Search

Glaciers represent an important water resource for Andean cities and mountain communities. However, a recent study has shown that Bolivian glaciers have shrunk by _43% in area over the last _30 years. If current rates of glacier recession are sustained then there could be potentially important consequences for downstream water supply, especially during the dry season. A first step in assessing the severity of this problem is to estimate the current volume of glacier ice in Bolivia, and how this has changed over recent decades. Here, we use VOLTA (created by James and Carrivick, 2016 – Computers & Geosciences), an ArcGIS tool requiring only a Digital Elevation Model (DEM) and glacier outlines to give a first-order ice thickness estimate and therefore derive volume changes for the entire Bolivian Andes between 1986 and 2017. The VOLTA tool also models bed topography, which we use to make a preliminary assessment of the locations of subglacial overdeepennings, which will become loci for future proglacial lakes capable of generating glacial lake outburst floods (GLOFs) and storing meltwater.

Glaciers of the Bolivian Andes have experienced areal shrinkage of _43% in the last three decades, which has been accompanied by the development of proglacial lakes, some of which could generate glacial lake outburst floods (GLOFs). We provide the first attempt to assess GLOF risk in Bolivia, and model potential GLOF inundation. There are _137 proglacial lakes in the Bolivian Andes, 25 of which have population and/or infrastructure downstream.We first developed a GLOF risk assessment strategy using Multi-Criteria Decision Analysis (MCDA) guidelines that could be used remotely and free-of-charge to identify glacial lakes that represent the greatest GLOF risk. This revealed that three lakes posed medium or high risk, and required further analysis. Secondly, we undertook a modelling study of potential GLOF inundation from these three lakes. This involved the generation of a 2m resolution Digital Elevation Model (DEM) from stereo and tri-stereo SPOT 6/7 satellite images; the 2D hydrodynamic model HEC-RAS 5.0.3 was used to model GLOF flow. The model was tested against field observations of a 2009 GLOF from Keara, in the Cordillera Apolobamba, and was shown to reproduce realistic flood depths and inundation. The model was then used to model GLOFs from Pelechuco lake (Cordillera Apolobamba), and Laguna Arkhata and Laguna Glaciar (Cordillera Real). In total, six villages could be affected by GLOFs if all three lakes were to burst. We ran the model for three scenarios (pessimistic, intermediate, optimistic) which give a range of 1589 and 2302 people affected by flooding; between 1107 and 2168 people would be exposed to damaging floods (flow depth _ 2m). We suggest that Laguna Arkhata and Pelechuco lake represent the greatest risk due to the higher numbers of people who live in the potential flood paths, and hence should be a priority for risk managers.

Glaciers across the world are thinning and receding in response to atmospheric warming. Glaciers tend to erode subglacial basins and deposit eroded materials around their margins as lateral-frontal terminal moraines. Recession into these basins and behind impounding moraines causes meltwater to pond as proglacial and supraglacial lakes. Consequently, there has been a general trend of increasing number and size of these lakes associated with glacier melting in many mountainous regions around the globe, in the last 30 years. Glacial lake outburst floods (GLOFs) then may occur where the glacial lake dam (ice, rock, moraine, or combination thereof) is breached, or overtopped, and thousands of people have lost their lives to such events in the last few decades, especially in the Andes and in the Himalaya. Given the ongoing and arguably increasing risk posed to downstream communities, and infrastructure, there has been a proliferation of GLOF studies, with many seeking to estimate GLOF hazard or risk in specific regions, or to identify ’potentially dangerous glacial lakes’. Given the increased scientific interest in GLOFs, it is timely to evaluate critically the ways in which GLOF risk has been assessed previously, and whether there are improvements that can be made to the ways in which risk assessment is achieved. We argue that, whilst existing GLOF hazard and risk assessments have been extremely valuable they often suffer from a number of key shortcomings that can be addressed by using different techniques as multi-criteria decision analysis and hydraulic modelling borrowed from disciplines like engineering, remote sensing and operations research.

Glaciers across the world are thinning and receding in response to atmospheric warming. Glaciers tend to erode subglacial basins and deposit eroded materials around their margins as lateral-frontal terminal moraines. Recession into these basins and behind impounding moraines causes meltwater to pond as proglacial and supraglacial lakes. Consequently, there has been a general trend of increasing number and size of these lakes associated with glacier melting in many mountainous regions around the globe, in the last 30 years. Glacial lake outburst floods (GLOFs) then may occur where the glacial lake dam (ice, rock, moraine, or combination thereof) is breached, or overtopped, and thousands of people have lost their lives to such events in the last few decades, especially in the Andes and in the Himalaya. Given the ongoing and arguably increasing risk posed to downstream communities, and infrastructure, there has been a proliferation of GLOF studies, with many seeking to estimate GLOF hazard or risk in specific regions, or to identify ‘potentially dangerous glacial lakes’. Given the increased scientific interest in GLOFs, it is timely to evaluate critically the ways in which GLOF risk has been assessed previously, and whether there are improvements that can be made to the ways in which risk assessment is achieved. We argue that, whilst existing GLOF hazard and risk assessments have been extremely valuable they often suffer from a number of key shortcomings that can be addressed by using different techniques as multi-criteria decision analysis and hydraulic modelling borrowed from disciplines like engineering, remote sensing and operations research.