In this study, we quantified the spatiotemporal variability and trends in observations of multiple snow characteristics in High Arctic Zackenberg in Northeast Greenland through 18 years. Annual premelt snow-depth observations collected in 2005-2014 along an elevation gradient showed significant differences in snow depth between vegetation types. The seasonal snow cover was characterized by strong interannual variability in the Zackenberg region. Particularly the timing of snow-cover onset and melt, and the annual maximum accumulation, varied up to an order of magnitude between years. Hence, apart from the snow-cover fraction registered annually on 10 June, which exhibits a significant trend of-2.3% per year over the 18-year period, we found little evidence of significant trends in the observed snow-cover characteristics. Moreover, SnowModel results for the Zackenberg region confirmed that the pronounced interannual variability in snow precipitations has persisted in this High Arctic setting since 1979 and may have masked potential temporal trends. In exception, a significant difference in interannual variability of snow-cover onset timing was observed through the period 1997-2014, which in the recent period since 2006 was 7.3 times more variable. We owe enormous gratitude to all GeoBasis field assistants, who have collected the snow observations in the Zackenberg Valley during the period 1997-2014. We wish to thank the logistics team at

Snow conditions are important drivers of the distribution and phenology of Arctic flora and fauna, but the extent and effects of local variation in snowmelt are still inadequately studied. We analyze snowmelt patterns within the Zackenberg valley in northeast Greenland. Drawing on landscapelevel snowmelt dates and meteorological data from a central climate station, we model snowmelt trends during 1998-2014. We then use time-lapse photographs to examine consistency in spatiotemporal snowmelt patterns during 2006-2014. Finally, we use monitoring data on arthropods and plants for 1998-2014 to investigate how snowmelt date affects the phenology of Arctic organisms. Despite large interannual variation in snowmelt timing, we find consistency in the relative order of snowmelt among sites within the landscape. With a slight overall advancement in snowmelt during the study period, early melting locations have advanced more than late-melting ones. Individual organism groups differ greatly in how their phenology shifts with snowmelt, with much variance attributable to variation in life history and diet. Overall, we note that local variation in snowmelt patterns may drive important ecological processes, and that more attention should be paid to variability within landscapes. Areas optimal for a given taxon vary between years, thereby creating spatial structure in a seemingly uniform landscape. ARTICLE HISTORY

Arctic tundra is a globally important store for carbon (C). However, there is a lack of reference sites characterising C exchange dynamics across annual cycles. Based on the Greenland Ecosystem Monitoring (GEM) programme, here we present 9–11 years of flux and ecosystem data across the period 2008–2018 from two wetland sites in Greenland: Zackenberg (74°N) and Kobbefjord (64°N). The Zackenberg fen was a strong C sink despite its higher latitude and shorter growing seasons compared to the Kobbefjord fen. On average the ecosystem in Zackenberg took up ∼−50 g C m−2 yr−1 (range of +21 to −90 g C m−2 yr−1), more than twice that of Kobbefjord (mean ∼−18 g C m−2 yr−1, and range of +41 to − 41 g C m−2 yr−1). The larger net carbon sequestration in Zackenberg fen was associated with higher leaf nitrogen (71%), leaf area index (140%), and plant quality (i.e. C:N ratio; 36%). Additional evidence from in-situ measurements includes 3 times higher levels of dissolved organic carbon in soils and 5 times more available plant nutrients, including dissolved organic nitrogen (N) and nitrates, in Zackenberg. Simulations using the soil-plant-atmosphere ecosystem model showed that Zackenberg’s stronger CO2 sink could be related to measured differences in plant nutrients, and their effects on photosynthesis and respiration. The model explained 69% of the variability of net ecosystem exchange of CO2, 80% for photosynthesis and 71% for respiration over 11 years at Zackenberg, similar to previous results at Kobbefjord (73%, 73%, and 50%, respectively, over 8 years). We conclude that growing season limitations of plant phenology on net C uptake have been more than counterbalanced by the increased leaf nutrient content at the Zackenberg site.

The Arctic is getting warmer and wetter. Here, we document two independent examples of how associated extreme precipitation patterns have severe implications for high Arctic ecosystems. The events stand out in a 23-year record of continuous observations of a wide range of ecosystem parameters and act as an early indication of conditions projected to increase in the future. In NE Greenland, August 2015, one-quarter of the average annual precipitation fell during a 9-day intensive rain event. This ranked number one for daily sums during the 1996–2018 period and caused a strong and prolonged reduction in solar radiation decreasing CO2 uptake in the order of 18–23 g C m−2, a reduction comparable to typical annual C budgets in Arctic tundra. In a different type of event, but also due to changed weather patterns, an extreme snow melt season in 2018 triggered a dramatic gully thermokarst causing rapid transformation in ecosystem functioning from consistent annual ecosystem CO2 uptake and low methane exchange to highly elevated methane release, net source of CO2, and substantial export of organic carbon downstream as riverine and coastal input. In addition to climate warming alone, more frequent occurrence of extreme weather patterns will have large implications for otherwise undisturbed tundra ecosystems including their element transport and carbon interactions with the atmosphere and ocean.