A new tool is evaluated in which the climate and surface mass balance (SMB) of the Greenland ice sheet are resolved at 55 and 28 km resolution, while the rest of the globe is modelled at ~110 km. The local refinement of resolution leads to improved accumulation (SMB > 0) compared to observations; however ablation (SMB < 0) is deteriorated in some regions. This is attributed to changes in cloud cover and a reduced effectiveness of a model-specific vertical downscaling technique.

Can we ignore the contribution of small ice leads when estimating turbulent heat flux? Combining bulk formulae and a fetch-limited model with surface temperature from MODIS and Landsat-8 Thermal Infrared Sensor (TIRS) images, we found small leads account for 25 % of the turbulent heat flux, due to its large total area. Estimated turbulent heat flux is larger from TIRS than that from MODIS with a coarser resolution and larger using a fetch-limited model than that using bulk formulae.

We present high-resolution bathymetry data that provide the most complete and detailed imagery of any Antarctic palaeo-ice stream bed. These data show how subglacial water was delivered to and influenced the dynamic behaviour of the ice stream. Our observations provide insights relevant to understanding the behaviour of modern ice streams and forecasting the contributions that they will make to future sea level rise.

In this study we investigate the historical (1980–2015) and projection scenario (2070–2099) components of freshwater runoff to Glacier Bay, Alaska, using a modeling approach. We find that many of the historically snow-dominated watersheds in Glacier Bay National Park and Preserve may transition towards rainfall-dominated hydrographs in a projection scenario under RCP 8.5 conditions. The changes in timing and volume of freshwater entering Glacier Bay will affect bay ecology and hydrochemistry.

Using scaling analysis we propose that the currently observed marine ice-sheet instability in the Amundsen Sea sector might be faster than all other potential instabilities in Antarctica.

The streamflow of Arctic rivers is changing. We analyzed available data (22 gauges, 1936–2015) in the basins of the Yana and Indigirka rivers completely located within the continuous permafrost zone. The results show that the main factor of increasing low flows is the shift from snow to rain due to warming. Other factors related to the release of water from permafrost, glaciers, or aufeis may fractionally contribute to streamflow increase but cannot be quantified based on available data.

A warm bias and higher total precipitation and snowfall were found in ERA5 compared with ERA-Interim (ERA-I) over Arctic sea ice. The warm bias in ERA5 was larger in the cold season when 2 m air temperature was < −25 °C and smaller in the warm season than in ERA-I. Substantial anomalous Arctic rainfall in ERA-I was reduced in ERA5, particularly in summer and autumn. When using ERA5 and ERA-I to force a 1-D sea ice model, the effects on ice growth are very small (cm) during the freezing period.

The Greenland Ice Sheet loses 30 %–60 % of its ice due to iceberg calving. Calving processes and their links to climate are not well understood or incorporated into numerical models of glaciers. Here we use a new 3-D calving model to investigate calving at Store Glacier, West Greenland, and test its sensitivity to increased submarine melting and reduced support from ice mélange (sea ice and icebergs). We find Store remains fairly stable despite these changes, but less so in the southern side.

The sea ice surface albedo parameterization of a coupled regional climate model was validated against aircraft measurements performed in May–June 2017 north of Svalbard. The albedo parameterization was run offline from the model using the measured parameters surface temperature and snow depth to calculate the surface albedo and the individual fractions of the ice surface subtypes. An adjustment of the variables and additionally accounting for cloud cover reduced the root-mean-squared error.

Changes in glacier mass are the direct response to meteorological conditions during the accumulation and melt seasons. We derived multi-year, seasonal mass balance from airborne laser scanning surveys and compared them to field measurements for six glaciers in the Columbia and Rocky Mountains, Canada. Our method can accurately measure seasonal changes in glacier mass and can be easily adapted to derive seasonal mass change for entire mountain ranges.

Conventionally, calving front positions have been manually delineated from remote sensing images. We design a novel method to automatically delineate the calving front positions of Jakobshavn Isbræ based on deep learning, the first of this kind for Greenland outlet glaciers. We generate high-temporal-resolution (about two measurements every month) calving fronts, demonstrating our methodology can be applied to many other tidewater glaciers through this successful case study on Jakobshavn Isbræ.

The apparent discrepancy between the Holocene δ¹⁸O records of the Guliya and the Chongce ice cores may be attributed to a possible misinterpretation of the Guliya ice core chronology.

To study the process of snow aging, we engineered a prototype instrument called the Near-Infrared Emitting and Reflectance-Monitoring Dome (NERD). Using the NERD, we observed rapid snow aging in experiments with added light absorbing particles (LAPs). Particulate matter deposited on the snow increased absorption of solar energy and enhanced snow melt. These results indicate the role of LAPs' indirect effect on snow aging through a positive feedback mechanism related to the snow grain size.