Extremely wet summer events enhance permafrost thaw for multiple years in Siberian tundra.

Permafrost thaw can accelerate climate warming by releasing carbon from previously frozen soil in the form of greenhouse gases. Rainfall extremes have been proposed to increase permafrost thaw, but the magnitude and duration of this effect are poorly understood. Here we present empirical evidence showing that one extremely wet summer (+100 mm; 120% increase relative to average June-August rainfall) enhanced thaw depth by up to 35% in a controlled irrigation experiment in an ice-rich Siberian tundra site. The effect persisted over two subsequent summers, demonstrating a carry-over effect of extremely wet summers. Using soil thermal hydrological modelling, we show that rainfall extremes delayed autumn freeze-up and rainfall-induced increases in thaw were most pronounced for warm summers with mid-summer precipitation rainfall extremes. Our results suggest that, with rainfall and temperature both increasing in the Arctic, permafrost will likely degrade and disappear faster than is currently anticipated based on rising air temperatures alone.

Design of the tundra rainfall experiment (TRainEx) to simulate future summer precipitation scenarios.

The majority of climate models predict severe increases in future temperature and precipitation in the Arctic. Increases in temperature and precipitation can lead to an intensification of the hydrologic cycle that strongly impacts Arctic environmental conditions. In order to investigate effects of future precipitation scenarios on ecosystems, precipitation manipulation experiments are being performed to simulate drought and extreme precipitation conditions. However, most of the existing research so far has been unevenly distributed, primarily focusing on temperate grasslands and woodlands. Despite large changes in the predicted precipitation and potentially high sensitivity of the Arctic tundra ecosystem to these changes, it is among the most understudied ecosystems for precipitation manipulation experiments. Gherardi and Sala (2013) presented a design for precipitation manipulation experiments that, relative to other methods at the time, was cheap, simplistic, and easily reproducible. In this study, we:•Present modifications to the original Gherardi and Sala (2013) design that are adapted to cold, harsh conditions, such as those present in the Siberian Arctic tundra.•Provide a detailed documentation of the improved design.•Validate our modified experimental design based on the first two years of our experiment.