Climate extremes in Svalbard over the last two millennia are linked to atmospheric blocking

Abstract

Arctic precipitation in the form of rain is forecast to become more prevalent in a warmer world but with seasonal and interannual changes modulated by natural modes of variability. Experiencing rapid hydroclimatic changes in the Arctic, Svalbard serves as an ideal study location due to its exposure to oceanic and atmospheric variability in the North Atlantic region. Here we use climate data from paleoproxies, observations, and a climate model to demonstrate that wet and warm extremes in Svalbard over the last two millennia are linked to the presence of atmospheric blocking regimes over Scandinavia and the Ural mountain region. Rainfall episodes lead to the deposition of coarse sediment particles and high levels of calcium in Linnévatnet, a lake in southwest Svalbard, with the coarsest sediments consistently deposited during atmospheric blocking events. A unique annually resolved sediment record from Linnévatnet confirms that this linkage has been persistent over the past 2000 years. Our record also shows that a millennial-scale decline in Svalbard precipitation ended around the middle of the 19th century, followed by several unprecedented extreme events in recent years. As warming continues and sea ice recedes, future Svalbard floods will become more intense during episodes of Scandinavian and Ural blocking.

Fig. 1. The relationship between Jun–Nov precipitation in Svalbard and atmospheric circulation features.

a Seasonal mean precipitation in Svalbard (PR_Sv; red rectangle in the inset) from June through November in two land-only observational datasets (see Methods): GPCC (solid black) and UDEL (dotted grey). Regression maps showing the relationship, for the period 1955–2019, between Jun-Nov monthly precipitation (GPCC PR_Sv) and (b) 500 hPa geopotential height (Z500) anomalies (colors) and winds (V500; arrows) from ERA5; (c) Sea surface temperature (SST) anomalies from HadISST dataset and (d) surface air temperature (SAT) anomalies from ERA5. Regression maps show changes for one standard deviation change in the monthly PR_Sv time series. Cross-hatching shows regions with significant regression coefficients at the 95% significance level.

Fig. 2. Linnévatnet’s sediment traps reveal the influence of atmospheric blocking in regulating flooding events.

a Air photo of Linnévatnet and location of intervalometer at mooring C (coring site). b Photograph of the gravity sediment core collected in 2019 showing the µ-XRF calcium variation throughout the first 41 cm. The year 2016 is characterized by highest Ca values. c Scarp on the eastern carbonate bedrock terrain (red square in (a)) triggered by anomalously warm weather and intense rainfall recorded on October 15, 2016. The worst mudflow in 40 years occurred that year with roads being completely flooded in Longyearbyen ((d), photograph from W.R. Farnsworth). e Grain-size data collected from the sediment trap at mooring C in 2015–2016 showing two days with coarsest grain-size peaking on September 11, 2015 and May 28, 2016, respectively. The background colors are from bottom to top: Green = summer 2015, yellow = fall 2015, purple = winter 2015/2016 and blue = spring and summer 2016. The sample symbols Blue = median, d50, red = mean and green = d90. f–g, Atmospheric pressure anomalies at 500 hPa Geopotential height (Z500) (relative to 1980–2010) during those two days calculated from daily National Centers for Environmental Prediction reanalysis data.

Fig. 3. Paleo evidence links Scandinavian blocking to Common Era flood events in Svalbard.

a The Scandinavian (Scand) reconstructed summer temperatures from the N-TREND tree-ring dataset (in red; data were area-weight averaged over 12–30°E and 65–70°N) and the Northern Scandinavian reconstructed temperature (in green; N. Scand) from Esper et al. compared to the Calcium variability at Linnévatnet (in black). For visibility only the 1-750CE time interval is shown for the N. Scand ((a); green) as the two Scand reconstructions exhibit strong co-variablity during their overlapping period (750 CE-2006 CE, annual: r = 0.87). b Spatial correlation between the Scand temperature reconstruction and atmospheric pressure at 500 hPa geopotential height (Z500). The blue star denotes the location of Linnévatnet in Svalbard. c Reconstructed sea ice concentration from the Nordic Seas (blue) compared to the Calcium at Linnévatnet. Periods of lower and higher sea ice concentration (SIC) are highlighted in red and blue, respectively. d Spatial correlation between the Scand temperature reconstruction and sea surface temperature from HadISST. Blue area delimits the location of the targeted reconstructed SIC from the Nordic Seas in (c). The tree-ring and sea-ice time series were filtered using 5-year running mean to improve visibility. Correlations shown (b, d) are significant with a P < 0.1.

Fig. 4. Relationship between regional circulation and Svalbard temperature and precipitation in MPI-ESM1-2-LR transient simulation over the Common Era (past2k experiment).

a Scandinavian Blocking (SB) pattern calculated by regressing Jun-Nov monthly 500 hPa geopotential height (Z500) anomaly field on the SB index (see Methods). b Regressions of Jun-Nov monthly temperature and the SB index. c Regressions of Jun-Nov monthly precipitation and the SB index. d Relationship between Jun-Nov monthly temperature and precipitation in over the period 1–1850 CE. The spatial average is calculated over the domain shown in Fig. 1a. e Relationship between the strength of the SB index and low-level moisture transport over the Greenland Sea. Cross-hatching shows regions with significant regression coefficients at the 95% significance level.

Fig. 5. Climate projections for Svalbard based on two CMIP6 models.

a Total change in Jun-Nov mean temperature and precipitation in Svalbard (see inset for area definition) between 2015 and 2099 based on linear trends for three emissions scenarios (SSP2-4.5, SSP3-7.0, SSP5-8.5). b Histograms showing the distributions of number of days with precipitation >10 mm across Svalbard (at 0.5˚ spatial resolution) over the period 2015–2034 (dashed line) and 2080–2099 (solid line) under SSP3-7.0 for IPSL-CM6A-LR (left) and MPI-ESM1-2-HR (right).