INFRARED IMAGING AND MODELING OF PROGLACIAL STREAM TEMPERATURE IN THE CORDILLERA BLANCA, PERU

In the tropical Andes, glacial meltwater is an important water resource for downstream communities. It is especially important because the region experiences distinct wet and dry seasons, with most precipitation occurring between October and May. However, climate change is continuing to accelerate the rate of glacial retreat, placing increased stress on the region’s already limited water resources. As this retreat continues, the ability of proglacial valleys to store groundwater and discharge it to the region’s rivers will be critical to sustaining river flows. In order to improve our understanding of the hydrologic fluxes in these proglacial valleys, we used high resolution infrared (IR) imagery coupled with in-stream temperature sensors to create an energy balance model for an ~1.2 km reach of the Quilcayhuanca River in the Cordillera Blanca, Peru. We positioned the IR camera 96 m above the stream on the southern valley wall and collected images every 10 minutes for 5 days, leading to the acquisition of over 700 IR images. We deployed 40 temperature sensors within the stream at approximately equal intervals. Stream temperature profiles were installed within the reach to determine that the water column is well mixed enough for the use of IR imagery. We installed 10 temperature sensors throughout the valley as additional control points. The temperature sensors all recorded data every 5 minutes for the length of the study. We obtained groundwater temperatures from 6 wells in the valley. The reach temperature varied diurnally from ~4 to 13°C. Temperature sensors installed in the tributaries that flow into the main channel showed various trends, with 4 exhibiting warmer temperatures than the main channel, one exhibiting temperatures similar to those of the groundwater (~9 to 11°C), and another with a range of ~18°C. We deployed a weather station that recorded air temperature, relative humidity, wind speed, and shortwave radiation every 10 minutes. IR image processing and calibration were performed in MATLAB. The field data were used as input for a spatial and temporal energy balance model of the reach, with the IR imagery increasing the spatial resolution of the model. Our study exhibits the value of IR imaging in stream temperature modeling and increases our knowledge of the heat fluxes occurring in this proglacial stream.