About OpenKnowledge@NAU | For NAU Authors

Three-component hydrograph separation for the glaciated Lake Peters catchment, Arctic Alaska

Ellerbroek, Rebecca Anne (2018) Three-component hydrograph separation for the glaciated Lake Peters catchment, Arctic Alaska. Masters thesis, Northern Arizona University.

[img] Text
Ellerbroek_R_2018_Three-component_hydrograph_separation_Lake_Peters.pdf - Published Version
Restricted to Repository staff only

Download (15MB) | Request a copy
[img] Spreadsheet
ELLERBROEK_THESIS_APPENDIX_I.xlsx
Restricted to Repository staff only

Download (1MB) | Request a copy
[img] Spreadsheet
ELLERBROEK_EKLUTNA_APPENDIX_II.xlsx
Restricted to Repository staff only

Download (17kB) | Request a copy

Abstract

Arctic lakes and rivers are fed by a variety of sources, including rain, snow, glacial melt, and groundwater. Due to sustained Northern Hemisphere warming associated with industrial- era atmospheric greenhouse gases, those water sources are undergoing rapid changes. However, there are relatively few studies quantifying the hydrological components of Arctic Alaskan watersheds. To estimate water source contributions in an Arctic Alaskan catchment, I conducted three years of hydrological and climatological monitoring in the glaciated Lake Peters watershed in the northeastern Brooks Range. Lake Peter’s two main inflows, Carnivore Creek and Chamberlin Creek, provide a useful comparison of catchment size, glacial coverage, surficial materials and slope. The Chamberlin watershed is small, steep, rocky, and heavily glaciated (8.1 km2, 25.9% glaciated) whereas the Carnivore watershed is larger and less glaciated (128.9 km2, 9.5%) and its channel flows through a wider and less steep valley bottom. Weather stations continuously recorded precipitation and temperature from May 2015 to August 2017, and Chamberlin and Carnivore Creeks were monitored from May through August of 2015, 2016 and 2017 to produce high-resolution discharge and conductivity records. The two streams were sampled intermittently, and source water samples were collected in the form of rain, snow, glacial ice and melt. Water samples (n = 187) were analyzed for stable oxygen isotopes, which were used in conjunction with conductivity as tracers for a three-component hydrograph separation. The three discharge components are defined as rainfall runoff (rainwater from recent summer storms), snowmelt (melting snowpack from within previous year), and an "old water" component (water residing in the catchment for longer than one year regardless of original source, including soil water, groundwater, and glacial melt). Carnivore Creek’s average discharge was an order of magnitude greater than Chamberlin Creek’s, at 9.4 and 0.7 m3/s respectively. Divided by catchment area, Chamberlin Creek’s specific discharge (1892 mm) was considerably larger than Carnivore Creek’s (1524 mm) for the three summers combined. Broad year-to-year similarities between the two rivers include a high responsivity to rain storms, a strong diurnal cycle, and an overall pattern of low discharge from May to mid-June increasing to peak flows in July and August. Differences between the two creeks are accentuated by higher temperatures. The two hydrographs are most similar in the coolest summer (2015) and diverge most during the warmest summer (2017), when Carnivore Creek recorded its lowest total discharge and Chamberlin Creek recorded its highest. This is attributed to Chamberlin Creek’s larger glacier coverage. The mixing model results show that, averaged over the three-summer study period, Chamberlin Creek is 43 ± 12% rainfall runoff, 26 ± 8% snowmelt, and 31 ± 19% old water, whereas Carnivore Creek is 38 ± 8% rainfall runoff, 19 ± 8% snowmelt, and 43 ± 13% old water. The percentages of each component vary from year to year, and by day of year (DOY). The proportion of old water in Carnivore Creek correlates positively with DOY (r = 0.48, p < 0.001), and snowmelt correlates negatively (r = −0.73, p < 0.001). There is no relation between DOY and rainfall runoff in Carnivore Creek, but Chamberlin Creek’s rainfall runoff increases linearly through the summer (r = 0.58, p < 0.001). Chamberlin Creek’s old water and snowmelt have nonlinear relations with DOY best described using second-order polynomial equations. The snowmelt component has a concave-downward trend (r = −0.78, p < 0.001), starting the summer with less than 15%, peaking at 42% in the mid-summer, then lowering to around 20% in August. Inversely, the old water component has a concave-upward trend (r = 0.79, p < 0.001), with peaks of 50−70% at the beginning and end of the summers, and lows of 10% in the mid- summer. Each modeled discharge component is compared with an independent observation of its associated water source. Cumulative summer rainfall is extrapolated from three weather stations along an elevational gradient throughout the catchment. Snowpack and glacier elevation changes are calculated by airborne photogrammetry, then converted to water equivalence (w.e.) using measured snow density (0.33 g/cm3) and known density of glacial ice (0.917 g/cm3). The rainfall observations are greater than, or within uncertainty of, the modeled rainfall discharge component for both rivers for all three summers. The snowpack w.e. observations also exceed their associated modeled discharge components for Carnivore Creek in 2015 and 2017, and Chamberlin Creek in 2015. For the other years, snowpack w.e. observations underestimate the modeled snowmelt by half. This study does not distinguish between glacier melt and subsurface flow in the old water component, because the isotopic composition of the two is indistinguishable. An additional tracer value is required to chemically differentiate glacial melt from subsurface flow. The old water percentage increased from the beginning to end of the summer and the highest proportion of glacier melt occurred during the warmest summer. With the seasonal increase in old water discharge being attributable to glacial melt, it is clear that the glaciers are a key source of water storage throughout warm and dry parts of the summer. If glacier contribution ceases completely, total discharge could be reduced by half. Combined with the documented mass loss of glaciers throughout the Arctic, these findings emphasize the need for further hydrology research.

Item Type: Thesis (Masters)
Publisher’s Statement: © Copyright is held by the author. Digital access to this material is made possible by the Cline Library, Northern Arizona University. Further transmission, reproduction or presentation of protected items is prohibited except with permission of the author.
Keywords: Alaska; Arctic; glacier; hydrograph separation; isotope; mixing model
Subjects: Q Science > QE Geology
NAU Depositing Author Academic Status: Student
Department/Unit: Graduate College > Theses and Dissertations
College of the Environment, Forestry, and Natural Sciences > School of Earth Sciences and Environmental Sustainability
Date Deposited: 22 Apr 2021 19:37
URI: http://openknowledge.nau.edu/id/eprint/5276

Actions (login required)

IR Staff Record View IR Staff Record View

Downloads

Downloads per month over past year