The Arctic Glacial Lakes project was funded by the US National Science Foundation, Arctic System Science Program from 2015 to 2018, with the goal of understanding how weather and climate are filtered through the glacier-hydrology-lake-sedimentation system and are recorded in physical components of lake deposits in glaciated catchments. The study was motivated by the need to improve interpretations about past and present environmental changes that are recorded by sediments that accumulate in Arctic lakes. The focus was on lakes in glaciated watersheds because their sediment often reflects changes in the size and melt rate of upstream glaciers, which are among the most dynamic components of the rapidly evolving Arctic system.
This project established a watershed observatory at Lake Peters, Alaska, where we observed multiple weather, glacier, river and lake processes. The resulting datasets are featured in the journal, Earth System Science Data, they are summarized in the “observatory data” section of this website, and are archived in full at Arctic Data Center. We also developed similar monitoring datasets at Eklutna Lake, and rescued seven years of similar datasets at Lake Linne.
The results show:
(1) Sediment transport to all three lakes exhibits significant inter-annual variability. This emphasizes the need for long-term monitoring studies and the challenge of deriving robust conclusions from short instrumental records. This also motivates the use of proxy records to infer long-term trends and variability.
(2) Rainfall dominates discharge hydrography and controls sediment transport at all three study lakes. This was expected for south Alaska, but is surprising in the Arctic, and has implications for the future of the Arctic where rainfall is projected to continue to increase. This finding also challenges some proxy-based reconstructions that relate sediment accumulation to summer melt temperature.
(3) River discharge, primarily driven by rainfall, is a key predictor variable for sediment transport in all three catchments, although sediment yield models developed for this project performed best when meteorological or temporal explanatory variables were also included.
(4) Our sediment transport model for the Lake Peters watershed was linked to a lake sedimentation model to simulate how changes in discharge, driven by precipitation and glacier melt, are ultimately recorded as sedimentation in the lake. The model does a good job of simulating the overall amount of annual and total sediment deposition from 2015-2017, as well as the relative amounts in each year, and the changes in deposition rates as a function of distance from the inlet.
We shared our project activities and results beyond of the scientific community. In 2017 we teamed up with PolarTREC to host high school teacher Rebecca Harris at Lake Peters. Her experiences are documented here. She and Ph.D. student Ellie Broadman shared our project with students at Kaktovik school and through webinars. We communicated regularly with the interested public, land managers, and to other scientists on the project facebook page.