| Summary: | The Arctic is currently experiencing rapid increases in air temperature and changes in the precipitation patterns. Consequently, an increase in winter snowfall may be expected in the Arctic. Due to enhanced thermal insulation of deepened snow, soil temperature during winter becomes warmer, which could accelerate soil nitrogen (N) cycling rates and increase the availability of N during the subsequent growing season. However, the magnitude of these responses to deepened snow varies between ecosystem type and systematical comparisons of contrasting tundra ecosystems across the Arctic have not been studied. The climate change in the Arctic also increases the frequency and severity of fire in many tundra regions. Fires can immediately remove vegetation, release carbon (C) and N into the atmosphere, and disrupt ecosystem N and other nutrient cycling. Despite these detrimental effects, the fate and distribution of post-fire C and N as well as post-fire greenhouse gas (GHG) emissions in Arctic tundra ecosystems remain largely unknown. This thesis addresses the effects of increased winter snow precipitation and tundra fires on soil N-cycle processes and GHG emissions in Arctic tundra ecosystems. Soil gross N cycling rates in response to experimentally deepened snow were investigated and compared among five contrasting tundra sites across the Arctic (i.e., Greenland (Disko Dry and Wet), Canada (Daring Lake), Norway (Svalbard Heath and Meadow)). The fate of post-fire mineral N was assessed using stable isotopic labelling (ammonium ( 15 NH 4 + -N) and nitrate ( 15 NO 3 − -N)) in a dry heath tundra (Disko, Greenland). The effects of experimental fire in combination with warming on soil GHG emissions were also investigated. The deepened snow across all sites enhanced gross nitrification, NO 3 − -N immobilization, and denitrification in winter, as well as gross N mineralization and denitrification in summer. The two wetter sites (Disko Wet and Daring Lake) showed more pronounced increases in N cycling rates than the other three drier sites. This suggests that the potential effects of increased winter snow precipitation on microbial N-cycle activities will be most pronounced in relatively moist tundra ecosystems. Tracing the fate of applied 15 N following the fire showed higher soil 15 N recovery in burned than unburned plots, with increased contribution of microbial 15N recovery to bulk soil 15N recovery post fire. This suggests an increase in post-fire soil N retention, which was probably due to the increased incorporation of N into microbial biomass. The responses of plant N uptake to the fire differed among dominant shrub species, suggesting a postfire shift in composition and structure of the dry heath tundra community in the long term. Moreover, soil NO 3 − -N, NH 4 + -N and phosphate (PO 4 3− -P) concentrations increased two years after the fire. Two years after the fire, soil nitrous oxide (N 2 O) and carbon dioxide (CO 2 ) production was higher in burned than unburned soils, while methane (CH 4 ) uptake remained unchanged. However, these post-fire increases in N 2 O and CO 2 production were only apparent when soils were at maximum water holding capacity, suggesting that fire effects can be masked under in situ dry soil conditions. There was a lack of moderate warming responses in soil respiration and CH 4 oxidation in burned soils. Altogether, our results suggest that fire potentially increases soil GHG emissions (e.g. N 2 O and CO 2 ) especially during episodes with wet soil conditions. On the other hand, the lack of warming responses in post-fire soil respiration may weaken this positive feedback to climate change.
|
|---|