In a decade-long soil warming experiment in a mid-latitude hardwood forest, we documented changes in soil carbon and nitrogen cycling in order to investigate the consequences of these changes for the climate system. Here we show that whereas soil warming accelerates soil organic matter decay and carbon dioxide fluxes to the atmosphere, this response is small and short-lived for a mid-latitude forest, because of the limited size of the labile soil carbon pool. We also show that warming increases the availability of mineral nitrogen to plants. Because plant growth in many mid-latitude forests is nitrogen-limited, warming has the potential to indirectly stimulate enough carbon storage in plants to at least compensate for the carbon losses from soils. Our results challenge assumptions made in some climate models that lead to projections of large long-term releases of soil carbon in response to warming of forest ecosystems.

We are conducting a field study to determine the long—term response of belowground processes to elevated soil temperatures in a mixed deciduous forest. We established 18 experimental plots and randomly assigned them to one of three treatments in six blocks. The treatments are: (1) heated plots in which the soil temperature is raised 5°C above ambient using buried heating cables; (2) disturbance control plots (cables but no heat); and (3) undisturbed control plots (no cables and no heat). In each plot we measured indexes of N availability, the concentration of N in soil solutions leaching below the rooting zone, and trace gas emissions (CO2, N2O, and CH4). In this paper we present results from the first 6 mo of this study. The daily average efflux of CO2 increased exponentially with increasing soil temperature and decreased linearly with increasing soil moisture. A linear regression of temperature and the natural logarithm of CO2 flux explained 92% of the variability. A linear regression of soil moisture and CO2 flux could explain only 44% of the variability. The relationship between soil temperature and CO2 flux is in good agreement with the Arrhenius equation. For these CO2 flux data, the activation energy was 63 kJ/mol and the Q10 was 2.5. The daily average uptake of CH4 increased linearly with increasing soil temperatures and decreased linearly with increasing soil moisture. Linear regression could explain 46% of the variability in the relationship between temperature and CH4 uptake and 49% of the variability in the relationship between soil moisture and CH4 uptake. We predicted the annual CO2 flux from our study site in 1991 using two empirical relationships: the relationship between air temperature and soil temperature, and the relationship between soil temperature and CO2 flux. We estimate that the annual CO2—C flux in 1991 was 712 g/m2 from unheated soil and 1250 g/m2 from heated soil. By elevating the soil temperature 5°C above ambient, we estimate that an additional carbon flux of 538 g°m—2°yr—1 was released from the soil as CO2.

SummaryÐCarbon dioxide and methane are important greenhouse gases whose exchange rates between soils and the atmosphere are controlled strongly by soil temperature and moisture. We made a laboratory investigation to quantify the relative importance of soil moisture and temperature on¯uxes of CO 2 and CH 4 between forest soils and the atmosphere. Forest¯oor and mineral soil material were collected from a mixed hardwood forest at the Harvard Forest Long-Term Ecological Research Site (MA) and were incubated in the laboratory under a range of moisture (air-dry to nearly saturated) and temperature conditions (5±258C). Carbon dioxide emissions increased exponentially with increasing temperature in forest¯oor material, with emissions reduced at the lowest and highest soil moisture contents. The forest¯oor Q 10 of 2.03 (from 15±258C) suggests that CO 2 emissions were controlled primarily by soil biological activity. Forest¯oor CO 2 emissions were predicted with a multiple polynomial regression model (r 2 =0.88) of temperature and moisture, but the ®t predicting mineral soil respiration was weaker (r 2 =0.59). Methane uptake was controlled strongly by soil moisture, with reduced¯uxes under conditions of very low or very high soil moisture contents. A multiple polynomial model accurately described CH 4 uptake by mineral soil material (r 2 =0.81), but only weakly (r 2 =0.45) predicted uptake by forest¯oor material. The mineral soil Q 10 of 1.11 for CH 4 uptake indicates that methane uptake is controlled primarily by physical processes. Our work suggests that inclusion of both moisture and temperature can improve predictions of soil CO 2 and CH 4 exchanges between soils and the atmosphere. Additionally, global change models need to consider interactions of temperature and moisture in evaluating e ects of global climate change on trace gas¯uxes. #