Quantifying supraglacial debris thickness at local to regional scales

Supraglacial debris thickness is a key control on the surface energy balance of debris-covered glaciers, which are common in temperate mountain ranges around the world. As such, it is an important input variable to the sorts of models that are used to understand and predict glacier change, which are essential for determining future water supply in glacierised regions and glacier contributions to sealevel rise. However, to quantify supraglacial debris thickness is difficult: making direct measurements is laborious and existing remote sensing approaches have not been thoroughly validated, so there is a general paucity of supraglacial debris thickness data. This thesis investigates methods of quantifying supraglacial debris thickness at local to regional scales. First, it makes in-situ field measurements of debris thickness at the local scale on glaciers in the Himalaya and the European Alps by manual excavation and by ground-penetrating radar (GPR). Second, it uses some of these field measurements to test and develop thermal remote sensing approaches to quantifying supraglacial debris thickness at the glacier scale. Third, it uses a dynamic energy-balance model in an inverse approach to quantify debris thickness on the glaciers of three watersheds in High Mountain Asia from thermal satellite imagery and high-resolution meteorological reanalysis data. At the local scale, GPR is found to be useful for measuring supraglacial debris thickness accurately and precisely, at least in the range 0.16-4.9 m. Debris thickness is highly variable over horizontal distances of < 10 m on individual glaciers due to gravitational reworking, which necessarily implies higher sub-debris ice melt rates than if debris thickness was spatially invariable. At the glacier scale, thermal remote sensing approaches can reproduce field measurements, and remote sensing estimates of supraglacial debris thickness can be used successfully to model sub-debris melting. If welldistributed field measurements are available, supraglacial debris thickness should be extrapolated using remote sensing-derived pseudo daily mean surface temperatures. Otherwise, it should be determined iteratively by minimising the mismatch between remotely sensed surface temperatures, preferably from night-time thermal images, and surface temperatures determined using a dynamic energy-balance model. At the regional scale, thermal satellite imagery and high-resolution meteorological reanalysis data can be used to provide reasonable estimates of supraglacial debris thickness. However, modelled uncertainties are not always able to explain ground-truth measurements, and there is a tendency towards underestimation due to problems associated with supraglacial ponds and ice cliffs and the spatial resolution of input data. The findings of this thesis will lead to improvements in the quantification of supraglacial debris thickness at a range of scales and, therefore, in the understanding and prediction of glacier change in temperate mountain ranges