(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Surface water area in a changing climate: Differential responses of Alaska’s subarctic lakes [1]

['Danielle L. Rupp', 'Central Alaska Network', 'National Park Service', 'Fairbanks', 'Alaska', 'United States Of America', 'Amy S. Larsen']

Date: 2022-07

Abstract Lake surface area in arctic and sub-arctic Alaska is changing in response to permafrost deterioration, changes in precipitation, and shifts in landscape hydrology. In interior Alaska, the National Park Service’s Central Alaska Network Shallow Lakes program studies lakes and ponds in a wide range of geomorphological settings ranging from alpine lakes to low lying lakes on fluvial plains. The purpose of this study was to determine if and how lake area was changing across this diverse environment. Using the USGS Dynamic Surface Water Extent product, we tested landscape-scale trends in surface water area from 2000–2019 in 32 distinct ecological areas, or ecological subsections, within the three parks. Surface water area declined in 9 subsections, largely in glaciated landscapes with coarse substrates and areas underlain by ice-rich permafrost. Surface water increase was seen in one subsection in the floodplain of the Copper River in Wrangell-St. Elias National Park. No net change was observed in many subsections, but individual lake analysis showed that within several ecological subsections some lakes were increasing in area while others decreased in area, masking changes in lake surface area within the subsection. Over the course of the study period, surface water area in all parks experienced similar fluctuations, likely due to oscillations in regional climate. Periods of high surface water area coincided with relatively warm, wet periods. Climate change models project increases in both temperature and precipitation in Alaska; our results suggest periods of regional wetting may mask longer-term declines in surface water area in some geomorphological settings. Overall, lake surface area declined over the study period; declines were greatest in the Glaciated Lowlands in Denali National Park and Preserve.

Citation: Rupp DL, Larsen AS (2022) Surface water area in a changing climate: Differential responses of Alaska’s subarctic lakes. PLOS Clim 1(6): e0000036. https://doi.org/10.1371/journal.pclm.0000036 Editor: Manish Kumar Goyal, Indian Institute of Technology, INDIA Received: February 22, 2022; Accepted: May 19, 2022; Published: June 17, 2022 This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication. Data Availability: The Dynamic Surface Water Extent product is available at https://www.usgs.gov/landsat-missions/landsat-dynamic-surface-water-extent-science-products. Geospatial data associated with ecological subsections are publicly available at the National Park Service IRMA Portal Data Store: codes 2216201, 2216297, and 2214226. Funding: The author(s) received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist.

1.0 Introduction The hydrologic environment in subarctic Alaska is undergoing rapid change [1], due in part to changes in precipitation patterns [2], groundwater flowpaths [3] and catchment characteristics [4]. A growing number of records indicate that lake surface area has declined over the past 50–100 years in the Subarctic [5–7] and low latitude regions of the Arctic [8]. Changes in lake surface area are attributed to a host of climate-related drivers including permafrost thaw [9], thermokarst [10], and shifts in precipitation regimes [11] as well as terrestrialization, or the infilling of lakes to peatland [7]. Lake surface area at high latitudes is highly dynamic and has historically been regulated by overwinter snowpack [12]. Overwintering conditions are changing in Alaska with warmer winter temperatures [13], a shorter snow season [14], increased seasonal thaw depth [15, 16], and in some areas increased precipitation [2]. However, not all lakes appear to be declining in surface area; for example, lakes in the Arctic found in moraine settings currently appear to be stable, with lake levels oscillating with precipitation [8]. In some regions, where glaciers are actively retreating, lakes are expanding [17, 18]. These differential responses to climate change, combined with the large acreage of subarctic shallow lake habitat (550,000 km2 and >57,000 lakes), make understanding lake succession in interior Alaska an important but challenging task. Interior Alaska lakes are concentrated in three major geomorphological settings: glacial moraines, thermokarst plains and floodplains. Source water inputs (precipitation, groundwater, glacial meltwater, etc.) vary among these three settings, and catchment characteristics (eg. vegetation, topography and permafrost) influence the timing and duration of surface water influx. Research in the low Arctic [8] shows that in glacial deposits lake surface area is correlated to precipitation, whereas lakes in thermokarst depressions are influenced by developmental stage: initial inception, expansion, and reduction by either catastrophic drainage or by becoming connected to groundwater [19–22]. Surface area extent in riverine systems is governed by groundwater or flooding and is dependent upon stage, sediment porosity and hydraulic conductivity [23]. These lakes tend to have discontinuous permafrost and once isolated can transition to thermokarst lakes or infill with vegetation [24]. These geomorphological differences create a complex mosaic of lakes—often within close proximity to each other—that will respond differently to change. Shallow lakes provide critical habitat for wetland obligate species and understanding long-term patterns in distribution and abundance of lakes is critical to land management agencies. To better understand the current status of lakes and to identify important drivers of lake change we have developed a study to track changes in lake surface area across three national parks and preserves in subarctic Alaska. We identified four primary objectives of this study: 1) detect trends in lake surface area annually from 2000–2019; 2) identify geomorphological patterns of change; 3) relate surface area change patterns to environmental climate variables; and 4) evaluate fine scale changes in lake surface area. Results from this study will be used to enhance our understanding of lake surface water dynamics in the parks and in subarctic Alaska. 1.1 Study area The study was conducted by the Central Alaska Network (CAKN) Inventory and Monitoring program which consists of three national park units located in interior Alaska: Yukon-Charley Rivers National Preserve (YUCH), Denali National Park and Preserve (DENA), and Wrangell-St. Elias National Park and Preserve (WRST, Fig 1). These subarctic parklands contain numerous lakes distributed across a large latitudinal and longitudinal gradient that extends from southeast Alaska to eastern interior Alaska, crossing a diverse range of permafrost conditions (continuous to sporadic), geologic settings, and climate regions. Surface water area trends were tested in pre-defined ecological subsections [25–28] rich in surface water, which were previously assigned based on geology, landforms, soils and vegetation. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. Ecological subsections within CAKN parks with >0.1% surface water area overlaid with Alaska climate regions [ Ecological subsections within CAKN parks with >0.1% surface water area overlaid with Alaska climate regions [ 29 ]. Basemap provided by the USGS National Map, https://basemap.nationalmap.gov/arcgis/rest/services/USGSShadedReliefOnly/MapServer. https://doi.org/10.1371/journal.pclm.0000036.g001 Lakes in these areas are concentrated in glacial moraines, thermokarst plains and floodplains. The Yukon River Valley and its associated floodplains contain most of the lakes found in YUCH, making up a large portion of riverine-influenced lakes within this study. Numerous lakes are also found in the Copper and Bremner River basins in WRST, and scattered along rivers throughout DENA. Glacial processes formed most lakes within DENA and WRST including large glacial scour lakes such as Tanada, Copper and Wonder Lakes, as well as small lakes distributed throughout the glacial moraines. Permafrost is present in all three parks, and thermokarst lakes are common in YUCH and DENA. The parks span a range of localized interior climates including the Northeast Interior, Central Interior, Southeast Interior, Cook Inlet, and Northeast Gulf climate regions (Fig 1) [29]. These climates vary from extreme annual temperature swings and low precipitation in the Interior to milder temperatures and high precipitation in the southern climate zones near the coast. However, the climate in these parks is changing. Mean annual temperature (MAT) increased significantly at long-term National Weather Service stations near YUCH and DENA (p<0.01, Mann-Kendall trend test) between 1980–2019, and the growing season (June-September) temperature has increased in WRST (p = 0.04) although MAT has not changed significantly (S1 Fig). However, 2014–2019 was significantly warmer than the long-term average [16]. No significant trends in precipitation have been observed between 1980–2019; however, potential evapotranspiration rates increased in all parks (p≤0.01, Mann-Kendall trend test, S2 Fig), and maximum snowpack in YUCH has decreased significantly (p = 0.02, S3 Fig). Discharge from snowmelt dominated rivers in Alaska has been declining for the past 50–60 years, largely due to decreased flow during the snowmelt period and summer months [30] (S4 Fig).

4.0 Discussion For central Alaska parklands, long-term regional climate trends may be more important than short term variations in precipitation and temperature in determining lake surface area extent. Between 2000 and 2019, we observed large scale oscillations in lake surface area throughout the entire study region. Surface water area increased during warm and wet periods, peaking at the end of the warm period, and decreased during cool periods, regardless of precipitation patterns (Fig 6). Surface water dynamics at the subsection-level, however, appear more dependent upon complex interactions among climate variables, local groundwater, glacial meltwater supplies, river discharge, etc. Several subsections within the parks undergoing surface water declines departed from this pattern in 2014 and continued to decline, despite the warming and wetting trend occurring on the landscape. In other areas, some individual lakes lost surface water while others gained surface water, likely resulting in a masking of change in some subsections. Overall, decreases in surface water area exceeded gains in both the ecological subsections and individual study lakes, corroborating recent findings of overall lake area loss in the Subarctic and Arctic over a similar time period [42]. Surface water area decreases occurred most often in glacier and thermokarst-affected landscapes. 4.1 Glacial lake succession Significant changes in lake surface area occurred in 10 of the 32 ecological subsections. Changes were most apparent in glaciated landscapes of DENA and WRST. In agreement, Riordan et al. (2006) [43] found surface area declines in interior Alaska including DENA and WRST between the 1950s and 2002. Arctic and subarctic landscapes, including DENA, are greening, with a documented increase in shrub and tree growth in glaciated regions [44–48]. Encroachment of shrubs and trees is a natural part of the succession of glaciated landscapes and is accelerated by warming temperatures. Trees and shrubs can intercept 15–30% of water moving across the landscape, increasing transpiration and biological water demand within watersheds and decreasing the volume ultimately received by water bodies [49]. Lake surface area declines in glaciated regions may be influenced by greening that is occurring here. Additionally, numerous historic lake records demonstrate that warming coincides with an intensification of lake terrestrialization and infilling rates in glacially-formed lakes [50–53]. This could be happening in the glaciated lowlands in DENA where the last glacial maximum was 26,500–19,000 years ago [54], as well as in subsections not influenced by glacial activity. 4.2 Thermokarst lake succession Significant declines in surface water area also occurred in three subsections containing thermokarst lakes or ice-rich permafrost. Wetting conditions in other regions have been shown to degrade permafrost in the long-term [15], and warm periods are directly responsible for the degradation of permafrost and glaciers [16]. More individual study lakes in YUCH increased in surface water area than in both WRST and DENA. This may be due to the more extensive ice-rich permafrost found here; lake expansion during this period may be an indication of additional permafrost degradation. At least four other subsections with discontinuous permafrost exhibited masking of change, including a thaw lake basin along the Yukon River and the Eolian Lowlands in DENA (Fig 7). Because thermokarst activity can expand lakes and create new ponds and wet areas, thermokarst-affected areas may be the most likely to mask overall change trends as increases and decreases in surface water area create no net surface water change over a geographic area. Masking has likely caused a significant underestimate of the changes occurring in permafrost affected landscapes in this study. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 7. An upland thermokarst lake in the Eolian Lowlands ecological subsection which lost surface water area 2000–2019. Denali National Park, 2008. https://doi.org/10.1371/journal.pclm.0000036.g007 4.3 Riverine lake succession Fourteen subsections within this study, containing at least 1,770 lakes, fell in areas of riverine influence. Of these, five subsections decreased and one subsection increased in surface water area over the study period. Subarctic riverine lakes are generally surrounded by discontinuous permafrost and typically in contact with groundwater for at least part of the year, which can drive lake surface area change [7]. The Yukon River Valley in YUCH contains approximately 567 lakes, over half of which are surrounded by permafrost. As permafrost thaws, groundwater connectivity to rivers including the Yukon is increasing [55–57]. This increased connectivity could potentially 1) increase underground recharge to floodplain lakes or 2) drain the landscape, exporting water to the river and carrying it downstream [3, 58]. These differential effects may be why masking of lake area change was seen in some riverine subsections in this study. 4.4 Other factors Lake surface areas were stable in much of the landscape; we detected no significant long-term trend in surface water area in 22 ecological subsections. The continued widespread presence of permafrost and glaciers still support a wet landscape via the prevention of subsurface drainage and addition of meltwater, respectively. Montane snow and glaciers hydrologically feed the landscape throughout the growing season and likely have an important impact on the parks’ surface water dynamics. However, glaciers throughout Alaska are receding and glacial extent has declined significantly over at least the last 60 years [59]. Lack of understanding of these flow paths impair our ability to determine if glacial meltwater is compensating for possible long-term landscape drying trends. As climate models project Alaska to become warmer and wetter, it is possible that we could see a landscape wetting before we cross a permafrost/glacial deterioration threshold and landscape-wide drying. Additionally, we did not consider the effects of wildfire history on lake surface area in this study, which has been shown to affect rates of lake surface area change in permafrost areas [60]. The accuracy of the DSWE product could influence surface water estimates; for example, we noticed that dense floating vegetation (e.g. pond lilies) obscured the detection of surface water in a few instances. However, the DSWE is available for the entire state which is difficult to attain in such a large landmass, and provides a promising resource for surface water change. The influence of any of these factors on lake area change in the Subarctic should continue to be pursued in future study.

5.0 Conclusions In Alaska’s interior parks, more ecological areas and individual lakes are losing surface water area than are gaining. Lakes found to be most susceptible to surface water area decreases appear to be related to the evolution of glacial and permafrost landscapes, although masking of overall change likely occurred in thermokarst-affected areas. Since 2014, mean annual air and ground temperature trends have dramatically increased, which will lead to significant permafrost degradation in Alaska’s national parks [16, 61]. Precipitation rates are also expected to increase in most of Alaska [2], which may accelerate permafrost thaw [15]. In this study, it appears that the interaction of these two climate variables (warming, wetting) could lead in the short-term to a wetter landscape overall as lakes endure a lag effect before expected surface water loss. As the climate continues to warm, enhanced groundwater connectivity could lead to both the stabilization or growth of some lakes and the drainage or infilling of others. The thaw of areas with coarse substrates (gravel, sand) are more likely to lose surface water area or perhaps transition to peatland. Whether situated in glacial, thermokarst, or river-affected landscapes, shallow lake ecosystems provide critical habitat for wildlife, including many species of migratory birds. The Park Service also values these habitats as subsistence, cultural, recreational, and aesthetic resources within its jurisdiction. The greatest surface water losses appear to be occurring in DENA, which is the most well-known and visited parkland in Alaska. Loss of aquatic habitat could affect important breeding populations of the many species that use shallow lakes and parkland users alike. As landscape infrastructures of the Subarctic—permafrost and glaciers—degrade with a changing climate, differential responses of surface water area to warming temperatures and increased precipitation will make surface water area predictions difficult. Modelling efforts to predict landscape change tend to generalize ecological shifts, and it is important to continue monitoring change in variable subarctic settings in a quantifiable manner. Landscape and water area change is complex and updating models and scenarios with observational data is essential for proactive management and adaptation.

Acknowledgments The authors would like to thank Dave Swanson and others who developed the ecological subsection maps. Dave also provided important technical support and guidance and provided internal review. Maggie MacCluskie provided internal review and guidance. Heidi Kristensen delineated individual study lake boundaries in ArcMap and developed key data management tools. Josh Schmidt provided suggestions and guidance for statistical analysis of environmental variables. Matt Macander evaluated the use of the DSWE for this application and partnered with us to modify existing standard operating procedures.

[END]
---
[1] Url: https://journals.plos.org/climate/article?id=10.1371/journal.pclm.0000036

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.