We develop and evaluate a method for estimating O₃ deposition and stomatal O₃ uptake at FLUXNET tower sites. The method combines standard micrometeorological measurements that are made with the eddy covariance method with a gridded dataset of surface O₃ concentration. We evaluate the method at 3 sites where O₃ fluxes were measured. This comparison shows that our method reproduces 69-92% of the variability with modest bias. The remaining unexplained variance is due to stomatal conductance filtering methods, non-stomatal parameterizations, representation error in the gridded O₃ data and measurement error in .

Using our O₃ deposition method, we calculate the stomatal O₃ uptake at 41 FLUXNET sites in the US and 55 FLUXNET sites in Europe. The spatial patterns in stomatal O₃ uptake are driven by the gradients in humidity, which promote stomatal conductance rather than spatial patterns in O₃ concentrations alone. Factors such as land cover, land management, and plant functional types are also important underlying factors in the spatial variability in the stomatal O₃uptake throughout the FLUXNET network.

With the synthetic O₃ flux data we examine the relationship with common concentration-based metrics for plant injuries due to O₃, such as mean concentration, AOT40 and W126. We show that concentration-based metrics are poor predictors of O₃ flux (R² ≤ 0.11), when applied across different biomes, but if they must be used, they should be limited to cases where VPD < 1.5 kPa (R² ≤ 0.28).

The synthetic O₃ flux data enable evaluation of atmospheric chemistry models at many additional sites.