A modeling approach to evaluate aircraft sampling strategies for estimation of terrestrial CO2 fluxes

Several aircraft sampling campaigns are being planned to estimate terrestrial fluxes of CO2 as a part of efforts to determine sources and sinks of CO2 in global and regional scales. These include, e.g., the COBRA project in USA, and the LBA project in Amazonia.

The modeling framework developed to help designing these aircraft campaigns is based on influence functions (footprints) calculated for both tracer concentration and vertical flux measurements. The influence function characterizes atmospheric transport from the point of view of a receptor. It provides spatial and temporal information on potential contributions from different source areas into measurement at a given receptor. Alternative sampling strategies are evaluated and compared with the aid of Bayesian inversion technique. A reduction of uncertainty in estimation of surface emissions is used as criteria in this evaluation.

The CO2 flux from the land surface shows a strong diurnal cycle related to the uptake of CO2 by photosynthesizing plants and the release of CO2 by microbial decomposition in the soil. The presentation will focus on the problem of designing aircraft sampling and tower measurements to correctly estimate the net flux of CO2. In particular, we are trying to determine the optimal time of the day for collecting air samples. The influence functions are derived for this purpose from a Lagrangian particle model linked to CSU RAMS (Regional Atmospheric Modeling System). The RAMS is used in different configurations: a mesoscale nested grid version for modeling of real summer episodes in Wisconsin, a 1-D version for an idealized study of multiday evolution of the boundary layer, and a LES configuration for selected regimes of the boundary layer.

The suggested modeling framework for evaluating atmospheric sampling strategies is general and can be applied to another problems as well.