چکیده (انگلیسی):

Changes in cryospheric snow accumulation, snowmelt, and soil freezing and thawing might influence the ground-surface CO2 flux and cumulative winter CO2 flux from agricultural land. We continuously observed CO2 flux using automatically closing chambers at an untreated control plot and a plot with snow removal in northern Japan. The CO2 in soil pores at 10-cm depth increased by 6.5 ppmv day−1 as soil began to freeze, but it increased dramatically (to 49 ppmv day−1) after snowmelt water infiltrated the soil and froze. The soil-frost layer constrained gas diffusion into the air, and the barrier strengthened as the frozen snowmelt water decreased the air volume in soil pores. Leached gas CO2 from the freezing snowmelt water also increased gas CO2 concentration in soil. As the soil thawed, the CO2 concentration decreased drastically, at 790 ppmv day−1. However, these changes had little effect on CO2 flux. The soil CO2 concentration remained stable after snow cover reached 30 cm in the control plot. Low CO2 flux in both plots occurred during the winter. No clear relation was found between CO2 flux and snow depth or soil-frost depth because of the small CO2 source at this site. We also considered how the presence of the chamber influenced soil temperatures and water contents. During the snow-free season, the chamber mitigated diurnal changes in soil temperature. The daily average soil temperature differed from that in the natural state by −1.7 °C to 6.3 °C. This fluctuation of temperature corresponded to the fluctuation of CO2 flux, which ranged from 91% to 143% of the CO2 flux in the natural state based on the temperature-response equations. The chamber had little influence on the soil temperature during the snow-cover period, and did not influence soil water content throughout the study period. Cumulative winter CO2 emissions were 17.2 gC m−2 (over 143 days) in the control plot and 13.4 gC m−2 (over 151 days) in the treated plot (10.0 and 7.5% of annual accumulation, respectively).