Regional estimates of net ecosystem-atmosphere exchange of carbon dioxide over a heterogeneous ecosystem

The net ecosystem-atmosphere exchange of CO2 (NEE) is estimated over a heterogeneous ecosystem in the 40x40km2 region centered at the WLEF tall tower. Flux aggregation and atmospheric boundary layer (ABL) budget methods are used. CO2 fluxes measured at three levels of the tower are decomposed in the growing season to infer the CO 2 fluxes for six stand types in the tower footprint area. The optimal solutions are estimated for the parameters in two ecosystem models for each stand type. Differences in the parameters and fluxes among stand types are consistent with general expectations. The inferred fluxes are compared to observations from two stand-level eddy-covariance flux towers, implying that the six stand classification scheme is insufficient to capture all variability in stand characteristics relevant to CO2 exchange. Significant uncertainty in aggregated fluxes exists due to the classification schemes used. A new approximate analytical model is introduced to predict the footprint for flux measurements in the lower part of the mixed layer. The ABL budget method is used to infer regional NEE by using the vertical profile of CO 2 mixing ratios measured at the tall tower. In the nocturnal boundary layer, the cases under very stable or nearly-neutral conditions are screened out due to systematic errors. Regional NEE is bounded by the current tower measurements. The range of the effects of horizontal advection of CO 2 on daytime NEE estimates is estimated by the measurements of the spatial distribution of water vapor. Regional NEE estimates from the two methods are compared with each other and with those reported in the literature. In the dormant season, differences among all estimates are small and NEE measured at the WLEF tower might be an acceptable approximation of NEE in the region. This is not true in the growing season. The cumulative regional NEE values are -103+/-50 and -175+/-60gC m-2 in 2000 and 2003, respectively, suggesting that the region is a net sink of CO2. Those values are significantly different from those measured at individual towers. It is inappropriate simply to extend fluxes measured at individual towers to a larger region.