Automatic Weather Station Observations Research Articles

Primary Impact Evaluation of Surface Temperature Observations for Microwave Temperature Sounding Data Assimilation over Land

Observations from the Advanced Microwave Sounding Unit-A (AMSU-A) onboard polar-orbiting satellites are considered to be the most effective satellite data in terms of obviously reducing operational prediction errors. However, there are still significant difficulties in the application of AMSU-A low-level channel data assimilation over land. One of them is the inaccurate surface skin temperature (SKT) of the background on land areas, which leads to significant uncertainty in the accuracy of simulating brightness temperature (BT) in these channels. Therefore, improving the accuracy of SKT in the background field is a direct way to improve the assimilation effect of these low-level channel data over land. In this study, both high-spatio-temporal-resolution automatic weather station (AWS) observation data from China in September 2021 and the AMSU-A observation data from NOAA-15/18/19 and MetOp-A were used. Based on the Advanced Research version of the Weather Research and Forecast model (WRF-ARW) and Gridpoint Statistical Interpolation (GSI) assimilation system, we first analyzed the differences in SKT between AWS observations and model simulations and then attempted to directly replace the simulated SKT with the observation data. On this basis, the differences in BT simulation effects over the land area of Southwest China before and after replacement were meticulously analyzed and compared. In addition, the impacts of SKT replacement in areas with different terrain elevations and in cloudy areas were also evaluated. The results indicate that the SKTs of background fields were generally lower than the surface observations, whereas the diurnal variation in SKT was not well simulated. After replacing the SKT of the background field with station observations, the BT differences between the observation and background (O–B, observation minus background) were remarkably reduced, especially for channels 3–5 and 15 of the AMSU-A. The volume of data passing the GSI quality control significantly increased, and the standard deviation of O–B decreased. Further analysis showed that the improvement effect was better in areas at an elevation above 1600 m. Moreover, introducing SKT observations leads to a significant and stable improvement over BT simulations in cloudy areas over land.

Effects of assimilation of YOPP‐SH additional radiosonde observations on analyses and forecasts over Antarctica in austral summer

AbstractRadiosonde observations over Antarctica and the surrounding oceans were enhanced during the Year of Polar Prediction in the Southern Hemisphere (YOPP‐SH) summer Special Observing Period (SOP). Observing System Experiments (OSEs) were conducted in a continuous cycling framework using the Weather Research and Forecasting (WRF) Model and its data assimilation system. Routinely available observations were assimilated in the CTL (control) experiment, and special radiosonde observations from the YOPP‐SH SOP were additionally assimilated in the YOPP experiment. The results were compared to investigate the effects of additional radiosonde observations on analyses and forecasts over and around Antarctica. Verifications against ERA5 re‐analysis, radiosonde observations, and Automatic Weather Station (AWS) observations show overall positive effects of additional radiosonde observations. These positive effects are most noticeable in temperature at lower levels at earlier forecast lead times; afterward, wind forecast improvements at upper levels are the most noticeable. Although routine and special radiosonde observations are concentrated over the eastern and coastal regions of Antarctica (compared to the western and inland regions), the effects of the extra data spread in longitudinal and latitudinal directions; therefore, the effects on the forecasts are not limited to only the areas near the radiosonde observations. A case study reveals how cyclone forecasts are improved through the assimilation of the additional YOPP‐SH SOP radiosonde observations. This study provides insights into future observation strategies in Antarctica, such as horizontal/vertical observation locations, observation variables, and so forth to maximize effects of new observations on forecasts over Antarctica.

The AntAWS dataset: a compilation of Antarctic automatic weather station observations

Abstract. A new meteorological dataset derived from records of Antarctic automatic weather stations (here called the AntAWS dataset) at 3 h, daily and monthly resolutions including quality control information is presented here. This dataset integrates the measurements of air temperature, air pressure, relative humidity, and wind speed and direction from 267 Antarctic AWSs obtained from 1980 to 2021. The AWS spatial distribution remains heterogeneous, with the majority of instruments located in near-coastal areas and only a few inland on the East Antarctic Plateau. Among these 267 AWSs, 63 have been operating for more than 20 years and 27 of them in excess of more than 30 years. Of the five meteorological parameters, the measurements of air temperature have the best continuity and the highest data integrity. The overarching aim of this comprehensive compilation of AWS observations is to make these data easily and widely accessible for efficient use in local, regional and continental studies; it may be accessed at https://doi.org/10.48567/key7-ch19 (Wang et al., 2022). This dataset is invaluable for improved characterization of the surface climatology across the Antarctic continent, to improve our understanding of Antarctic surface snow–atmosphere interactions including precipitation events associated with atmospheric rivers and to evaluate regional climate models or meteorological reanalysis products.

Integration of shapley additive explanations with random forest model for quantitative precipitation estimation of mesoscale convective systems

Mesoscale convective cloud systems have a small horizontal scale and a short lifetime, which brings great challenges to quantitative precipitation estimation (QPE) by satellite remote sensing. Combining machine learning models and geostationary satellite spectral information is an effective method for the QPE of mesoscale convective cloud, while the interpretability of machine learning model outputs remains unclear. In this study, based on Himawari-8 data, high-density automatic weather station observations, and reanalysis data over the North China Plain, a random forest (RF) machine learning model of satellite-based QPE was established and verified. The interpretation of the output of the RF model of satellite-based QPE was further explored by using the Shapley Additive Explanations (SHAP) algorithm. Results showed that the correlation coefficient between the predicted and observed precipitation intensity of the RF model was .64, with a root-mean-square error of .27 mm/h. The importance ranking obtained by SHAP model is completely consistent with the outputs of random forest importance function. This SHAP method can display the importance ranking of global features with positive/negative contribution values (e.g., current precipitation, column water vapor/black body temperature, cloud base height), and can visualize the marginal contribution values of local features under interaction. Therefore, combining the RF and SHAP methods provides a valuable way to interpret the output of machine learning models for satellite-based QPE, as well as an important basis for the selection of input variables for satellite-based QPE.

Channelling flows in the Hunter Valley

The Hunter Valley is well known for the strong westerly winds in winter and elevated fire danger arising from hot and dry north-westerly winds in summer. These hazards are closely related to the valley channelling in the region, and the connection between the two has been an interest to weather forecasters, emergency service personnel, and the aviation industry. In this paper, the climatology of valley winds is constructed to identify the dominant types of channelling in the Hunter Valley and their preferred directions using the 10-year data of Automatic Weather Station observations, upper air sounding and ERA5 reanalysis data between July 2010 and June 2020. Particular attention is given to the conceptual model of pressure-driven channelling of westerlies in winter and its mechanics, as the climatology shows that it is the main cause of wind-warning conditions in the Hunter.

Investigation of Atmospheric Dynamic and Thermodynamic Structures of Typhoon Sinlaku (2020) from High-Resolution Dropsonde and Two-Way Rawinsonde Measurements

Profiling the vertical atmospheric structure for typhoons remains challenging. Here, the atmospheric dynamic and thermodynamic structures were investigated during the passage of Typhoon Sinlaku (2020) over Xisha Islands in the South China Sea for the period 28 July to 2 August 2020, mainly based on two-way rawinsonde and dropsonde measurements in combination with surface-based automatic weather station observations, disdrometer measurements, and Himawari-8 geostationary satellite images. The study period was divided to three stages: the formation stage of tropical depression (pre-TD), tropical depression (TD), and tropical storm (TS). The wind speed and local vertical wind shear reached the maximum value at 3 km above mean sea level (AMSL) before the typhoon approached the Xisha islands. Pseudo-equivalent potential temperature (θse) was found to decrease with the altitude below 2 km AMSL; temperature inversions occurred frequently within this altitude range, particularly during the TS stage. This seemed a typical capping inversion that indicated a downward motion above 2 km AMSL. The temperature increased slightly with the development of Typhoon Sinlaku (2020) at altitudes of 8–10 km AMSL. This indicated that our observations presumably captured the air mass warmed by the condensation, which was a good signature of an upper air in the tropical cyclone. In addition, wind speed (particularly in the lower stratosphere), specific humidity, and equivalent potential temperature escalated significantly when the tropical depression strengthened into Typhoon Sinlaku (2020), which indicated that the typhoon constantly obtained energy from the sea surface during its passage over the study region. The thermodynamic and dynamic structures of atmosphere advance our understanding of the inner structure of typhoons during the different evolutionary stages.

Predicting Frigid Mixed‐Phase Clouds for Pristine Coastal Antarctica

AbstractSupercooled water is common in the clouds near coastal Antarctica and occasionally occurs at temperatures at or below −30°C. Yet the ice physics in most regional and global numerical models will glaciate out these clouds. This presents a challenge for the simulation of highly supercooled clouds that were observed at McMurdo, Antarctica during the Atmospheric Radiation Measurement (ARM) West Antarctic Radiation Experiment (AWARE) project during 2015–2017. The polar optimized version of the Weather Research and Forecasting model (Polar WRF) with the recently developed two‐moment P3 microphysics scheme was used to simulate observed supercooled liquid water cases during March and November 2016. Nudging of the simulations to observed rawinsonde profiles and Antarctic automatic weather station observations provided increased realism and much greater cloud water amounts. Sensitivity tests that adjust the ice physics for extremely low ice nucleating particle (INP) concentrations decrease cloud ice and increases the cloud liquid water closer to observed amounts. In these tests, a liquid layer near cloud top is simulated, in agreement with observations. Accurate representation of INP concentrations appears to be critical for the simulation of coastal Antarctic clouds.

Thunderstorms and extreme rainfall in south Norfolk, 16 August 2020: meteorological analysis

A multicell thunderstorm cluster affected parts of the Breckland region in south Norfolk on the afternoon of 16 August 2020, producing as much as 240mm of rain. A detailed mesoanalysis is constructed to analyse the meteorological conditions present using conventional observations, reanalysis and privately owned automatic weather station observations. While the instantaneous rain rates observed are not unprecedented, when maintained over several hours result in this event ranking as the highest August daily rainfall on record in the United Kingdom.

Impact of Different Nesting Methods on the Simulation of a Severe Convective Event Over South Korea Using the Weather Research and Forecasting Model

AbstractIn this study, the effects of different nesting methods on simulating a flash‐flood‐producing severe convective event over Cheongju, South Korea on July 16, 2017 was examined. This event developed as part of a mesoscale convective system (MCS) accompanied by frontal forcing. Numerical experiments were conducted using the Weather Research and Forecasting model (WRF) employing one‐way concurrent (OWC), one‐way sequential (OWS), and two‐way (TW) nesting approaches with advanced physics options from Korean Integrated Model (KIM). Analysis of model simulations against Tropical Rainfall Measuring Mission (TRMM) and Automatic Weather Station (AWS) observations suggests that the TW nesting method performs better than both OW nesting methods in simulating rainfall. Large‐scale features, moisture, instability in the boundary layer, and the vertical distribution of meteorological parameters favorable for convection are better represented by TW nesting. Probability distribution Function (PDF) analysis from AWS/WRF reveals that the local‐scale distribution of surface meteorological parameters which affect storm intensity were well captured using TW. Further assessment of Equitable Threat score (ETS) also showed better precipitation forecast skill over different thresholds in TW. Vertical velocity in the innermost domain simulated using TW nesting is more consistent with ERA5 reanalysis. An additional MCS (11–13 July 2006) simulated using a similar numerical setup also benefited from TW nesting, increasing confidence in the initial findings. Along with frequent lateral boundary conditions, TW nesting allows multi‐scale interactions between parent and nested domains and improves the representation of both synoptic and local‐scale features, enhanced cloud hydrometeor, vertical velocity distributions, and subsequent rainfall.

Initiation of an Elevated Mesoscale Convective System With the Influence of Complex Terrain During Meiyu Season

AbstractA mesoscale convective system (MCS) was initiated on the lee side of the Wuling Mountain in central‐east China in the late afternoon (1700 LT) on June 27, 2016, and led to an early morning precipitation peak. We investigated the convective initiation (CI) mechanism of this MCS using radars and automatic weather station observations, WRF model simulation, and analysis from the radar data assimilation system Variational Doppler Radar Analysis System. It was found that the convection initiation was elevated as indicated by an elevated convergence zone and updraft located above 1,625 m from the ground. A zero convective inhibition layer was located at 850 hPa, over a stable boundary layer (BL) resulting from the daytime rainfall in the plain region, and provided a favorable condition for the elevated CI. Our study suggested that the transition of mountain‐plain thermal wind near sunset was the trigger for the convective initiation. An early afternoon mountain‐plain thermal circulation was disrupted when the east‐facing lee side slope was in the shadow of sunshine after 4 p.m. As a result, a local‐scale northwest downslope wind accelerated and led to a downstream convergence above the stable BL, where the elevated convection was initiated near 1,625 m above ground level. By combining the model simulation, the analysis from data assimilation, and the observations, this study is the first to provide a detailed analysis of CI mechanism of elevated convection in a complex topography region during the Meiyu season in China.

The characteristics and temporal variability of föhn winds at King Edward Point, South Georgia

AbstractThe mountainous subantarctic island of South Georgia lies within the belt of strong westerly winds that blow around the Southern Ocean. Interaction of the prevailing circumpolar westerly winds with the island's central mountain chain, which rises to nearly 3,000 m, generates warm, dry föhn winds on the downwind side of the island, which are a significant feature of the local climate. We make use of 10 years of automatic weather station observations from King Edward Point (KEP), on the northeastern (climatologically downwind) coast of South Georgia, to develop a climatology of the occurrence and properties of föhn winds in this region. Using objective criteria to identify the onset and cessation of föhn, we find that KEP experiences föhn conditions for around 30% of the time. Föhn events last for about 30 hr on average and are associated with significant increases in temperature and wind speed, and decreases in relative humidity. On average, a föhn event is observed every 4 days, with little seasonal variation seen in the frequency of occurrence. However, föhn events observed during the austral summer season are, on average, both longer and more intense (as measured by changes in temperature, relative humidity, and wind speed) than those occurring in other seasons. The fraction of the time for which föhn conditions are observed at KEP increases (decreases) during months when the prevailing westerlies are strengthened (weakened). Monthly mean temperatures at KEP are positively correlated with the föhn fraction but variations in the latter only explain 16% of the variance of monthly mean temperature. We conclude that, while föhn plays an important role in shaping the local climate of South Georgia, other processes may be of greater importance in controlling regional climate variability and change.

Cluster analysis of tropical cyclones affecting the Taiwan Strait

Automatic weather station (AWS) observations are used to select tropical cyclones (TCs) affecting the Taiwan Strait during 2007–2016. Applying these data, the influence of TCs on the Taiwan Strait is measured and TCs with weak or no impact are removed. Objective statistical properties of the remaining TCs are analysed by clustering them into three groups using the fuzzy C‐means method. Clusters are differentiated as (a) TCs affecting the Taiwan Strait but recurving at the edge of the TC alert area, (b) TCs moving straight across The Philippines and the South China Sea, and (c) TCs striking Taiwan and eastern China and recurving after landfall. Differences among the clusters include in the spatial range and duration of gales monitored by AWS, and the track, intensity, and genesis location of TCs. Also analysed are the influence of large‐scale atmospheric environments on TC motion, sea‐surface temperature responses to TCs, and the influence of El Niño‐Southern Oscillation on TC classifications.

The importance of accurate glacier albedo for estimates of surface mass balance on Vatnajökull: evaluating the surface energy budget in a regional climate model with automatic weather station observations

Abstract. A simulation of the surface climate of Vatnajökull ice cap, Iceland, carried out with the regional climate model HIRHAM5 for the period 1980–2014, is used to estimate the evolution of the glacier surface mass balance (SMB). This simulation uses a new snow albedo parameterization that allows albedo to exponentially decay with time and is surface temperature dependent. The albedo scheme utilizes a new background map of the ice albedo created from observed MODIS data. The simulation is evaluated against observed daily values of weather parameters from five automatic weather stations (AWSs) from the period 2001–2014, as well as in situ SMB measurements from the period 1995–2014. The model agrees well with observations at the AWS sites, albeit with a general underestimation of the net radiation. This is due to an underestimation of the incoming radiation and a general overestimation of the albedo. The average modelled albedo is overestimated in the ablation zone, which we attribute to an overestimation of the thickness of the snow layer and not taking the surface darkening from dirt and volcanic ash deposition during dust storms and volcanic eruptions into account. A comparison with the specific summer, winter, and net mass balance for the whole of Vatnajökull (1995–2014) shows a good overall fit during the summer, with a small mass balance underestimation of 0.04 m w.e. on average, whereas the winter mass balance is overestimated by on average 0.5 m w.e. due to too large precipitation at the highest areas of the ice cap. A simple correction of the accumulation at the highest points of the glacier reduces this to 0.15 m w.e. Here, we use HIRHAM5 to simulate the evolution of the SMB of Vatnajökull for the period 1981–2014 and show that the model provides a reasonable representation of the SMB for this period. However, a major source of uncertainty in the representation of the SMB is the representation of the albedo, and processes currently not accounted for in RCMs, such as dust storms, are an important source of uncertainty in estimates of snow melt rate.

Automatic Weather Station Observations of the April 2014 Mount Everest Avalanche

ABSTRACTInstrumental records indicate a warming of approximately 0.8 °C has occurred in the Mount Everest region since the 1980s, which has resulted in a 100–300 m rise in the height at which the ground is permanently frozen as well as a retreat and thinning of Everest’s glaciers. For some time, there have been concerns that this warming and the resultant changes in the region’s glaciers may be increasing the risks for travellers to Mount Everest as well as the indigenous populations who support them. On 18 April 2014, an avalanche caused by the collapse of a large serac swept down Mount Everest’s Khumbu Ice Fall resulting in the deaths of 16 Sherpa. Although satellite imagery has been used to estimate the size of the serac, in situ data on the avalanche itself has not been available. Here we show that this event coincided with an approximate 15-min-long wind, thermal, and moisture anomaly, which was observed at the Nepal Climate Observatory-Pyramid situated 10 km from Mount Everest. We argue that this anomaly was associated with the avalanche and thereby provides some information on its scale and duration as well as a potential mechanism to monitor future events in this remote and data-sparse region.

Measurement and modelling of temperature cold pools in the Cerdanya valley (Pyrenees), Spain

ABSTRACTThe Cerdanya valley in northeast Spain experiences intense cold air pooling which decouples the valley atmosphere from the regional circulation, especially in winter. This makes air temperature prediction a challenge. A network of 40 temperature sensors was installed in 2012 along seven elevational transects to collect hourly temperatures throughout the cold pool, enabling measurement of the detailed cold pool structure for the first time. Sensors were also installed in the upper Conflent valley to the northeast for comparison, where previous research has shown that there is reduced cold air pooling. Sensor data were validated against automatic weather station observations at two locations. Through calculation of hourly lapse rates in various elevation bands for 2 years, frequent inversions developing up to 1450 m are shown, and sometimes extending much higher than this, concentrating in winter. Case studies of two intense episodes in December 2012 and January 2013 show that model simulations, despite being able to simulate broad mechanisms of cold air pooling formation, underestimate the amount of cooling. This is due to over‐enthusiastic simulation of a low level jet in the first case and mountain waves in the second. Solving these model problems has important consequences for the future ability to predict episodes of extreme low temperatures and associated hazards (frost, fog) in Cerdanya and mountain valleys elsewhere.

Using remotely sensed data from AIRS to estimate the vapor flux on the Greenland ice sheet: Comparisons with observations and a regional climate model

Mass loss from the Greenland ice sheet (GrIS) in recent years has been dominated by runoff from surface melt. It is currently being studied extensively, while little interest has been given to the smallest component of surface mass balance (SMB): the vapor flux. Although poorly understood, it is not negligible and could potentially play a larger role in SMB in a warming climate where temperature, relative humidity, and precipitation changes remain uncertain. Here we present an innovative approach to estimate the vapor flux using the Atmospheric Infrared Sounder (AIRS) version 6 data and a modified vapor flux model (BMF13) over the GrIS between 2003 and 2014. One modification to the BMF13 model includes a new Multiangle Imaging SpectroRadiometer surface aerodynamic roughness product, which likely produces more accurate estimates of the drag coefficient on the ice sheet. When comparing AIRS data with GC-Net and Programme for Monitoring of the Greenland Ice Sheet automatic weather station observations of skin temperature, near-surface air temperature, and humidity, they agree within 2 K, 2.68 K, and 0.34 g kg−1. Largest differences occur in the ablation zone where there is significant subgrid heterogeneity. Overall, the average vapor flux from the GrIS between 2003 and 2014 was found to be 14.6 ± 3.6 Gt yr−1. No statistically significant trends were found during the data record. This data set is compared to the Regional Atmospheric Climate Model (RACMO2.3) vapor flux, and BMF13 produced smaller vapor fluxes in the summer (~0.05 Gt d−1) and slightly more deposition in the winter (~9.4 × 10−3 Gt d−1). Annually, differences between BMF13 and RACMO2.3 were only 30 ± 15%.

МЕТОДИКА РАСЧЕТА ВОДНОГО БАЛАНСА ДЛЯ СУХОЙ СТЕПИ ЮГА ЗАПАДНОЙ СИБИРИ ПО ДАННЫМ МЕЖДУНАРОДНОЙ МОНИТОРИНГОВОЙ СЕТИ

<p>The study of the water balance is not only of great practical importance, but also of interest from the standpoint of basic science for the researchers of the steppe zone, which has traditionally attracted people for its climate and soil resources. The article presents the results obtained from a network of automatic stations deployed in the framework of the Russian-German scientific cooperation in the field of climate, soil, and hydrological monitoring for 2012-2015. The instrumental data were obtained and the method of soil water balance calculation was performed on the basis of data of automatic weather station observations and lysimetric stations of new generation in the conditions of Western Siberia.</p><p> </p>

Föhn winds on South Georgia and their impact on regional climate

Average summer temperatures over South Georgia have risen by 1°C along with an increase in the rate of glacial retreat since the 1920s. Glaciers on the northeast of the island are retreating faster than those on the southwest side. These asymmetrical changes are thought to be linked with the fohn effect. To explore the link between fohn and the climate of South Georgia, we show results from a climatological analysis using Automatic Weather Station observations and high-resolution simulations of fohn events with the Weather Research and Forecasting model. We find that fohn events significantly contribute to the asymmetry of South Georgia's regional climate.

Modeling ground surface temperature by means of remote sensing data in high-altitude areas: test in the central Tibetan Plateau with application of moderate-resolution imaging spectroradiometer Terra/Aqua land surface temperature and ground-based infrared radiometer

Ground surface temperature (GST) is a crucial parameter of surface energy budgets and controls the thermal state of the active layer and permafrost in permafrost regions. However, with limited observed datasets available for the Tibetan Plateau, a greater bias existed for GST products from remote sensing data. Model validation (the whole year 2012 data) showed that all three models performed well, with a determination (R2), mean error, mean absolute error, and root mean squared error of 0.86 to 0.93, −0.61 to 1°C, 2.28 to 3.06°C, and 2.96 to 3.83°C, respectively. The model established by observations of Terra and Aqua satellites during the daytime and nighttime showed the highest correlation, with R2 values ranging from 0.91 to 0.93, as well as the lowest MAE and RMSE of 2.28 to 2.42 and 2.96 to 3.05°C, respectively. However, the application of this model substantially reduced the available pixels. Models established with the automatic weather station observations at the satellite overpass times performed better than those using the moderate-resolution imaging spectroradiometer land surface temperature observations. The results might be useful to produce a more reliable dataset for monitoring and modeling permafrost changes.

Surface energy budget on Larsen and Wilkins ice shelves in the Antarctic Peninsula: results based on reanalyses in 1989–2010

Abstract. Ice shelves in the Antarctic Peninsula have significantly disintegrated during recent decades. To better understand the atmospheric contribution in the process, we have analysed the inter-annual variations in radiative and turbulent surface fluxes and weather conditions over Larsen C Ice Shelf (LCIS) and Wilkins Ice Shelf (WIS) in the Antarctic Peninsula in 1989–2010. Three atmospheric reanalyses were applied: ERA-Interim by ECMWF, Climate Forecast System Reanalysis (CFSR) by NCEP, and JRA-25/JCDAS by the Japan Meteorological Agency. In addition, in situ observations from an automatic weather station (AWS) on LCIS were applied, mainly for validation of the reanalyses. The AWS observations on LCIS did not show any significant temperature trend, and the reanalyses showed warming trends only over WIS: ERA-Interim in winter (0.23 °C yr−1) and JRA-25/JCDAS in autumn (0.13 °C yr−1). In LCIS from December through August and in WIS from March through August, the variations of surface net flux were partly explained by the combined effects of atmospheric pressure, wind and cloud fraction. The explained variance was much higher in LCIS (up to 80%) than in WIS (26–27%). Summer melting on LCIS varied between 11 and 58 cm water equivalent (w.e.), which is comparable to previous results. The mean amount of melt days per summer on LCIS was 69. The high values of melting in summer 2001–2002 presented in previous studies on the basis of simple calculations were not supported by our study. Instead, our calculations based on ERA-Interim yielded strongest melting in summer 1992–1993 on both ice shelves. On WIS the summer melting ranged between 10 and 23 cm w.e., and the peak values coincided with the largest disintegrations of the ice shelf. The amount of melt on WIS may, however, be underestimated by ERA-Interim, as previously published satellite observations suggest that it suffers from a significant bias over WIS.