdown capacity. In the context of climate change, this work underscores the value of long-term monitoring and modeling of XCO₂ to track changes in CO₂ drawdown and to identify environmental stresses affecting terrestrial carbon sinks.