1. Introduction
Greenland is the largest island on Earth with most of its surface (∼85%) covered by ∼1500-m-thick ice. The loss of mass from the Greenland Ice Sheet (GrIS) presently contributes the largest amount of meltwater to global sea level rise since the 1990s (Gardner et al. 2013; van den Broeke et al. 2016; IMBIE Team 2020), and the mass loss is currently accelerating (IMBIE Team 2020; Mankoff et al. 2021). Warmer air temperature is considered to be the dominating factor for the rapid mass loss of the GrIS (Box et al. 2012; Hanna et al. 2008, 2020, 2021). Precipitation determines the accumulation and generally drives the interannual variability of the GrIS mass balance (Koenig et al. 2016; Hofer et al. 2019; Huai et al. 2021). Thus, deciphering the variations of temperature and precipitation is important for better understanding the behavior of the GrIS. Such an investigation is aided by the establishment of observational networks [e.g., Programme for Monitoring of the Greenland Ice Sheet (PROMICE) and Greenland Climate Network (GC-Net)] and high-resolution regional climate simulations [e.g., MAR, Regional Atmospheric Climate Model (RACMO), HIRHAM] over Greenland.
Observations from coastal stations suggested a significant warming trend since the 1990s in all seasons, which contributes to the melting of the GrIS at low elevations (Tedesco et al. 2016; Hanna et al. 2021; Zhang et al. 2022). A similar warming trend is observed at inland stations over Greenland in summer, though with smaller magnitude relative to coastal stations (Hanna et al. 2021). The strongest warming occurs in west and northwest Greenland from 1991 to 2019, which reaches up to ∼6°–6.5°C in winter (Hanna et al. 2021). In addition to increased greenhouse gases, the interannual variation of temperature over Greenland is suggested to be largely tied to change in Greenland blocking, which modulates temperature by altering warm-air advection and cloud cover (Nghiem et al. 2012; Hofer et al. 2017; Preece et al. 2022). Precipitation over Greenland experienced a major shift around 1960s, manifested as wetter conditions along coastal Greenland and drier conditions over the ice sheet (Mernild et al. 2015). However, the trend in seasonal precipitation varies greatly across Greenland. For example, summer precipitation shows an increasing trend over northwest Greenland (i.e., coastal Thule Air Base), but a decreasing trend over southern coastal regions (Hawley et al. 2014; Wong et al. 2015; Mernild et al. 2015). Over the recent two decades (2000–19), decreased summer precipitation was observed across the western percolation zone, contributing to the increased surface melting (Lewis et al. 2019).
With rising mean temperature, Greenland experienced multiple extreme weather events in recent decades that led to significant melting of the ice sheet and caused large damages to coastal communities (Mote 2007; Nghiem et al. 2012; Tedesco et al. 2014; Hanna et al. 2014; Moon et al. 2020; Wei et al. 2022). In summer 2012, a warm spell over Greenland caused a record-breaking surface melt extending over nearly the entire ice sheet, which also led to new records for low albedo in Greenland (Nghiem et al. 2012; Hanna et al. 2014). A similar extreme warm event occurred in summer 2019, with approximately 90% of the surface of Greenland reaching the melting point. This led to more extensive bare ice and saturated snow/firn and a rise of global mean sea level by 1.5 mm (Tedesco and Fettweis 2020; Cullather et al. 2020; Sasgen et al. 2020). Additionally, August 2021 was noted by an extreme rain event that occurs at the Summit Station of Greenland for the first time on record, and the amount of ice mass lost was 7 times higher than the daily average for this time of year (Moon et al. 2021; https://nsidc.org/greenland-today/greenland-surface-melt-extent-interactive-chart/).
These events highlight the high sensitivity of the GrIS to climate change, given the accompanied massive surface melting. Moreover, extreme events may affect Greenland ecosystems by either crossing a critical threshold or occurring simultaneously with other events (e.g., Walter et al. 2014; Lamarche-Gagnon et al. 2019; Walsh et al. 2020; Culberg et al. 2021). And the harsh conditions induced by extreme events pose new challenges for the observation systems. Thus, it is of great importance to understand the variations of extreme weather/climate events over Greenland. Scientists are unlocking the characteristics of several extreme events and identifying the underlying processes (Bennartz et al. 2013; Mattingly et al. 2020; Sasgen et al. 2022). The intensity and frequency of extreme surface mass balance over the GrIS have been examined recently (Wei et al. 2022). But a comprehensive evaluation of extreme climate events over Greenland has not yet been performed, partially owing to the lack of long-term continuous daily observations.
In this study, we complied daily in situ observations from automatic weather stations (AWSs) and staffed stations to evaluate the RACMO2.3p2. Next, we examine the spatiotemporal variations of extreme temperature and precipitation events over Greenland during 1958–2019 in terms of the long-term trend and interannual variability, using daily outputs from the RACMO2.3p2. Such an investigation may advance our understanding of Greenland’s distinct climatic regimes and provides insights into climate risks across the Arctic and their links with global climate change. The remainder of the paper is organized as follows. In section 2, we introduce the datasets and methods used. We assess the performance of RACMO2.3p2 in depicting extreme climate events over Greenland and examine the variations of extreme temperature and precipitation events in section 3. The possible dynamic mechanisms are discussed in section 4. We summarize the main results in section 5.
2. Data and methods
a. In situ observations
1) Data sources
We collect daily mean temperature, maximum/minimum temperature, and precipitation from 176 observational stations over Greenland (Fig. 1). These station-based measurements are obtained from the GC-Net, PROMICE, Global Historical Climatology Network daily (GHCNd), and Global Surface Summary of the Day (GSOD) (Table 1).
Locations of Greenland meteorological stations used in this study. Stations where only extreme temperature or extreme precipitation can be calculated are indicated by black and blue dots, respectively. Stations where both extreme temperature and precipitation can be calculated are indicated by red dots. Sites where the trend can be calculated are indicated by circles.
Citation: Journal of Climate 36, 14; 10.1175/JCLI-D-22-0488.1
Brief description of the observational-based datasets used in this study.
The GC-Net was established in 1994 and includes 23 AWSs that are spatially scattered over the GrIS and monitor climatological and glaciological parameters (Steffen et al. 1996). Each AWS samples hourly air temperature, wind speed, wind direction, humidity, pressure, and radiation, and the length of observations varies by station.
The PROMICE was initiated in 2007 and consists of 25 AWSs over Greenland. In contrast with the GC-Net, the PROMICE network is placed on the margins of the GrIS and peripheral glaciers in order to complement the spatial distribution of existing ice sheet stations. Measurements at each AWS are taken every 10 min and are postprocessed (e.g., filtering unlikely values using thresholds) to obtain daily products from 2007 to 2020 (Fausto et al. 2021).
The GHCN consists of daily meteorological variables from >100 000 land stations that have been integrated and gone through a common suite of quality assurance checks (Alexander et al. 2006). A total of 70 stations that are located in Greenland and contain daily temperature and precipitation measurements are selected for this study. The record length of the chosen stations spans from 1 to 127 years.
The GSOD is derived from the Integrated Surface Hourly dataset that was obtained from the U.S. Air Force climatology center. The GSOD consists of daily meteorological records from >9000 global weather stations (Lott et al. 2008). Here, we use historical datasets from 58 stations over Greenland that provide daily maximum temperature, minimum temperature, and/or precipitation.
2) Quality control
We merge the aforementioned observational datasets from 48 AWSs and 128 staffed stations and keep only one station based on data integrity if stations from different sources overlap. The number of overlapped stations from different sources is 5 and accounts for ∼3% of total stations. Next, we perform a three-step data quality-control procedure, following the methods of Wei et al. (2019). Briefly, we first discard unrealistic daily values at each station using a threshold method (i.e., 3 times the standard deviation). Note that the choice of 3 times the standard deviation as the threshold may incorrectly exclude a few extreme observations, but this has very limited influence for studying climatological states. Next, we require that the success rate in each month of a year (defined as the ratio between the days with available measurements and the total number of days in a month) is larger than 80%. Last, the length of records that pass the two-step quality control at a station is ≥5 years, but not necessarily continuous.
Data from 91 of the 176 stations meet these criteria and are used to calculate the spatial distribution of temperature-related extremes during 1958–2019 over Greenland, and 35 stations are available for precipitation-related extremes (Table 1). The sites used for the evaluation of temperature-related extremes are located from coastal to inland Greenland, whereas the stations used for precipitation-related extremes are all distributed along the margins of Greenland (Fig. 1). When computing the temporal trend, we further require that the number of years with quality-controlled datasets at a station should be ≥20 years (missing values do not exceed 10%). The number of stations meeting this criterion sharply reduces to 17 and 14 for temperature and precipitation-related extremes, respectively (Fig. 1). Note that if a longer period is required (e.g., 30 years), the available stations for temperature and precipitation extremes further decrease to 9.
b. RACMO2.3p2 simulations
The RACMO model is developed and maintained at the Royal Netherlands Meteorological Institute (Van Meijgaard et al. 2008). RACMO adopts physical schemes developed by the European Centre for Medium-Range Weather Forecasts (White 2001) and dynamical core from High Resolution Limited Area Model (Undén et al. 2002). Here, RACMO version 2.3p2 that incorporates a drifting snow physics (Lenaerts et al. 2012) is run at 5.5-km horizontal resolution over Greenland, driven by ERA-40 (Uppala et al. 2005) for the period 1958–78, ERA-Interim (Stark et al. 2007) for the period 1979–89, and ERA5 (Hersbach et al. 2020) for the period 1990–2019. The previous evaluations have illustrated that RACMO2.3p2 is skillful in depicting the observed characteristics of surface climates over Greenland (e.g., Noël et al. 2019; Fettweis et al. 2020). More information concerning the model and experimental design is given in Noël et al. (2019).
c. Extreme indices
Here, we use three indices to measure the intensity of temperature and precipitation-related extreme events over Greenland, following the guidance of Expert Team on Climate Change Detection and Indices (ETCCDI; Karl et al. 1999). Specifically, annual maximum value of daily maximum temperature (TXx; an indicator for the hottest daytime temperature) and annual minimum value of daily minimum temperature (TNn; an indicator for the coldest nighttime temperature) are used to measure the intensity of temperature-related extremes. The annual maximum daily precipitation amount (Rx1day) is adopted to measure the intensity of precipitation-related extremes. These indices have been proven useful in depicting extreme climate events and hence widely used to identify extreme events across the world (e.g., Kharin et al. 2007; Sillmann et al. 2013; Fischer and Knutti 2015; Seneviratne et al. 2021). In this study, these indices are applied to the observational dataset and the simulated gridded data in the same ways. Notably, the extreme value in the observation records does not always correspond to the same day as in the model output. This is attributed to the fact that the internal variability of climate systems cannot be fully captured in RACMO2.3p2.
It should be mentioned that there are generally gaps in the observed time series owing to the harsh environment over Greenland, which greatly affect the calculation of the threshold indices that rely on the reference data selected and hence their reliability in representing extreme events. On the other hand, global/regional climate models have been shown to have relatively lower skills in capturing the characteristics of the threshold indices than absolute indices (e.g., Sillmann et al. 2013; Guo et al. 2021; Kim et al. 2020). Thus, the variations of threshold indices (e.g., TX90p defined as the percentage of days when daily maximum temperature > 90th percentile according to ETCCDI) that are used to measure the frequency of extreme events are only briefly discussed here.
3. Results
a. Model evaluation
Figure 2 shows the scatterplot of the observed and modeled temperature and precipitation extremes over Greenland from 1958 to 2019. Based on the observations, annual maximum temperature (i.e., TXx) over Greenland ranges from ∼0°C at the summit of the ice sheet to up to 24°C along low-lying coastal regions, suggesting a general inland-to-coastal increasing pattern (Fig. 2a). The climatological distribution of TXx is well reproduced in the model. The correlation coefficient between observations and simulations reaches 0.87, with a root-mean-square error (RMSE) of 3.4°C and a cold bias of −1.5°C. As shown in Fig. 2d, 16 of the 17 sites show a long-term positive trend in TXx, indicating intensified extreme warm events over Greenland. The increasing trend is generally reproduced in RACMO2.3p2, with an RMSE of 0.25°C decade−1, though the simulated trend is larger (by ∼24% on average) than the observations. The observed annual minimum temperature (TNn) over Greenland spans from −12°C at Qaqortoq Heliport station to −58°C at Ngrip station, which seems dependent on both latitude and elevation (Fig. 2b). The RACMO2.3p2 captures the mean state of the TNn (r = 0.93), with an RMSE of 4.8°C and a warm bias of ∼2.7°C. The model also reproduces the increasing trend of the TNn at most stations (r = 0.64), though the magnitude is generally underestimated (Fig. 2e).
Scatterplot of the observed and modeled climatological (a) annual maximum temperature (TXx; °C), (b) annual minimum temperature (TNn; °C), and (c) annual maximum daily precipitation amount (Rx1day; mm w.e. day−1) and trend of (d) TXx (°C yr−1), (e) TNn (°C yr−1), and (f) Rx1day (mm w.e. day−1 year−1). The insert map in each subplot shows the spatial distributions of climate extremes based on the observations.
Citation: Journal of Climate 36, 14; 10.1175/JCLI-D-22-0488.1
Regarding extreme precipitation, Rx1day over Greenland spans from nearly zero to ∼80 mm day−1 at the available stations, with the relatively stronger extreme events occurring over the southeast coast. The pattern of the modeled Rx1day is broadly in agreement with the observations (r = 0.73). However, RACMO2.3p2 overestimates the intensity of extreme precipitation, with an RMSE of 32.2 mm day−1 and a wet bias of 26.3 mm day−1 (106.3%) (Fig. 2c). The signs of the temporal trend in 10 of the 14 stations are reproduced in the simulations. On average, there is an RMSE of 0.23 mm day−1 decade−1 and a bias of −0.1 mm day−1 decade−1 (−60.7%) between the modeled and observed trends in Rx1day at all stations.
Additionally, we compare the modeled daily temperature/precipitation with the observations at 11 sites with quality-controlled records longer than 30 years. We find that the model not only captures the long-term trend of the observed temperature and precipitation extremes, but also reasonably reproduces the interannual variability at most sites during 1958–2019 (Fig. 3). The RMSE in interannual variability between simulations and observations is ∼0.5° and 0.9°C for TXx and TNn, respectively, with a relative bias of −18% and +3%. For the Thule Air Base station that has the longest records for extreme temperature (62 years), the model–data correlation coefficient reaches 0.58 (0.53) for TXx (TNn), with an RMSE of 2.1°C (2.7°C). The temporal evolution of extreme precipitation is well captured, but the model overestimates the observed interannual variability on average by ∼62%. The overestimated interannual variability may be partially attributed to the uncertainty in model parameterizations (e.g., representing solid precipitation; Huai et al. 2021) and the horizontal resolution (Niwano et al. 2021; Huai et al. 2022), which leads to too intense precipitation along coastal regions where the sites with long-term records are generally located.
The time series of the modeled (color lines) and observed (black lines) TXx (°C), TNn (°C), and Rx1day (mm w.e. day−1) at each subplot (from top to bottom) at 9 stations with long-term (≥30 years) extreme temperature or precipitation records during 1958–2019. The scatterplot is the observed and modeled interannual variability of TXx, TNn and Rx1day at the 9 stations. Note that two stations have only extreme temperature or extreme precipitation records.
Citation: Journal of Climate 36, 14; 10.1175/JCLI-D-22-0488.1
Overall, RACMO performs well in representing the climatological TXx and TNn, and although it generally overestimates daily maximum precipitation, its simulated values are the correct order of magnitude. Additionally, it should be noted that the relatively short observational records make it difficult to evaluate long-term trends, but for the available data, RACMO seems to perform well, especially for the sign of the temporal trend. Therefore, in the following sections, we use the outputs of RACMO2.3p2 to perform a detailed examination of the spatiotemporal features of extreme events across Greenland.
b. Temperature-related extremes from 1958 to 2019
During the past six decades (1958–2019), extreme warm events measured by TXx on average experience a significant increasing trend (0.13°C decade−1; significant at the 95% confidence level using the two-tailed Student’s t test; Fig. 4a), especially after the 1990s during which the trend has more than doubled (0.28°C decade−1). To examine the spatial variation, we divide Greenland into ice-free coastal region and the GrIS, which is further divided into seven subregions according to Mouginot et al. (2019). As shown in Fig. 4c and Table 2, the warming trend in TXx is observed almost over the entire Greenland during 1958–2019. The most intense increase occurs over the ice-free coastal region (0.35°C decade−1), followed by the northern parts of the GrIS (∼0.1°C decade−1), and the least warming is observed over the southwest GrIS (0.04°C decade−1). The warming trend is generally intensified after the 1990s (i.e., 1990–2019) over Greenland, especially over the ice-free coastal region and northern Greenland (Fig. 4d; Table 2). Interestingly, the warming trend in TXx is smaller than that of mean surface air temperature over the GrIS during 1958–2019, which is opposite to the changes over the majority of Northern Hemisphere lands where extreme warm events increase more rapidly than the mean temperature in recent decades (e.g., Seneviratne et al. 2021). Besides, we found that the variation of the TX90p (defined as the percentage of days when daily maximum temperature > 90th percentile) show large similarities with the TXx in terms of interannual variation and long-term trend (Figs. S1 and S3 in the online supplemental material).
(a) The time series of TXx averaged over Greenland (dotted lines) from 1958 to 2019 and its 9-yr running mean (black line). The green (orange) line shows the long-term linear trend from 1958 to 2019 (1990–2019). The gray line shows the NASA global land–ocean temperature anomaly relative to 1951–80. (b) The time series of standardized detrended TXx. The distributions of the trend in TXx during (c) 1958–2019 and (d) 1990–2019. (e) Composite difference in TXx between high and low TXx years [i.e., red and blue dots in (b)]. The dotted areas in (c)–(e) are statistically significant at the 95% level. All data used here are calculated from RACMO2.3p2 results.
Citation: Journal of Climate 36, 14; 10.1175/JCLI-D-22-0488.1
Linear trend of each extreme index during 1958–2019 (1990–2019). Asterisks indicate that the trends are statistically significant at the 95% level. T2m means 2-m air temperature.
With respect to interannual variability, we focus on the difference in temperature extremes between strong and weak extreme years. According to the threshold of ±1 standard deviation of the detrended TXx, nine years (1958, 1959, 1966, 1968, 1975, 1990, 2002, 2012, 2019) are defined as strong TXx years during 1958–2019 and nine years (1973, 1977, 1985, 1986, 1992, 1993, 1996, 2001, 2013) are chosen as weak TXx years (Fig. 4b). Based on composite analysis (i.e., the difference in the averaged TXx between the selected strong and weak TXx years), TXx is significantly higher (>2°C) over the ice-free coastal region during the strong TXx years than the weak TXx years (Table 3). Larger TXx difference (>1°C) is also observed over the north and central parts of the GrIS (Fig. 4e), especially at the relatively higher elevations (>2000 m). In contrast, there is slight change and even cooler TXx over southern GrIS during the strong TXx years (Fig. 4e). This pattern shares large similarity with the distribution of the long-term trend during 1990–2019 (Fig. 4d). Additionally, although the selection of extreme years depends on the indices we used (e.g., TXx versus TX90p), the associated spatial differences between strong and weak extreme years are broadly similar, though with local differences (Fig. S3).
The differences between strong and weak years for each extreme index during 1958–2019. Asterisks indicates that the differences are statistically significant at the 95% level.
A warming trend is also observed in TNn during 1958–2019, indicating weakened extreme cold events over Greenland. Specifically, TNn averaged over Greenland exhibits a significant long-term upward trend from 1958 to 2019 (0.24°C decade−1), together with clear decadal fluctuations (Fig. 5a). Similar to TXx, the regions with the largest warming trend of TNn during 1958–2019 are the ice-free coastal Greenland (0.42°C decade−1; Table 2), followed by the northern GrIS (0.17°–0.29°C decade−1). After 1990, TNn over the entire Greenland shows an intensified increasing trend, which reaches 1.24°C decade−1 (Figs. 5a,d). Notably, the regions with the most intense change shift from the ice-free coastal Greenland to southwest GrIS (Table 2). Additionally, the increasing trend in TNn during 1990–2019 is considerably larger than the trends of mean air temperature and TXx across Greenland, similar to the conditions over Northern Hemisphere lands (e.g., Seneviratne et al. 2021).
(a) The time series of TNn averaged over Greenland (dotted lines) from 1958 to 2019 and its 9-yr running mean (black line). The green (orange) line shows the long-term linear trend from 1958 to 2019 (1990–2019). The gray line shows the NASA global land–ocean temperature anomaly relative to 1951–80. (b) The time series of standardized detrended TNn. The distributions of the trend in TNn during (c) 1958–2019 and (d) 1990–2019. (e) Composite difference in TNn between strong and weak TNn years [i.e., red and blue dots in (b)]. The dotted areas in (c)–(e) are statistically significant at the 95% level. All data used here are calculated from RACMO2.3p2 results.
Citation: Journal of Climate 36, 14; 10.1175/JCLI-D-22-0488.1
Additionally, TNn over Greenland shows clear interannual variability during 1958–2019. To examine the spatial difference, we define 9 years (1971, 1972, 1983, 1989, 1992, 1993, 1995, 2002, 2015) and 11 years (1958, 1962, 1963, 1980, 1985, 2003, 2005, 2010, 2014, 2016, 2018) as the low and high TNn years based on ±1 standard deviation of the detrended TNn, respectively (Fig. 5b). Compared with the weak TNn years, TNn is significantly lower over the entire Greenland during the strong TNn years based on the composite analysis (Fig. 5e). The maximum cooling is located over the western parts of GrIS (<−5°C), and the least cooling occurs over the ice-free coastal region (Table 3). This is opposite to the interannual variation of TXx, which exhibits more profound change over the ice-free marginal region than over the ice sheet. Besides, these characteristics associated with TNn are generally observed in the variation of TN10p that is defined as the percentage of days when daily minimum temperature < 10th percentile, in terms of interannual variability and long-term trend (Figs. S1 and S4).
c. Precipitation-related extremes from 1958 to 2019
During the past six decades, Rx1day on average exhibits an increasing trend (0.26 mm day−1 decade−1) over Greenland (Fig. 6a). In contrast to the accelerated increase in extreme temperatures after 1990, there is a significant increasing trend in Rx1day before 1990 (0.69 mm day−1 decade−1), whereas no obvious trend is observed afterward. However, change in extreme precipitation shows high regional variability (Fig. 6c). During 1958–2019, the maximum increasing trend in Rx1day is observed over the northeast GrIS (0.52 mm day−1 decade−1; Table 2), followed by the central-east GrIS and ice-free coastal regions (0.37–0.39 mm day−1 decade−1). In contrast, Rx1day experiences a weakening trend over the southeast and central-west GrIS, though not passing the significance test at the 95% level. During the recent three decades (1990–2019), Rx1day shows a decreasing trend over the north and southwest GrIS (−0.56 and −1.29 mm day−1 decade−1), which is opposite to the long-term upward trend (Table 2). In contrast, the central-west and northeast GrIS experiences a larger increasing trend after 1990. However, the trends during 1990–2019 do not pass the statistical significance test, which may be attributed to the large discrepancy in trends across each subregion. The small-scale variability along the coastal regions may be partially linked with the uneven distribution of precipitation due to the complex topography. On the other hand, over the regions where the interannual variability of precipitation is larger, there is generally no significant long-term trend, which hence contributes to the uneven distribution of the insignificant trend (e.g., in the interior part). Additionally, the long-term trend of extreme precipitation measured by R95 (annual total precipitation when precipitation amount >95th percentile) aligns with the change in Rx1day (Figs. S2 and S5).
(a) The time series of Rx1day averaged over Greenland (dotted lines) from 1958 to 2019 and its 9-yr running mean (black line). The green line shows the long-term linear trend. (b) The time series of the standardized detrended Rx1day. (c) The distribution of the trend in annual Rx1day during 1958–2019. (d) Composite difference in Rx1day between strong and weak Rx1day years [i.e., red and blue dots in (b)]. The dotted areas in (c) and (d) are statistically significant at the 95% level. All data used here are calculated from RACMO2.3p2 results.
Citation: Journal of Climate 36, 14; 10.1175/JCLI-D-22-0488.1
Regarding the interannual variability, we define 9 years (1964, 1967, 1972, 1976, 1984, 1986, 1997, 2008, 2017) and 10 years (1958, 1962, 1963, 1966, 1968, 2004, 2006, 2010, 2011, 2019) as strong and weak Rx1day years, according to the ±1 standard deviation of the detrended Rx1day, respectively. Based on the composite analysis (Table 3), the largest difference between the strong and weak Rx1day years occurs in the ice-free coastal region (10.2 mm day−1), followed by the southern and western GrIS where Rx1day is ∼8–9 mm day−1 higher during the strong Rx1day years. In northern GrIS, the difference in Rx1day between strong and weak Rx1day years is 1.9–2.8 mm day−1, accounting for 10%–29% of multiyear averaged Rx1day over there. This is opposite to the pattern of the long-term trend, which shows a significant wetting trend over the northeast GrIS and a drying trend over southwest GrIS (Figs. 6c,d). Similar spatial differences are also observed in the interannual variation of R95, though the extreme years defined are different (Figs. S2 and S5).
4. Discussion
In this section, we attempt to explore the large-scale atmospheric circulations that provide favorable conditions for temperature and precipitation extremes from 1958 to 2019. For the temperature-related extremes, we note that the long-term increase of TXx and the weakening of TNn in Greenland align with global warming since the 1990s (Figs. 4a and 5a), which implies the potential role of global warming in the long-term increasing (decreasing) TXx (TNn) as small increase in temperature can cause large upticks in the probability of extreme weather events (Diffenbaugh 2020). Moreover, the variations of extreme events are often closely linked with abnormal atmospheric circulations (Herring et al. 2018; Agel et al. 2019). Thus, we then examine the variations of large-scale atmosphere circulations that contribute to changes in extreme climate events over Greenland. As shown in Fig. 7a, atmospheric circulations in summer when annual TXx generally occurs experience significant changes since 1990, with an increasing trend in geopotential height at 500 hPa over Greenland, indicating an intensified Greenland blocking and the associated anticyclonic circulation. Composite differences between the strong and weak TXx years show a similar pattern in atmospheric circulations (Fig. 7b). The intensified Greenland blocking (a warm high pressure system) is conducive to the maintenance of warm air over Greenland and favors the development of extreme warm events (Hanna et al. 2021; Pfahl and Wernli 2012; Schaller et al. 2018), which partly explains the increasing TXx after 1990 and higher TXx during the strong extreme years.
The linear trend of (a) 500-hPa geopotential height (shadings; gpm decade−1) and 850-hPa winds (vectors; m s−1 decade−1) in summer and (c) 100-hPa geopotential height (gpm decade−1) in winter during the period of 1990–2019 based on EAR5 reanalysis. (b),(d) As in (a) and (c), but for the composite differences between the strong and weak TXx and TNn years, respectively, during the period of 1959–2019 based on EAR5 reanalysis. The climatological mean 100-hPa geopotential height is shown as cyan lines (gpm) in (c) and (d). The trends/differences for geopotential height and winds that are statistically significant at the 95% level are shown as white dotted and gray shaded area, respectively.
Citation: Journal of Climate 36, 14; 10.1175/JCLI-D-22-0488.1
Furthermore, the polar vortex, which is a low pressure area with a wide expanse of swirling cold air, experiences a significant weakening during the period of 1990–2019 in winter (Fig. 7c), when annual minimum temperature in Greenland generally occurs. The weakening polar vortex could cause weaker cold air in the Arctic and is unfavorable for southward penetration of cold air, contributing to the increasing TNn (i.e., weaker extreme cold events) over Greenland from 1990 to 2019. In contrast, the composite differences between the strong and weak TNn years illustrate a significantly stronger polar vortex, with a low pressure anomaly over the Arctic (Fig. 7d). The intensified polar vortex provides favorable conditions for the occurrence of extreme cold events over Greenland.
Next, we explore the atmospheric circulations linked with the long-term trend and interannual variability of extreme precipitation over Greenland. Figure 8 shows the linear trend of atmospheric circulations during 1958–2019. At upper levels, zonal wind at 300 hPa is intensified along the jet axis from 1958 to 2019 (Fig. 8a), especially at the east of 60°W, suggesting an eastward shift of the jet stream. This could alter the tracks of storms that generally move along the jet axis (Coumou and Rahmstorf 2012), leading to fewer storms entering southern Greenland (Lewis et al. 2019). Together with an intensified southward moisture transport (Fig. 8b), these suggest unfavorable conditions for precipitation extremes over southern Greenland. In contrast, an anomalous anticyclonic water vapor flow appears over the Arctic Ocean, which creates favorable conditions for precipitation extremes via transporting water vapor from Baffin Bay and the Greenland Sea to northern Greenland (Fig. 8b).
The linear trend of (a) zonal wind at 300 hPa (shading; m s−1 decade−1) and its climatology (blue lines; m s−1), (b) vertically integrated water vapor flux from surface to 300 hPa (kg m−1 s−1 decade−1) during the period of 1958–2019 based on EAR5 reanalysis. (c),(d) As in (a) and (b), but for the composite differences between the strong and weak Rx1day years during the period of 1958–2019 based on EAR5 reanalysis. Dotted values in (a) and (c) and shaded areas in (b) and (d) are significant at the 95% confidence level.
Citation: Journal of Climate 36, 14; 10.1175/JCLI-D-22-0488.1
For the interannual variability, we calculate the differences in atmosphere circulations between strong and weak Rx1day years using the composite analysis. At the upper troposphere (300 hPa), a westerly wind anomaly dominates the south of Greenland and an easterly wind anomaly prevails over the midlatitudes of the North Atlantic, indicating a northward shift of the polar jet stream and the associated storm tracks (Fig. 8c). This favors water vapor traveling northward to southern Greenland. Furthermore, according to the integrated water vapor fluxes, an anomalous cyclonic circulation over Greenland brings abundant moisture from the North Atlantic and Baffin Bay to southeast and southwest Greenland (Fig. 8d). These conditions contribute to enhanced precipitation over southern Greenland during strong Rx1day years.
5. Conclusions
Greenland has experienced multiple extreme weather events in recent decades that led to significant melting of the ice sheet and caused large damages to coastal communities. In this study, we assess the performance of RACMO2.3p2 in depicting extreme climate events over Greenland against available observations from staffed stations and AWSs, and then investigate the spatiotemporal variations of extreme temperature- and precipitation-related extremes over Greenland during 1958–2019.
Compared with 176 in situ observations over Greenland, the RACMO2.3p2 model reproduces the observed extreme temperatures/precipitations over Greenland well, in terms of climatological distribution, interannual variation, and long-term trend. Based on the RACMO2.3p2, our results illustrate that annual maximum temperature exhibit a statistically significant increasing trend (∼0.13°C decade−1) during 1958–2019, with the most intense change occurring over the ice-free coastal region and the northern GrIS. In contrast, there is a weakening trend (−0.24°C decade−1) in annual minimum temperature over Greenland, especially after the 1990s (−1.24°C decade−1), and the largest declining trend is observed over the southwestern GrIS during the later period. In terms of interannual variability, the distributions of composite differences in TXx/TNn between strong and weak extreme years share large similarities with the distributions of long-term trends for TXx/TNn. Extreme precipitation events measured by annual maximum daily precipitation amount shows a significant increasing trend over northeastern Greenland (0.52 mm day−1 decade−1). In terms of the difference between strong and weak extreme years, the most intense change in Rx1day occurs in the ice-free coastal region and southern Greenland.
Further analysis indicates that apart from global warming from increased greenhouse gases, the increasing trend of TXx during 1990–2019 is linked with the enhanced Greenland blocking, which also accounts for the higher daily maximum temperature during the strong TXx years relative to the weak TXx years. The decreasing TNn during 1990–2019 partially benefits from the weakening of the polar vortex. In contrast, the intensified polar vortex provides favorable conditions for the occurrence of extreme cold events over Greenland in terms of interannual variability. For the extreme precipitation during 1959–2019, an eastward shift of the jet center and associated storm tracks contribute to the weakened precipitation extremes over southern Greenland. Compared with the weak Rx1day years, a poleward shift of the jet stream is partially responsible for the enhanced extreme precipitation over Greenland during the strong Rx1day years.
However, there are some limitations to be considered. Given the spatially scattered observational sites, we investigate the spatiotemporal features of extreme events over Greenland using the RACMO2.3p2 model. The quantitative results reported here may be sensitive to the regional climate model and lateral climatic forcing used (i.e., reanalysis datasets), but the long-term trend may largely hold unchanged, as temperature shows a consistently increasing trend across different reanalysis datasets, though with different warming rate (Hersbach et al. 2020). Notably, the dominating circulation factor for extreme events may vary with different extreme indices and time intervals. Moreover, the mechanisms reported here are just favorable environmental conditions that support the long-term trend and interannual variation of extreme events, which cannot be simply applied to local sites or a single event that should be examined via computing climatic anomalies, for example, between the 15 days before an extreme event at a site and the climatological mean. Additionally, Greenland compound extremes, i.e., multiple events occurring simultaneously or in close sequence (e.g., concurrent heatwaves in multiple breadbaskets), may lead to larger risks to the GrIS, which deserves additional investigations in future works. Nevertheless, we for the first time provide a possible scenario on the variations of temperature- and precipitation-related extremes over Greenland, which may advance our understanding of Greenland’s distinct climatic regimes, guide the establishment of future observational networks, and provide insights into climate risks across the Arctic.
Acknowledgments.
We sincerely thank the three reviewers for their extremely valuable and constructive comments, which have helped us greatly improve the quality of our study. This study was funded by the National Natural Science Foundation of China (41975120) and the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA2010030807). B. Noël was funded by the NWO VENI Grant VI.Veni.192.019.
Data availability statement.
All observational datasets required for the analysis are freely available at their official websites. Codes and simulation datasets are available upon request to the corresponding author.
REFERENCES
Agel, L., M. Barlow, F. Colby, H. Binder, J. L. Catto, A. Hoell, and J. Cohen, 2019: Dynamical analysis of extreme precipitation in the US northeast based on large-scale meteorological patterns. Climate Dyn., 52, 1739–1760, https://doi.org/10.1007/s00382-018-4223-2.
Alexander, L. V., and Coauthors, 2006: Global observed changes in daily climate extremes of temperature and precipitation. J. Geophys. Res., 111, D05109, https://doi.org/10.1029/2005JD006290.
Bennartz, R., and Coauthors, 2013: July 2012 Greenland melt extent enhanced by low-level liquid clouds. Nature, 496, 83–86, https://doi.org/10.1038/nature12002.
Box, J. E., X. Fettweis, J. C. Stroeve, M. Tedesco, D. K. Hall, and K. Steffen, 2012: Greenland Ice Sheet albedo feedback: Thermodynamics and atmospheric drivers. Cryosphere, 6, 821–839, https://doi.org/10.5194/tc-6-821-2012.
Coumou, D., and S. Rahmstorf, 2012: A decade of weather extremes. Nat. Climate Change, 2, 491–496, https://doi.org/10.1038/nclimate1452.
Culberg, R., D. M. Schroeder, and W. Chu, 2021: Extreme melt season ice layers reduce firn permeability across Greenland. Nat. Commun., 12, 2336, https://doi.org/10.1038/s41467-021-22656-5.
Cullather, R. I., and Coauthors, 2020: Anomalous circulation in July 2019 resulting in mass loss on the Greenland Ice Sheet. Geophys. Res. Lett., 47, e2020GL087263, https://doi.org/10.1029/2020GL087263.
Diffenbaugh, N. S., 2020: Verification of extreme event attribution: Using out-of-sample observations to assess changes in probabilities of unprecedented events. Sci. Adv., 6, eaay2368, https://doi.org/10.1126/sciadv.aay2368.
Fausto, R. S., and Coauthors, 2021: Programme for Monitoring of the Greenland Ice Sheet (PROMICE) automatic weather station data. Earth Syst. Sci. Data, 13, 3819–3845, https://doi.org/10.5194/essd-13-3819-2021.
Fettweis, X., and Coauthors, 2020: GrSMBMIP: Intercomparison of the modelled 1980–2012 surface mass balance over the Greenland Ice Sheet. Cryosphere, 14, 3935–3958, https://doi.org/10.5194/tc-14-3935-2020.
Fischer, E. M., and R. Knutti, 2015: Anthropogenic contribution to global occurrence of heavy-precipitation and high-temperature extremes. Nat. Climate Change, 5, 560–564, https://doi.org/10.1038/nclimate2617.
Gardner, A. S., and Coauthors, 2013: A reconciled estimate of glacier contributions to sea level rise: 2003 to 2009. Science, 340, 852–857, https://doi.org/10.1126/science.1234532.
Guo, D., Y. Zhang, X. Gao, N. Pepin, and J. Sun, 2021: Evaluation and ensemble projection of extreme high and low temperature events in China from four dynamical downscaling simulations. Int. J. Climatol., 41 (Suppl. 1), E1252–E1269, https://doi.org/10.1002/joc.6765.
Hanna, E., and Coauthors, 2008: Increased runoff from melt from the Greenland Ice Sheet: A response to global warming. J. Climate, 21, 331–341, https://doi.org/10.1175/2007JCLI1964.1.
Hanna, E., and Coauthors, 2014: Atmospheric and oceanic climate forcing of the exceptional Greenland Ice Sheet surface melt in summer 2012. Int. J. Climatol., 34, 1022–1037, https://doi.org/10.1002/joc.3743.
Hanna, E., and Coauthors, 2020: Mass balance of the ice sheets and glaciers—Progress since AR5 and challenges. Earth-Sci. Rev., 201, 102976, https://doi.org/10.1016/j.earscirev.2019.102976.
Hanna, E., and Coauthors, 2021: Greenland surface air temperature changes from 1981 to 2019 and implications for ice-sheet melt and mass-balance change. Int. J. Climatol., 41 (Suppl. 1), E1336–E1352, https://doi.org/10.1002/joc.6771.
Hawley, R. L., Z. R. Courville, L. M. Kehrl, E. R. Lutz, E. C. Osterberg, T. B. Overly, and G. J. Wong, 2014: Recent accumulation variability in northwest Greenland from ground-penetrating radar and shallow cores along the Greenland inland traverse. J. Glaciol., 60, 375–382, https://doi.org/10.3189/2014JoG13J141.
Herring, S. C., N. Christidis, A. Hoell, J. P. Kossin, C. J. Schreck III, and P. A. Stott, 2018: Introduction to Explaining Extreme Events of 2016 from a Climate Perspective [in “Explaining Extreme Events of 2016 from a Climate Perspective”]. Bull. Amer. Meteor. Soc., 99 (1), S54–S59, https://doi.org/10.1175/BAMS-D-17-0284.1.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hofer, S., A. J. Tedstone, X. Fettweis, and J. L. Bamber, 2017: Decreasing cloud cover drives the recent mass loss on the Greenland Ice Sheet. Sci. Adv., 3, e1700584, https://doi.org/10.1126/sciadv.1700584.
Hofer, S., A. J. Tedstone, X. Fettweis, and J. L. Bamber, 2019: Cloud microphysics and circulation anomalies control differences in future Greenland melt. Nat. Climate Change, 9, 523–528, https://doi.org/10.1038/s41558-019-0507-8.
Huai, B. J., M. R. van den Broeke, C. H. Reijmer, and J. Cappellen, 2021: Quantifying rainfall in Greenland: A combined observational and modeling approach. J. Appl. Meteor. Climatol., 60, 1171–1188, https://doi.org/10.1175/JAMC-D-20-0284.1.
Huai, B. J., M. R. van den Broeke, C. H. Reijmer, and B. Noël, 2022: A daily 1-km resolution Greenland rainfall climatology (1958–2020) from statistical downscaling of a regional atmospheric climate model. J. Geophys. Res. Atmos., 127, e2022JD036688, https://doi.org/10.1029/2022JD036688.
IMBIE Team, 2020: Mass balance of the Greenland Ice Sheet from 1992 to 2018. Nature, 579, 233–239, https://doi.org/10.1038/s41586-019-1855-2.
Karl, T. R., N. Nicholls, and A. Ghazi, 1999: CLIVAR/GCOS/WMO Workshop on Indices and Indicators for Climate Extremes: Workshop summary. Climatic Change, 42, 3–7, https://doi.org/10.1023/A:1005491526870.
Kharin, V. V., F. W. Zwiers, X. Zhang, and G. C. Hegerl, 2007: Changes in temperature and precipitation extremes in the IPCC ensemble of global coupled model simulations. J. Climate, 20, 1419–1444, https://doi.org/10.1175/JCLI4066.1.
Kim, Y.-H., S.-K. Min, X. Zhang, J. Sillmann, and M. Sandstad, 2020: Evaluation of the CMIP6 multi-model ensemble for climate extreme indices. Wea. Climate Extremes, 29, 100269, https://doi.org/10.1016/j.wace.2020.100269.
Koenig, L. S., and Coauthors, 2016: Annual Greenland accumulation rates (2009–2012) from airborne snow radar. Cryosphere, 10, 1739–1752, https://doi.org/10.5194/tc-10-1739-2016.
Lamarche-Gagnon, G., and Coauthors, 2019: Greenland melt drives continuous export of methane from the ice-sheet bed. Nature, 565, 73–77, https://doi.org/10.1038/s41586-018-0800-0.
Lenaerts, J. T. M., M. R. van den Broeke, W. J. van de Berg, E. van Meijgaard, and P. Kuipers Munneke, 2012: A new, high-resolution surface mass balance map of Antarctica (1979–2010) based on regional atmospheric climate modeling. Geophys. Res. Lett., 39, L04501, https://doi.org/10.1029/2011GL050713.
Lewis, G., and Coauthors, 2019: Recent precipitation decrease across the western Greenland Ice Sheet percolation zone. Cryosphere, 13, 2797–2815, https://doi.org/10.5194/tc-13-2797-2019.
Lott, N., R. Vose, S. A. Del Greco, T. F. Ross, S. J. Worley, and J. Comeaux, 2008: The integrated surface database: Partnerships and progress. 24th Conf. on Interactive Information and Processing Systems, New Orleans, LA, Amer. Meteor. Soc., 3B.5, https://ams.confex.com/ams/88Annual/techprogram/paper_131387.htm.
Mankoff, K. D., and Coauthors, 2021: Greenland Ice Sheet mass balance from 1840 through next week. Earth Syst. Sci. Data, 13, 5001–5025, https://doi.org/10.5194/essd-13-5001-2021.
Mattingly, K. S., T. L. Mote, X. Fettweis, D. van As, K. Van Tricht, S. Lhermitte, C. Pettersen, and R. S. Fausto, 2020: Strong summer atmospheric rivers trigger Greenland Ice Sheet melt through spatially varying surface energy balance and cloud regimes. J. Climate, 33, 6809–6832, https://doi.org/10.1175/JCLI-D-19-0835.1.
Mernild, S. H., and Coauthors, 2015: Greenland precipitation trends in a long-term instrumental climate context (1890–2012): Evaluation of coastal and ice core records. Int. J. Climatol., 35, 303–320, https://doi.org/10.1002/joc.3986.
Moon, T. A., A. S. Gardner, B. Csatho, I. Parmuzin, and M. A. Fahnestock, 2020: Rapid reconfiguration of the Greenland Ice Sheet coastal margin. J. Geophys. Res. Earth Surf., 125, e2020JF005585, https://doi.org/10.1029/2020JF005585.
Moon, T. A., and Coauthors, 2021: Arctic Report Card 2021: Rapid and pronounced warming continues to drive the evolution of the Arctic environment. NOAA Tech. Rep., 126 pp., https://arctic.noaa.gov/Portals/7/ArcticReportCard/Documents/ArcticReportCard_full_report2021.pdf.
Mote, T. L., 2007: Greenland surface melt trends 1973–2007: Evidence of a large increase in 2007. Geophys. Res. Lett., 34, L22507, https://doi.org/10.1029/2007GL031976.
Mouginot, J., and Coauthors, 2019: Forty-six years of Greenland Ice Sheet mass balance from 1972 to 2018. Proc. Natl. Acad. Sci. USA, 116, 9239–9244, https://doi.org/10.1073/pnas.1904242116.
Nghiem, S. V., and Coauthors, 2012: The extreme melt across the Greenland Ice Sheet in 2012. Geophys. Res. Lett., 39, L20502, https://doi.org/10.1029/2012GL053611.
Niwano, M., J. E. Box, A. Wehrlé, B. Vandecrux, W. T. Colgan, and J. Cappelen, 2021: Rainfall on the Greenland Ice Sheet: Present‐day climatology from a high‐resolution non‐hydrostatic polar regional climate model. Geophys. Res. Lett., 48, e2021GL092942, https://doi.org/10.1029/2021GL092942.
Noël, B., W. J. van de Berg, S. Lhermitte, and M. R. van den Broeke, 2019: Rapid ablation zone expansion amplifies north Greenland mass loss. Sci. Adv., 5, eaaw0123, https://doi.org/10.1126/sciadv.aaw0123.
Pfahl, S., and H. Wernli, 2012: Quantifying the relevance of atmospheric blocking for co-located temperature extremes in the Northern Hemisphere on (sub-)daily time scales. Geophys. Res. Lett., 39, L12807, https://doi.org/10.1029/2012GL052261.
Preece, J. R., L. J. Wachowicz, T. L. Mote, M. Tedesco, and X. Fettweis, 2022: Summer Greenland blocking diversity and its impact on the surface mass balance of the Greenland Ice Sheet. J. Geophys. Res. Atmos., 127, e2021JD035489, https://doi.org/10.1029/2021JD035489.
Sasgen, I., and Coauthors, 2020: Return to rapid ice loss in Greenland and record loss in 2019 detected by the GRACE-FO satellites. Commun. Earth Environ., 1, 8, https://doi.org/10.1038/s43247-020-0010-1.
Sasgen, I., A. Salles, M. Wegmann, B. Wouters, X. Fettweis, B. P. Y. Noël, and C. Beck, 2022: Arctic glaciers record wavier circumpolar winds. Nat. Climate Change, 12, 249–255, https://doi.org/10.1038/s41558-021-01275-4.
Schaller, N., J. Sillmann, J. Anstey, E. M. Fischer, C. M. Grams, and S. Russo, 2018: Influence of blocking on northern European and western Russian heatwaves in large climate model ensembles. Environ. Res. Lett., 13, 054015, https://doi.org/10.1088/1748-9326/aaba55.
Seneviratne, S. I., and Coauthors, 2021: Weather and climate extreme events in a changing climate. Climate Change 2021: The Physical Science Basis, V. Masson-Delmotte et al., Eds., Cambridge University Press, 1513–1766.
Sillmann, J., V. V. Kharin, X. Zhang, F. W. Zwiers, and D. Bronaugh, 2013: Climate extremes indices in the CMIP5 multimodel ensemble: Part 1. Model evaluation in the present climate. J. Geophys. Res. Atmos., 118, 1716–1733, https://doi.org/10.1002/jgrd.50203.
Stark, D., S. van Hal, D. Marriott, J. Ellis, and J. Harkness, 2007: Irritable bowel syndrome: A review on the role of intestinal protozoa and the importance of their detection and diagnosis. Int. J. Parasitol., 37, 11–20, https://doi.org/10.1016/j.ijpara.2006.09.009.
Steffen, K., J. E. Box, and W. Abdalati, 1996: Greenland Climate Network: GC-Net. Glaciers, ice sheets and volcanoes: Tribute to M. Meier, USACE CRREL Special Rep. 96-97, 98–103.
Tedesco, M., and X. Fettweis, 2020: Unprecedented atmospheric conditions (1948–2019) drive the 2019 exceptional melting season over the Greenland Ice Sheet. Cryosphere, 14, 1209–1223, https://doi.org/10.5194/tc-14-1209-2020.
Tedesco, M., J. E. Box, T. S. Jensen, T. Mote, A. K. Rennermalm, L. C. Smith, R. S. W. van de Wal, and J. Wahr, 2014: Greenland Ice Sheet [in “State of the Climate in 2002”]. Bull. Amer. Meteor. Soc., 95 (7), S136–S138, https://doi.org/10.1175/2014BAMSStateoftheClimate.1.
Tedesco, M., S. Doherty, X. Fettweis, P. Alexander, J. Jeyaratnam, and J. Stroeve, 2016: The darkening of the Greenland Ice Sheet: Trends, drivers, and projections (1981–2100). Cryosphere, 10, 477–496, https://doi.org/10.5194/tc-10-477-2016.
Undén, P., and Coauthors, 2002: HIRLAM-5 scientific documentation. Swedish Meteorological and Hydrological Institute Doc., 146 pp., https://repositorio.aemet.es/bitstream/20.500.11765/6323/1/HIRLAMSciDoc_Dec2002.pdf.
Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131, 2961–3012, https://doi.org/10.1256/qj.04.176.
van den Broeke, M. R., E. M. Enderlin, I. M. Howat, P. Kuipers Munneke, B. P. Y. Noël, W. J. van de Berg, E. van Meijgaard, and B. Wouters, 2016: On the recent contribution of the Greenland Ice Sheet to sea level change. Cryosphere, 10, 1933–1946, https://doi.org/10.5194/tc-10-1933-2016.
Van Meijgaard, E., L. H. van Ulft, W. J. van de Berg, F. C. Bosveld, B. J. J. M. van den Hurk, G. Lenderink, and A. P. Siebesma, 2008: The KNMI Regional Atmospheric Climate Model RACMO version 2.1. KNMI Tech. Rep. 302, 50 pp., https://cdn.knmi.nl/knmi/pdf/bibliotheek/knmipubTR/TR302.pdf.
Walsh, J. E., T. J. Ballinger, E. S. Euskirchen, E. Hanna, J. Mård, J. E. Overland, H. Tangen, and T. Vihma, 2020: Extreme weather and climate events in northern areas: A review. Earth-Sci. Rev., 209, 103324, https://doi.org/10.1016/j.earscirev.2020.103324.
Walter, F., J. Chaput, and M. P. Lüthi, 2014: Thick sediments beneath Greenland’s ablation zone and their potential role in future ice sheet dynamics. Geology, 42, 487–490, https://doi.org/10.1130/G35492.1.
Wei, T., Q. Yan, and M. Ding, 2019: Distribution and temporal trends of temperature extremes over Antarctica. Environ. Res. Lett., 14, 084040, https://doi.org/10.1088/1748-9326/ab33c1.
Wei, T., B. Noël, M. Ding, and Q. Yan, 2022: Spatiotemporal variations of extreme events in surface mass balance over Greenland during 1958–2019. Int. J. Climatol., 42, 8008–8023, https://doi.org/10.1002/joc.7689.
White, P. W., Ed., 2001: Physical processes (CY23R4). ECMWF IFS Doc., 166 pp., https://www.ecmwf.int/sites/default/files/elibrary/2003/77030-ifs-documentation-cy23r4-part-iv-physical-processes_1.pdf.
Wong, G. J., E. C. Osterberg, R. L. Hawley, Z. R. Courville, D. G. Ferris, and J. A. Howley, 2015: Coast-to-interior gradient in recent northwest Greenland precipitation trends (1952–2012). Environ. Res. Lett., 10, 114008, https://doi.org/10.1088/1748-9326/10/11/114008.
Zhang, Q., B. Huai, M. R. van den Broeke, J. Cappelen, M. Ding, Y. Wang, and W. Sun, 2022: Temporal and spatial variability in contemporary Greenland warming (1958–2020). J. Climate, 35, 2755–2767, https://doi.org/10.1175/JCLI-D-21-0313.1.
