Show simple item record

dc.contributor.authorRoth, Aurora
dc.date.accessioned2017-02-14T02:17:41Z
dc.date.available2017-02-14T02:17:41Z
dc.date.issued2016-12
dc.identifier.urihttp://hdl.handle.net/11122/7310
dc.descriptionThesis (M.S.) University of Alaska Fairbanks, 2016en_US
dc.description.abstractMass loss from glaciers in Southeast Alaska is expected to alter downstream environmental conditions such as streamflow patterns, riverine and coastal ecological systems, and ocean properties. To investigate these potential changes under future climate scenarios, accurate climate data are needed to drive glacier mass balance models. However, assessing and modeling precipitation in mountainous regions remains a major challenge in glacier mass balance modeling. We have used a linear theory of orographic precipitation model (LT model) to downscale precipitation from both the Weather Research and Forecasting (WRF) model and the European Centre for Medium-RangeWeather Forecasts interim reanalysis (ERA-Interim) to the Juneau Icefield, one of the largest icefields in North America (4149 km2), over the period 1979--2013. The LT model is physically-based, combining airflow dynamics and simple cloud microphysics to simulate precipitation in complex terrain. Cloud microphysics is parameterized as a function of user-defined snow and rain fall speeds which are then used to calculate the cloud time delay, t, at every time step. We established a model reference run using literature values of snow fall speed and rain fall speed. The model was run using a 1 km digital elevation model and 6 hour timesteps. Due to a lack of precipitation observations, we validated the model with point net accumulation observations along an 8.5 km transect on Taku glacier, one of the largest and best-studied outlet glaciers of the icefield. The observations occurred in late July of 1998, 2004, 2005, 2010, and 2011. We extracted the snow portion from the modeled precipitation and accounted for melt using a temperature-index model prior to comparing results to the observations. The latter was necessary since the observations were taken when substantial melt of the winter snow cover had occurred. The results of the reference run show reasonable agreement with the available glaciological observations (r2 = 0.89). We assessed the LT model results in terms of the icefield-wide average winter (October-March) precipitation amount and its spatial pattern for the 1979-2013 time period. To express the latter we calculated a precipitation index map where each grid cell of average winter precipitation was divided by the icefield-wide spatial mean. The downscaled precipitation pattern produced by the LT model is consistent with the expected orographic precipitation pattern with substantially reduced precipitation on the eastern lee-side portion of the icefield, a pattern that is absent in the coarse resolution WRF and ERA-Interim precipitation fields. To investigate the robustness of the LT model results, we performed a series of sensitivity experiments varying the LT model parameters of snow fall speed and rain fall speed, as well as the horizontal resolution of the underlying grid, and the climate input data. The precipitation pattern produced by the LT model was stable regardless of the parameter combination, horizontal resolution, and climate input data, but the precipitation amount varied strongly with these factors. For the range of snow fall speeds tested and holding all other parameters constant, the average winter precipitation spatial mean varied from 2.5 m to 4.4 m. We were unable to constrain the precipitation amount due to the scarcity of validation data. However, given the stability of the winter precipitation pattern produced by the LT model, we suggest a winter precipitation index map calculated from the LT model reference run results be used in combination with a distributed mass balance model for future mass balance modeling studies of the Juneau Icefield. More observations of total precipitation are needed to further validate the precipitation pattern of the LT model results, constrain the model parameters, and improve the estimation of total precipitation amounts by the LT model. We suggest three locations for potential weather stations that would be most beneficial for validating LT model results. The LT model could be applied to other regions in Alaska and elsewhere with strong orographic effects for improved glacier mass balance modeling and/or hydrological modeling.en_US
dc.language.isoen_USen_US
dc.titleApplying a model of orographic precipitation to improve mass balance modeling of the Juneau Icefielden_US
dc.typeThesisen_US
dc.type.degreemsen_US
dc.identifier.departmentDepartment of Geosciencesen_US
dc.contributor.chairHock, Regine
dc.contributor.committeeTruffer, Martin
dc.contributor.committeeAschwanden, Andy
refterms.dateFOA2020-03-05T12:18:36Z


Files in this item

Thumbnail
Name:
Roth_A_2016.pdf
Size:
6.019Mb
Format:
PDF

This item appears in the following Collection(s)

Show simple item record