IAim Winter snow has been suggested to regulate terrestrial carbon (C) cycling
by modifying microclimate, but the impacts of change in snow cover on the
annual C budget at a large scale are poorly understood. Our aim is to quantify
the C balance under changing snow depth.
Location Non-permafrost region of the northern forest area.
Methods Here, we used site-based eddy covariance flux data to investigate the
relationship between depth of snow cover and ecosystem respiration (Reco)
during winter. We then used the Biome-BGC model to estimate the effect of
reductions in winter snow cover on the C balance of northern forests in the
non-permafrost region.
Results According to site observations, winter net ecosystem C exchange
(NEE) ranged from 0.028 to 1.53 gCm22
day21
, accounting for 44 6 123% of
the annual C budget. Model simulation showed that over the past 30 years,
snow-driven change in winter C fluxes reduced non-growing season CO2
emissions, enhancing the annual C sink of northern forests. Over the entire
study area, simulated winter Reco significantly decreased by 0.33
gCm22
day21
year21 in response to decreasing depth of snow cover, which
accounts for approximately 25% of the simulated annual C sink trend from
1982 to 2009.
Main conclusion Soil temperature is primarily controlled by snow cover
rather than by air temperature as snow serves as an insulator to prevent
chilling impacts. A shallow snow cover has less insulation potential, causing
colder soil temperatures and potentially lower respiration rates. Both eddy
covariance analysis and model-simulated results show that both Reco and NEE
are significantly and positively correlated with variation in soil temperature
controlled by variation in snow depth. Overall, our results highlight that a
decrease in winter snow cover restrains global warming as less C is emitted to
the atmosphere.
Key