Partitioning the net ecosystem carbon balance of a semiarid steppe into biological and geological components

Recent studies have highlighted the need to consider geological carbon sources when estimating the net ecosystem carbon balance (NECB) of terrestrial ecosystems located in areas potentially affected by geofluid circulation. We propose a new methodology using physical parameters of the atmospheric boundary layer to quantify the CO2 coming from deep ground origin in a steppe ecosystem located in the SE of Spain. Then, we compared published NECB estimates at the site with seasonal patterns of soil CO2 efflux and biological activity measured by satellite images over a 2-year period (2007/2008). The alpha grass ecosystem was a net carbon source (93.8 and 145.1 g C m(-2) year(-1), in 2007 and 2008, respectively), particularly as a result of large amounts of carbon released over the dry period that were not related to biological activity. While the highest ecosystem CO2 emission rates were measured over the dry period (reaching up to 15 mu mol m(-2) s(-1)), soil CO2 efflux rates (ca. 0.5 mu mol m(-2) s(-1)) and plant productivity were minimal during this period. After using a linear relationship between NECB and wind speed for different stability conditions and wind sectors, we estimated the geological flux F (GEO) (217.9 and 244.0 g C m(-2) in 2007 and 2008, respectively) and subtracted it from the NECB to obtain the biological flux F (BIO) (-124.0 and -98.9 g C m(-2) in 2007 and 2008, respectively). We then partitioned F (BIO) into gross primary productivity and ecosystem respiration and proved that, after removing F (GEO), ecosystem respiration and soil CO2 efflux followed similar seasonal patterns. The annual contribution of the geological component to NECB was 49.6 and 46.7 % for the year 2007 and 2008, respectively. Therefore, it is clear that geological carbon sources should be quantified in those ecosystems located in areas with potential natural emission of geological gases to the surface.