Effects of Experimental Warming On Heterotrophic Soil Respiration In A Cultivated Andisol In Japan: First Two-Year Results.

A large proportion of the mitigation potential in agricultural sector arises from soil carbon sequestration through improved management practices. However, some of the soil carbon may be vulnerable to loss upon future global warming. Uncertainty about the complex biological and ecological processes involved in soil organic carbon dynamics, especially the sensitivities of labile and recalcitrant carbon to rising temperature, currently limits our ability to evaluate the long-term effectiveness of mitigation practices. In order to study the effects of warming on soil carbon dynamics under Asia monsoon climate and soil properties, we designed a field soil warming experiment for an Andisol, the dominant soil type of agricultural lands in Japan. The aim of this study was to assess effects of warming on the decomposition of soil organic carbon using the first two-year results of the soil warming experiment.

    Six plots, each 2×2 m² with three as heated plots and the other as non-heated control plots, were established in a wheat-soybean double cropping field in the winter of 2007. For each heated plot we used two infrared lamps to warm the soil from 150-cm above the ground, and established a feedback ON/OFF control system to keep the soil temperature at 5 cm depth in the heated plots 2ºC higher than the non-heated control plot. Each plot established two chambers to determine the heterotrophic soil respiration (Rh) using a newly-developed automatic opening and closing chamber system (AOCC). The soil warming started at the end of July in 2008 through the cropping seasons.

    Raising the soil temperature by 2ºC in the heated plots appeared to increase Rh during the cool seasons (winter to early summer) by approximately 10% compared to the non-heated control plots. In contrast, during the warm seasons (summer to Autumn), soil warming induced the decreased Rh in the heated plots; about 17% in 2008 (the extremely hot and dry summer) and 8.5% in 2009 lower relative to the control plots. These results indicate that Rh was stimulated by soil warming during the lower temperature regime (below 20ºC) than in the higher temperature regime (above 20ºC). Over the two-year soil warming treatment (compiling data from July 2008 to October 2009), different types of relationship between Rh and soil temperature (Ts) has been tested. The relationship between Rh and Ts was best described with an exponential function: Rh(Ts) = aexp[bTs]. The heated plots showed a lower value of apparent Q₁₀ (2.62) than that of the non-heated control plots (2.92). Lower apparent temperature sensitivity of Rh in the heated plots appeared to be linked to soil moisture. We found a significant correlation between normalized Rh (the ration of measured Rh to predicted Rh by Rh(Ts)) and soil moisture using a second-order polynomial function. That is, normalized Rh was suppressed when soil moisture was extremely low or high. Thus, the extremely low soil moisture (volume soil water content around 0.3) driven by the +2ºC heating during summer likely led to the reduction in Rh. Overall, soil warming resulted in a lower annual Rh compared to the non-warming treatment (478 and 535 g C m^-2 yr^-1 for heated and non-heated plots, respectively). Our results suggest that 2ºC temperature raise can have drastically different impact on soil organic matter decomposition between warm and cold season most likely through its effect on soil moisture regime, highlighting the importance of accounting for temperature vs. moisture interaction in the models when predicting soil carbon dynamic under climate warming scenario.