Soil CO₂Production And Efflux Under A Winter Wheat-Summer Maize Double Cropping System

Soil respiration refers to a process in which the CO2produced by respiration of soil microorganism and plant roots is released from the soil to the atmosphere. This process plays a large role in global carbon cycling and directly influences the global carbon balance. Soil CO2efflux contains two processes:CO2production and transport. However, previous studies mainly focus on soil CO2efflux only, which provides no information on the vertical distribution of the CO2sources and their seasonal dynamic variation, thus not permitting to study the effect of environmental factors on soil CO2production. In this study, we estimated the soil CO2production and efflux with gradient method by measuring soil CO2concentration, temperature, water content, and porosity at various depths continuously through2011-2012in a winter-wheat-summer-maize farmland. This enables to explore the temporal and spatial patterns of CO2production and their relationship with soil temperature and water content, the effect of straw returned on soil CO2production at different depths and the influence of the variation in soil porosity on estimating soil CO2production. The major findings are as follows:First, the distribution of CO2production at different depths for the winter wheat season follows a generally monotonic trend that larger depth produces less CO2. Specifically, the0-10depth corresponds to3601kg ha-1, follows by10-20cm depth (2364kg ha-1),20-30cm depth (267kg ha-1), and <30cm depth (518kg ha-1), accounting for53%,35%,4%and8%of cumulated soil-surface CO2efflux (6748kg CO2ha-1), respectively. Diurnal change of soil CO2production at different depths was similar to soil temperature at the corresponding depths, but soil CO2production at0-10cm depth reached the perk value at about13:00-1400h, earlier than soil temperature at5cm depth. In winter wheat growing season, the amplitude of diurnal variation of soil CO2production was not consistent with soil temperature; soil water content and temperature had different influences on soil CO2production at various depths and seasons. In early stage (DOY89-133) and late stage (DOY134-175), soil CO2production was controlled by soil temperature and soil water content, respectively. The water content for optimum CO2production at0-10cm depth was smaller than that at10-20depth. Thus, the application of the gradient method allows us to understand the subsurface CO2processes comprehensively by estimating not only soil-surface CO2efflux but also CO2production.Second, the cumulated soil-surface CO2efflux in RTS (reduce tillage with straw) was3058kg ha-1higher than that in RT (reduce tillage without straw) during the winter wheat growing season, which was mainly due to the difference in soil CO2production (1984kg ha-1) at0-10depth between RTS and RT. Though, the absolute soil CO2production at each depth in RTS was higher than that in RT, the relative contribution of soil CO2productions at each depth (i.e.,0-10,10-20,20-30, and <30cm) to the total soil-surface CO2efflux in RTS and RT were similar (i.e.57%28%,6%,9%in RTS and53%,35%,4%,8%in RT, respectively). Cumulated soil CO2production at different depths showed significant linear correlation with the concentration of soil organic carbon, indicating that the difference in soil organic matter between RTS and RT was responsible for the difference between soil CO2production at different depths in RTS and RT. The Q10at0-10and10-20cm depths in RTS were higher than those in RT in both early stage and whole winter wheat season, suggesting that RTS leads to the increase of soil organic carbon in surface soil layer, although such increased soil organic carbon may be easily released to the atmosphere as it is vulnerable to temperature changes.Third, soil porosity changed from0.49m3m-3(the initial measured soil porosity φi at the time of planting) to0.43m3m-3(the final measured soil porosity φf at the time of harvest) during the summer maize season. Compared with soil air-filled porosity, soil gas diffusivity, and soil CO2production estimated with φV (dynamic soil porosity φV (Eq.5-1)), the values estimated with φf were on average6%,11%, and22%smaller, and those estimated with φi were on average17%,36%, and70%larger. In addition, surface soil CO2effluxes calculated with φV showed close agreement with those measured with surface chambers. Therefore, our results support the use of φV in soil CO2production estimations.In conclusion, we demonstrate that the gradient method can be used to effectively estimate soil CO2production, so as to explore the temporal and spatial patterns of CO2production at different depths, and study the relationship between soil CO2production and soil water content and temperature directly. Our work sheds light on the in-depth understanding of the soil respiration processes, especially on subsurface CO2processes.