Lakes Churup and Churupito

Lakes Churup (4,480 m asl) and Churupito (4,580 m asl) are located in the Churup valley embedded into the west-facing slopes of the central Cordillera Blanca massif. The formation of both these lakes is associated with glacial activity – larger Lake Churup fills the depression carved by repeated cycles of glacier re-advances in the deep past, while smaller Lake Churupito is dammed by moraine material deposited by the glacier during its last advance, so called Little Ice Age (LIA) terminating in the second half of 19th Century.

Fig. 1: Upstream located moraine-dammed Lake Churupito and downstream located bedrock-dammed Lake Churup as seen from SW-facing slopes of Nevado Churup.

This glacier originates on SW-facing slopes of Nevado Churup (5,495 m asl) and underwent evident retreat since its maximum extent during the LIA. While the LIA extent was reconstructed being 1.05 ± 0.1 km2, this number dropped to 0.63 km2 in 1948, 0.49 km2 in 1986 and 0.18 km2 in 1995. Current extent of glacierised area is less than 0.05 km2.

Fig. 2: Lake Churup and the whitish peak of Nevado Churup (5,495 m asl) with remaining glacier (in 2015) as seen from the dam of Lake Churup. Lake Churupito is located behind the moraine visible in the centre of the image.

Being located just 1 hour bus ride + 4 hours walk from regional capital Huaráz, both lakes have attracted increasing attention of tourists and have became a popular visiting spot during the past years. Climate change-induced glacier retreat, however, changes the landscape of the Cordillera Blanca in a way that future generations will likely not enjoy the panoramic view of the whitish peak towering high above the mirroring lake water level.

Further reading: Emmer, A., Juřicová, A., Veettil, B.K. (2019). Glacier retreat, rock weathering and the growth of lichens in the Churup Valley, Peruvian Tropical Andes. Journal of Mountain Science, not yet assigned to an issue.

WP5: GLOF implications & impacts

Using the CORDEX simulations outlined in WP2 and the GLOF simulations from WP4, at selected sites in Peru we will undertake a detailed analysis of the potential downstream economic and societal impacts of future GLOFs, including the impact on communities, infrastructure and agriculture. We will make recommendations for mitigation strategies. Finally, as an output from this research project we will make recommendations for GLOF hazard assessment protocols and mitigation strategies in lower income countries globally.

WP4: GLOF modelling

We will (a) establish the physical processes that govern GLOF behaviour such as moraine-dam failure, ice-dammed lake drainage and slope instability; (b) analyse flood hydrographs of selected former GLOFs to establish downstream impacts. We will undertake research to understand the physical processes that govern the development of Glacial Lake Outburst Floods (GLOFs) in Peru: From the inventory in WP1, we will then identify currently developing GLOF sites and establish the locations of potential future GLOF sites. We will use these observations to establish the physical processes that govern GLOF behaviour in Peru including sites that may be prone to moraine-dam failure, the locations of sites likely to be sensitive to ice-dammed lake drainage and sites that are susceptible to rock slope failure. We will analyse flood hydrographs (where available) of selected former GLOFs and use these to establish the patterns of downstream impacts.

WP3: GLOF hazard assessment

Using remote-sensing (WP1a), geomorphological mapping (WP1b) and fieldwork (WP1c), we will identify sites of past GLOF sites around the major glaciers of Peru. We will use these sites to identify probable physical triggers of the identified GLOFs in order to better understand the processes that impact on glacial lake systems. Using this information we will enhance existing methods of hazard assessment (e.g. Reynolds, 2014) and identify sites at which glacial hazards are assessed as ‘severe’ or ‘very severe’. We will use these observations to establish the physical processes that govern GLOF behaviour in Peru including sites that may be prone to moraine-dam failure, the locations of sites likely to be sensitive to ice-dammed lake drainage and sites that are sensitive to rockslide and debris flow activity. We will analyse flood hydrographs (where available) of selected former GLOFs and use these to establish the patterns of downstream impacts. We will objectively assess the potential for damage to communities and infrastructure downstream. In tandem, we will also develop a numerical model based on the CORDEX simulations outlined in WP2 and conduct numerical simulations of downstream impacts for selected potential GLOF sites using these physically based numerical flood models. We will identify current and likely future glacier hazards focusing on the developing landslide and debris flow risk as glaciers recede; establish the locations of potential future vulnerable lakes and potential GLOF sites.

WP1: Climate modelling

We will use an ensemble of 14 downscaled Coupled Model Intercomparison Project phase 5 (CMIP5) models which use the RCP8.5 ‘business as usual scenario’ and cover a wide range of climate sensitivities. The CMIP5 models were downscaled using HadGEM3-A and the EC-EARTH-HR atmosphere only global climate models, providing us with climate data on a high resolution (approximately 50 km2). This will help capture the variability in climate over complex mountain topography. The high resolution climate data has been bias corrected using a trend preserving statistical bias method that was developed for the first Inter-Sectoral Impact Model Intercomparison Project (ISI-MIP) (Hempel et al., 2013). To explore the response of the MAAT proxy to scenarios where carbon emissions are reduced though mitigation, we will use the CORDEX-South America regional climate model datasets for scenarios RCP4.5 and RCP2.6 (Solman, 2013). Using the HELIX and CORDEX ensembles allow us to generate a range of future MAAT responses which include uncertainty in both climate change scenarios and climate model sensitivities. This Work Package feeds directly into WP5.