Predicting Lake Depths from Topography to Map Global Lake Volume

The depth of a lake affects its role in climate and biogeochemical cycling. There is a lack of lake depth data due to the difficulty of measuring bathymetry, which impedes the accurate inclusion of lakes in climate models and the assessment of global water resources and carbon storage. However, lake depths can be estimated from land topography, for which remotely-sensed DEM data is available. We develop a simple statistical model to predict lake depth from two explanatory variables: the mean relief above the lake surface of an area around the lake, and whether the lake’s location was glaciated in the last ice age. The model is based on 328 lakes with known depths, located on all continents but Antarctica, and has an r2 of 0.57. We then apply this model to a database of over 200,000 lakes to produce global gridded maps of predicted total lake volume and average depth. The realistic depth estimates provided by our model can improve the accuracy of future studies of climate and water resources.Bachelor of Scienc