Arctic Deltas (ADs) are generally characterized by flashier, snowmelt controlled streamflow regimes then their temperature counterparts, geomorphic inactivity through the winter, and abundant thermokarst lakes. Projected climate change will change streamflow regimes and thaw permafrost, but it is not yet clear how these changes will manifest themselves in complex deltaic environments. As ADs contain significant quantities of labile carbon and modulate how riverine freshwater, sediment, and nutrient fluxes reach the Arctic Ocean, these geomorphic changes will have broader implications.
One question we have asked is how the spatial distribution of permafrost may influence seasonal lake variability and hydrologic connectivity. Specifically, we hypothesized that active layer thickness increases closer to river channels, resulting in greater near-surface flow from lakes to channels during flood recession and therefore greater summertime lake area variability. We measured summertime lake area variability using 20 years of classified Landsat imagery over the Yukon and Colville and found that sumemrtime lake area shrank faster closer to the channel network compared with farther away. Our results have been published in Geophysical Research Letters.
The second overarching question we have asked is since thermokarst lakes grow due to the thawing of ice-rich permafrost, do thermokarst lake patterns carry a signature of climate within them? An analysis of 12 arctic deltas spanning Siberia, Canada, and Alaska, has revealed that lake sizes on arctic deltas universally follow a lognormal distribution, which is associated with a proportionate growth model and average and large lake size increases on colder deltas. Moreover, non-thermokarst waterbodies (i.e. ephemeral wetlands) have a power law size distribution and exhibit no relationship with climate. This climate relationship is associated with thicker permafrost in colder deltas which prevents lakes from connecting to the sub-permafrost aquifer and becoming more ephemeral, decreasing their lateral thermal impact. These results have been published in Geophysical Research Letters.
To further probe this second question, we have been analyzing how lake spatial distribution varies within and across deltas by analyzing local lake spacing at the individual lake, island, and delta scale. This work is utilizing statistical pattern analysis techniques and has thus far found that typically, lakes are found in clusters of similarly spaced lakes, and these clusters likely relate to physical process differences across the deltas. We are corroborating this with further field and remote sensing data, and plan to submit our work for publication soon.
Vulis, L., A. Tejedor, I. Zaliapin, J. C. Rowland, and E. Foufoula-Georgiou, Climate signatures on lake and wetland size distributions in arctic deltas, Geophysical Research Letters, 48, e2021GL094437, doi:10.1029/2021GL094437, 2021.
Vulis, L., A. Tejedor, J. Schwenk, A. Piliouras, J. C. Rowland, and E. Foufoula-Georgiou, Channel network control on seasonal lake area dynamics in arctic deltas, Geophysical Research Letters, 46, doi:10.1029/2019GL086710, 2020
Lakes are non-randomly distributed on major arctic deltas. The real lake distribution of the Kolyma delta is shown along with 4 examples ofrandomly reshuffled lakes, with each lake having a unique, random color. (a) The probability density function (PDF) of the neighborhood distance for each set of reshufflings (black) compared with the true neighborhood distance PDF (red), with significantly smaller mean and larger standard deviation in the real lakes. (b) On every deltas lakes exhibit clustering as evidenced by a smaller mean and higher larger standard deviation (i.e. higher coefficient of variation) in the observed distance distributions compared with the distribution of randomly shuffled lakes (black points). Bars on the black points represent the 95% confidence intervals of the distance distributions for the reshuffled lakes.