B53A-0168:
Observing Mean Annual Mediterranean Maquis Ecosystem Respiration

Friday, 19 December 2014
Serena Marras1, Veronica Bellucco2, Simone Mereu1, Costantino Sirca1 and Donatella Spano2, (1)University of Sassari; CMCC, Sassari, Italy, (2)University of Sassari, Sassari, Italy
Abstract:
In semi arid ecosystems, extremely low Soil Water Content (SWC) values may limit ecosystem respiration (Reco) to the point of hiding the typical exponential response of respiration to temperature. This work is aimed to understand and model the Reco of an evergreen Mediterranean maquis ecosystem and to estimate the contribution of soil CO2 efflux to Reco.

The selected site is located in the center of the Mediterranean sea in Sardinia (Italy). Mean annual precipitation is 588 mm and mean annual temperature is 15.9 °C. Vegetation cover is heterogeneous: 70% covered by shrubs and 30% of bare soil. Net Ecosystem Exchange (NEE) is monitored with an Eddy Covariance (EC) tower since April 2004. Soil collars were placed underneath the dominant species (Juniperus phoenicea and Pistacia lentiscus) and over the bare soil. Soil CO2 efflux was measured once a month since April 2012. Soil temperature and SWC were monitored continuously at 5 cm depth in 4 different positions close to the soil collars. Six years of EC measurements (2005–2010) and two years of soil CO2 efflux (2012–2013) measurements were analysed. Reco was estimated from the measured EC fluxes at night after filtering for adequate turbulence (u* > 1.5). Reco measurements were then binned into 1°C intervals and median values were first fitted using the Locally Estimated Scatterplot Smoothing (LOESS) method (to determine the dominant trend of the experimental curve) Reco shows an exponential increase with air and soil temperature, until SWC measured at 0.2 m depth remains above 19% vol. Secondly, the coefficients of the selected Lloyd and Taylor (1994) were estimated through the nonlinear least square (nls) method: Rref (ecosystem respiration rate at a reference temperature of 10 °C was equal to 1.65 μmol m-2 s-1 and E0 (activation energy parameter that determines the temperature sensitivity) was equal to 322.46.

In addition, bare and drier soils show a reduced response of measured CO2 efflux to increasing temperature than moist under canopy ones. Soil CO2 efflux measurements were binned in 1°C classes. Mean total site CO2 efflux was estimated scaling measurements for percentages of soil cover as 1 μmol m-2 s-1. The mean difference between Reco and CO2 efflux increases linearly with increasing temperatures (R2 = 92%) even if soil efflux remains on average about 38% of Reco.