The near-surface weathering crust is a thin (<0.5 m), low density ice layer that develops on glacier surfaces during the ablation season and is formed by internal melting driven by the penetration of shortwave radiation (SWR) into polycrystalline glacier ice. This ‘photic zone’ hosts microbial communities, mediates biogeochemical processes and routes meltwater to the channelised supraglacial drainage network. Despite these critical roles, direct field measurements of weathering crust formation and evolution are scarce—rather, current understanding is largely derived from modelling approaches. Here, we present in situ measurements of weathering crust evolution at five sites on the western Greenland ice sheet, each over a 19–25 hour period. Shallow ice cores revealed weathering crust ice densities of 420–910 kg m−3, demonstrating dynamic evolution of the weathering crust linked to diurnal SWR receipt. We compare our empirical data with two existing weathering crust models, neither of which fully reproduce the observed ice density or its temporal variability. Additionally, we reveal that the density of the uppermost 0.1 m of the weathering crust is a key control upon bare-ice albedo. Our findings highlight the need for improved process-level-understanding and parameterisations of weathering crust dynamics in surface energy balance models.

Near-surface ice layers in the lower accumulation zone of Arctic glaciers and ice sheets may significantly affect deep meltwater percolation and runoff availability. This study presents a framework to assess three methods for characterizing near-surface ice layer permeability and its influence on runoff and mass balance on the Devon Ice Cap using field data. In the most effective method, ice layer permeability depends on temperature and thickness: they remain permeable above a threshold temperature (Tth = −0.15°C) and become impermeable once exceed a critical thickness (Himp = 1 m). Our modelling replicates ice layers that are typically thinner in the upper accumulation zone and thicker in the lower accumulation zone. Additionally, we simulate an observed increase in the number of ice layers in the upper accumulation zone after 2007. The evolution of thicker (>1 m) ice layers (or ice slab) in the lower accumulation zone reduces meltwater retention through refreezing, making surface mass balance (SMB) and runoff sensitive to climate changes. Simulated mean SMB ranged from −0.09 to 0.26 m w.e. a−1 from 1999 to 2022. Our model can be applied to simulate the long-term evolution of ice slab and project its impact on ice sheet runoff.

New experiments have revealed that a thin layer of granular ice bonded to salty and to salt-free columnar-grained ice increases flexural strength when the composite material is rapidly bent to the point of failure through brittle fracture. When bent slowly within the regime of ductile behavior, the layer has no detectable effect. Strengthening is attributed to the suppression of cracking; its absence, to dislocation creep.

Glacier surficial melt rates are commonly modelled using surface energy balance (SEB) models, with outputs applied to extend point-based mass-balance measurements to regional scales, assess water resource availability, examine supraglacial hydrology and to investigate the relationship between surface melt and ice dynamics. We present an improved SEB model that addresses the primary limitations of existing models by: (1) deriving high-resolution (30 m) surface albedo from Landsat 8 imagery, (2) calculating shadows cast onto the glacier surface by high-relief topography to model incident shortwave radiation, (3) developing an algorithm to map debris sufficiently thick to insulate the glacier surface and (4) presenting a formulation of the SEB model coupled to a subsurface heat conduction model. We drive the model with 6 years of in situ meteorological data from Kaskawulsh Glacier and Nàłùdäy (Lowell) Glacier in the St. Elias Mountains, Yukon, Canada, and validate outputs against in situ measurements. Modelled seasonal melt agrees with observations within 9% across a range of elevations on both glaciers in years with high-quality in situ observations. We recommend applying the model to investigate the impacts of surface melt for individual glaciers when sufficient input data are available.

Light absorbing impurities (LAI) initiate powerful snow albedo feedbacks, yet due to a scarcity of observations and measurements, LAI radiative forcing is often neglected or poorly constrained in climate and hydrological models. To support physically-based modeling of LAI processes, daily measurements of dust and black carbon (BC) stratigraphy, optical grain size, snow density and spectral albedo were collected over the 2013 ablation season in the Rocky Mountains, CO. Surface impurity concentrations exhibited a wide range of values (0.02–6.0 mg g−1 pptw) with 98% of mass being deposited by three episodic dust events in April. Even minor dust loading initiated albedo decline, and the negative relationship between dust concentrations and albedo was log-linear. As melt progressed, individual dust layers coalesced and emerged at the snow surface, with minimal mass loss to meltwater scavenging. The observations show that the convergence of dust layers at the surface reduced albedo to 0.3 and snow depth declined ~50% faster than other years with similar depth but less dust. The rapid melt led to an unexpected reduction in both grain size and density in uppermost surface layers. BC concentrations co-varied with dust concentrations but were several orders of magnitude lower (<1–20 ppb).