Thermokarst distribution and relationships to landscape characteristics in the Feniak Lake region, Noatak National Preserve, Alaska

Balser, A; Gooseff, M; Jones, J; Bowden, W

1.0 ABSTRACT Climate warming in the arctic is leading to loss of permafrost and wide-scale ecosystem alteration(ACIA 2004). Permafrost thaw often results in thermokarst, involving intensive and catastrophic loss of soil structure and subsidence(Jorgenson et al. 2008). Regional-scale distribution of thermokarst features is poorly documented throughout the arctic, and correlations with landscape variables are poorly understood(Osterkamp 2007). The Feniak Lake area, within the Noatak Basin in northwestern Alaska’s Brooks Range mountains, harbors a transitional landscape from arctic and alpine tundra to boreal shrubland (Young 1974). Field investigations augmented by photogrammetric measurements in 2006 reveal patterns in the distribution of classifiable thermokarst failure types, and provide data on the physical and chemical impacts these features have on aquatic systems. Distinct thermokarst types show significant relationships with local landscape variables such as slope and vegetation, and with regional variables including lithology, surficial geology, glacial history and landcover. Frequency of thermokarst features has increased markedly within this study area in the past 25 years. Analysis of current and historical aerial photography shows a two fold increase in number of thermokarst features, and in total surface area of affected landscape. The majority of these previously unreported features occur in headwaters of Noatak River tributaries, where they can have marked impacts on small headwater streams which have less buffering capacity for disturbance. These results demonstrate that thermokarst, especially in headwater regions, has been vastly under-reported in the Noatak Basin. These findings also suggest that similar phenomena may be under-reported in other permafrost regions as well, due in part to the logistical difficulty of conducting quantitative surveys in remote areas with rugged topography. Significant associations among thermokarst types and landscape parameters suggest that a predictive model to identify thermokarst-susceptible terrain should be viable in the future.