Mountain Scholar :: Browsing by Author "Fassnacht, Steven R., advisor"

Browsing by Author "Fassnacht, Steven R., advisor"

Now showing 1 - 20 of 20

Open Access
A modeling approach to estimating snow cover depletion and soil moisture recharge in a semi-arid climate at two NASA CLPX
(Colorado State University. Libraries, 2004) Holcombe, Julie D., author; Fassnacht, Steven R., advisor
Snow cover depletion and soil moisture recharge are small segments, but crucial hydrological components for cryospheric regions of the earth. The abilities of a one-dimensional mass and energy balance model (SNTHERM) to predict snow cover depletion and Fast All season Soil STrength (FASST) to model the evolution of soil moisture recharge based on observed data from two NASA Cold Land Processes Experiment (CLPX) sites were evaluated. The objective was to investigate both model accuracies in predicting the observed parameters at Buffalo Pass near Steamboat and Illinois River located in North Park, both of which are located in the Colorado Rocky Mountains and are known for their differences in terrain and weather conditions. The results from SNTHERM and FASST and the model performance statistics illustrate that the models overall fit to the observations were excellent at both locations. SNTHERM predicted the snow cover depletion date two days later than the observations at Buffalo Pass and only one day prior to the observations at Illinois River. The timing of snow accumulation and melt at Illinois River was in agreement with the observations at Illinois River, but the magnitude of snow depth was incorrect. The shallow and patchy nature of snow cover and the inconsistent meteorological parameters were problematic for SNTHERM. FASST correctly predicted the magnitude of seasonal soil moisture storage at both sites, but soil moisture recharge prediction was challenging for the model. A lateral flow module and thorough soil data are thought to improve FASST's capability to predict the timing of soil moisture change. SNTHERM and FASST prove to possess the ability to predict snow cover depletion and seasonal soil moisture storage at two radically different field sites.
Open Access
Assessing flow alteration and channel enlargement due to dam management at Hog Park Creek, Wyoming
(Colorado State University. Libraries, 2016) Carleton, Tyler J., author; Fassnacht, Steven R., advisor; Butters, Gregory, committee member; Stednick, John D., committee member
As part of a complex water exchange agreement, Little Snake River water is piped through the Continental Divide and released into Hog Park Creek to replace over-appropriated North Platte River piped to Cheyenne, Wyoming. The Little Snake River water, in addition to native flows, has used Hog Park Creek as a conduit since the 1960s. As a result, Hog Park Creek has continued to enlarge. This study assesses flow alterations and channel enlargement at Hog Park Creek due to dam management. To assess flow alterations at Hog Park Creek without a pre-dam daily flow record, the Precipitation-Runoff Modeling System (PRMS) simulated natural flows from 1995 to 2015. A regionalization technique transferred calibrated parameters to Hog Park Creek model parameterization from Encampment River model parameterization. Along with the simulated natural flows, reference flows were used to compare to the post-dam flow record. All comparisons indicate the greatest flow alterations were winter and spring monthly flows and low flows. The April median flows and 7-day low flows more than tripled. To a lesser degree of deviation, significant flow alterations included peak flow alterations such as greater magnitude, longer duration, increased frequency, earlier peak flow timing, and faster fall rates. In addition, flow alterations due to climate were assessed. The climate trends reflect warmer-wetter climate change with a shift to earlier peak flows. However, these flow alterations are minor compared to those by dam management. The climate projections compared historic (1980-1999) and future (2040-2059) PRMS simulated natural flows using warmer-wetter and -drier scenarios. Both scenarios project more frequent, flashier peak flows. The warmer-wetter scenario also projects a shift to earlier peak flows. This projected shift of peak flows to mid-May is earlier than the current artificial peak flows in late-May and the natural peak flows in early June. Channel enlargement measured at Hog Park Creek is consistent with qualitative channel response for increased flows and sediment loads less than sediment transport capacity. Stream surveys from 2006 and 2015 measured irregular channel widening and bed degradation. The riffle cross-sections (XSs) measured little change while pool XSs at the maximum point of scour measured extensive widening (+ 3.6 m). Ecologic implications of continued channel enlargement were evaluated by modeling changes in water surface elevations using the Hydrologic Engineering Center River Analysis System (HEC RAS). Between 2006 and 2015, modeling indicated a decrease in water surface elevation by 3 cm per decade and a decrease in flood inundation area of 70 m2 per 1 m of stream length per decade. Additionally, the hydraulic modeling results support the theory that alluvial channel form is most influenced by bankfull flow, which in this case is the 1.5-year flood. Based on this agreement, modeling indicated channel enlargement began near a pre-dam bankfull flow of 3.8 m3 s-1 (135 ft3 s-1) and has since increased to 5.5 m3 s-1 (195 ft3 s-1) in 2015. A possible trajectory of channel enlargement is to a bankfull flow of 5.8 m3 s-1 (205 ft3 s-1), which is based on the 1.5-year flood since dam enlargement in the 1980s. However, without a stable flow regime, a stable channel form is not possible. Thus, to improve aquatic and riparian habitat, a stable flow regime and channel form will be necessary. For this reason, recommendations for a modified flow regime based on the findings of this study are developed and can be used as guidance for adaptive management.
Open Access
Assessment of digital land cover maps for hydrological modeling of the Yampa River Basin, Colorado, USA
(Colorado State University. Libraries, 2005) Repass, Julie Mae, author; Fassnacht, Steven R., advisor
In order to produce satisfactory results from hydrologic models, it is imperative to use good input data. Today there is a multitude of different land cover maps available, and determining which input data map for the model can be unclear. The goal of this study was to quantify the differences between several readily available land cover maps to determine their relative suitability for hydrological modeling of the Yampa River Basin, Colorado. The land cover maps compared in this study are derived from Advanced Very High Resolution Radiometer (AVHRR), Landsat Thematic Mapper (TM), and Moderate Resolution Imaging Spectroradiometer (MODIS) imagery. These maps were compared to a 30-m land cover map modeled from ground data, Landsat imagery, and MODIS imagery, all collected in 2004. This map was regarded as "truth" in this study due to its fine resolution and use of recent ground data and imagery, and was used to rank the public domain land cover data sets. In order to compare the different land cover data sets, all data were first degraded to a common spatial resolution (~30-m) and a common species resolution. Once this was accomplished, the maps were assessed on four levels. The four assessments were based on: (i) the relative agreement of the total aggregated land class percentages after the data had been cross-walked with respect to the reference map; (ii) pixel accuracy; (iii) scene accuracy; and (iv) cumulative streamflow model output from the United States Geological Survey (USGS) Precipitation-Runoff Modeling System (PRMS) in relation to observed cumulative streamflow. The results showed that the pixel and scene accuracies did not correlate with model performance within the Yampa River Basin using the PRMS model. The qualitative comparison of the total aggregated land class percentages helped explain the general trends in the simulation results. It was found that maps with the correct proportion of forested and non-forested areas generally had simulated cumulative streamflow that matched closest to observed cumulative streamflow. Overall, the MODIS-derived land cover maps performed the best in terms of hydrological modeling using PRMS in the Yampa River Basin. However, the model was not found to be particularly sensitive to accurate land cover conditions. As a result, the scene and pixel accuracy results would not necessarily correlate with the model results.
Open Access
Evaluation of ultrasonic snow depth sensors for automated surface observing systems (ASOS)
(Colorado State University. Libraries, 2005) Brazenec, Wendy Ann, author; Fassnacht, Steven R., advisor; Doesken, Nolan, committee member; Kelly, Gene, committee member; Stednick, John, committee member
In the 1990's the National Weather Service deployed automated surface observing systems at hundreds of airport locations across the country. Prior to the automation, human observers made snow observations every six hours. Once the automated systems were deployed, snow measurements ceased due to the lack of an automated sensor to measure snow. This study explored how well ultrasonic snow depth sensors compared to manual snow observations at nine sites across the country. This study had four objectives: 1.) Develop a method of quality assurance and quality control 2.) Identify factors which affect sensor performance 3.) Compare automated sensors to manual observations of snow depth 4.) Derive an algorithm to estimate six hour snowfall from automated sensor snow depth. A reliable data smoothing/processing technique was achieved using filtering of large variability and smoothing with a moving average to smooth small variations in snow depth. Factors found to affect sensor performance included: snow crystal type, wind speed, blowing/drifting snow, uneven snow surface, extremely low temperatures, and intense snowfall. The Judd and Campbell sensors both did a satisfactory job measuring snow beneath the sensor within ±0.4 inches. Two separate algorithms were created due to differing degrees of precision between the two sensors. It was found that the Campbell sensor did a better job at estimating six hour snowfall than the Judd using an algorithm that calculated snowfall over 5 minute periods and applying a temperature based compaction model to the estimated snowfall. The Campbell agreed with the manual data with an average mean absolute error between measurements of 0.23 inches. The Judd sensor results improved by using an algorithm which calculated snowfall using the change in snow depth over sixty minutes, however, the Campbell results were better using the five minute snowfall algorithm. Overall, both sensors accurately depicted the snow depth on the ground, however the Campbell sensor was more accurate at predicting six hour snowfall using the algorithms presented in this research.
Open Access
Examining trends in snowmelt contribution to streamflow in the southern Rocky Mountains of Colorado
(Colorado State University. Libraries, 2016) Pfohl, Anna K. D., author; Fassnacht, Steven R., advisor; Stednick, John D., committee member; Niemann, Jeff, committee member
Snowmelt contribution to streamflow in snow-dominated watersheds has largely been limited to using the Center of Volume method, which looks at the day at which a certain amount of flow has passed, typically 20%, 50%, and 80%, referred to as tQ20, tQ50, and tQ80, respectively. We developed a new method to measure streamflow timing in the Southern Rocky Mountains of Colorado for 39 gauging stations from 1976 to 2015. We first manually extracted start and end days from the annual hydrograph of a small, medium, and large watershed to use as "truth." We then looked at the cumulative annual hydrograph and then found average spring and late fall baseflow. Using these average baseflows, we plotted the cumulative baseflow against the cumulative hydrograph and determined that the start and end of snowmelt contribution, tstart and tend, occurred when the cumulative hydrograph departed from the cumulative baseflow by a given baseflow factor. Using NSE and RMSE values, we determined that 10x and 17.5 baseflow were able to best represent the manually extracted values. NSE values ranged from 0.59 to 0.6 and 0.53 to 0.69 for tstart and tend, respectively; RMSE values ranged from 5.42 to 7.7 and 6.32 to 8.00, for tstart and tend, respectively. In comparison, NSE values ranged from -4.73 to -25.35 and -5.87 to -13.25 for tQ20 and tQ80, respectively; RMSE values ranged from 29.33 to 43.19 and 33.01 to 34.94 for tQ20 and tQ80, respectively. This new automated method was able to better predict values of start and end than what has been commonly used in the literature. We identified other variables related to snowmelt timing to streamflow, including the percent of flow and volume at the estimated tstart and tend, as well as the total duration of contribution. We used the correlation coefficient to help explain the variance in the observed trends of the different snowmelt timing variables, using different physiographic characteristics (mean slope, mean elevation, mean solar radiation, latitude, and longitude) as well as trends in winter precipitation and summer NDVI. Most of these trends were not statistically significant, but mean slope was best able to explain the variance in trends for tend, Q100, Qend, Qduration, %Qtend, and tQ80 (p < 0.05).
Open Access
Exploration of a geometric approach for estimating snow surface roughness
(Colorado State University. Libraries, 2015) Kamin, David Jeffrey, author; Fassnacht, Steven R., advisor; Stednick, John D., committee member; Bauerle, William, committee member
The roughness of a surface that influences atmospheric turbulence is estimated as the aerodynamic roughness length (Z0), and is used to understand the flow of air, temperature, and moisture over a surface. Z0 is a critical variable for estimating latent and sensible fluxes at the surface, but most land surface models treat Z0 simply as a function of land cover and do not address the variability of this value, such as due to changing snow surfaces. This is due in large part to the difficulty and cost of obtaining reliable estimates of Z0 under field conditions. This work addresses the need for versatile methods to evaluate snow surface roughness on a plot-scale. This study used anemometric data from a meteorological tower near Fort Collins, Colorado over two winters (2013-2014). Thorough screening yielded 153 wind-speed profiles which were used to calculate the aerodynamic roughness length at different times and under different snow conditions. The anemometric Z0 values observed in this study with changing surface conditions ranged by 2.5 orders of magnitude from 0.2 to 52 x 10-3m. Concurrently, a terrestrial laser scanner was used periodically to measure surface geometry and generate point clouds across the study site. Point clouds were processed and interpolated onto a regular grid for estimation of Z0 based on the geometry and distribution of surface roughness elements. Two different geometric evaluations, the Lettau and Counihan methods, were used for the estimation of Z0. The estimates based on surface geometry were evaluated and compared to anemometric Z0 values calculated from field observations of wind turbulence across the surface of the study site. The Lettau method Z0 values compared well to the measured anemometric results, with low but acceptable Nash-Sutcliffe Efficiency Coefficient (NSE) of 0.14 and a strong coefficient of determination (R2 = 0.90). While the NSE was small, the Lettau Z0¬ values could easily be scaled to the anemometric Z0. The Counihan method yielded less accurate results compared to the anemometric data, with a NSE of -1.1. The data also showed a strong correlation between Z0 and changing snow cover. The coefficient of determination between Z0 and snow-covered area for both the anemometric and Lettau methods was greater than 0.7, indicating that both methods responded well to changing surface conditions.
Open Access
From the tree to the forest: the influence of a sparse canopy on stand scale snow water equivalent
(Colorado State University. Libraries, 2007) Ewing, Patrick John, author; Fassnacht, Steven R., advisor
The canopy of an individual tree has a negative effect on the accumulation of snow around tree boles, resulting in a decrease in snow depth inward from the edge of the canopy to the tree trunk. This influence of trees on snow distribution affects the total volume of water stored in the snowpack, especially for a sparse forest stand. However, snow measurements, in particular depth, are typically made between trees, and this neglects the decreased accumulation around trees. As well, little is known about changes in snowpack density under the canopy compared to between trees. Sparse individual trees have their own microclimate (energy balance, wind profiles, etc.) that could produce directional variations in snowpack properties. To establish how the decreased snow depth and possibly change in snowpack density under the canopy can affect estimates of stand scale SWE, depth and density measurements were taken in the four cardinal directions around three Picea engelmanii and two Abies lasiocarpa during the winters of 2005 and 2007 near Cameron Pass, northern Colorado. These near tree measurements were assessed against existing snow depth models and superimposed on a 50-m transect of depth measurements taken at 0.5-m intervals. Three scenarios of a sparse forest were considered: one tree with a 1-m canopy radius, one tree with a 2-m canopy radius, and three trees each with a 2-m canopy radius. Directionality was observed in the snow depth increasing away from each tree. An increasing trend in snowpack density was observed outward from each tree. The estimated average snow water equivalent for the transect decreased by 14.4% with the addition of three trees with 2-m canopy radii.
Open Access
Geostatistical methods for estimating snowmelt contribution to the seasonal water balance in an alpine watershed
(Colorado State University. Libraries, 2006) Hultstrand, Douglas M., author; Fassnacht, Steven R., advisor; Stednick, John, advisor; Doesken, Nolan, committee member; Musselman, Robert, committee member
The performance of nine spatial interpolation models was evaluated to estimate snowmelt contributions to streamflow in the West Glacier Lake watershed (0.61 km2), in the Snowy Range Mountains of Wyoming. Streamflow from the West Glacier Lake watershed has been previously estimated at 40% to 130% greater than measured precipitation inputs. Additional input into the watershed had been attributed to a permanent snowfield in the upper portion of the watershed covering approximately 2.4% of the watershed area. However, the excess output may be a result of inaccurate estimation of water quantities using current precipitation and stream gauging methods. In April 2005, near peak accumulation snow depth measurements and snow density measurements were collected within West Glacier Lake watershed. The distribution of snow water equivalent (SWE) was calculated as the product of snow depth, snow density, and snow-covered-area (SCA). Snow depths were spatially distributed throughout the watershed through nine spatial interpolation models. Snow densities were spatially distributed through a multiple linear regression. The nine spatial snow depth models explained 18% to 94% of the observed variance in the measured snow depths. Co-kriging with solar radiation produced the best results explaining 94% of the observed variance in snow depth measurements. The annual water balance, expressed as equivalent water depths for water year 2005, was total precipitation (1,481 mm), snowpack sublimation (251 mm), and streamflow (1,000 mm), resulting in an evapotranspiration estimate of 230 mm. Estimated SWE from the field survey data was 67% greater than precipitation gauge estimates and accounted for 85% of the annual streamflow. Summer precipitation was not a significant contributor to the annual hydrograph and was also less than snowpack sublimation. Precipitation gauge values were unrepresentative of actual precipitation depths, and several spatially distributed snow depth models provided better estimates of precipitation inputs.
Open Access
Instream flow methodologies: an evaluation of the Tennant method for higher gradient streams in the national forest system lands in the western U.S.
(Colorado State University. Libraries, 2006) Mann, Jennifer, L., author; Fassnacht, Steven R., advisor; Merritt, David, committee member; Rathburn, Sara, committee member
In 1976 Donald Tennant introduced a method for determining instream flow requirements for fish, known as the 'Montana method', or more commonly the Tennant method. The method uses a percentage of average annual flow (AAF) to determine fish habitat quality. From 58 cross sections from 11 streams in Montana, Nebraska, and Wyoming, Tennant concluded that 10% of AAF is the minimum for short term fish survival, 30% of AAF is considered to be able to sustain fair survival conditions, and 60% of AAF is excellent to outstanding habitat. These quantities are employed internationally, regardless of physical and hydrologic setting, due to the simplicity of using only the average annual hydrograph. The purpose of the current study was to determine under what conditions Tennant's fixed percent AAF values apply, to specifically evaluate Tennant's original width, depth, and velocity measurements, to evaluate the applicability of Tennant's percent of AAF, as compared to other methods of determining minimum instream flows, and to determine if there are regional characteristics that relate to the applicability of the Tennant method. Tennant's method was tested to see if percent AAF actually can be used as a surrogate for other hydraulic measures, such as width, depth, and velocity. These physical parameters have been used in other studies to quantify instream flow used for fish. The two other methods that were used in the comparisons were the wetted perimeter method and the physical habitat simulation system (PHABSIM). A set of regional characteristics were used to look for region specific patterns. These characteristics including: stream type, state, ecoregion, and hydro-climatic regime. A total of 151 cross sections were analyzed on seventy river segments throughout the western U.S. (California, Colorado, Idaho, Montana, Oregon, Utah, and Washington). The streams were classified as pool-riffle, plane bed, step-pool, and dune-ripple. This study will offer resource managers additional information on the applicability of the Tennant method for determining instream flow needs for the physical, biological, and social setting. This study concluded that Tennant's original dataset was not representative of streams in the western United States. Data collected from lower gradient streams in Nebraska followed the patterns set forth by Tennant much more closely, and therefore the Tennant method is more applicable in similar low gradient streams (slope less than 1%). In higher gradient streams the use of the Tennant method should be with caution and be restricted to planning stages of instream flow recommendations. Further validation and method adaptation is recommended when using the Tennant method for higher gradient stream types. The Tennant method should be used in instream flow protection scenarios and not in restoration scenarios because of the method's assumption that the current average annual hydrograph represents the optimal fish habitat.
Open Access
Quantifying scale relationships in snow distributions
(Colorado State University. Libraries, 2007) Deems, Jeffrey S., author; Fassnacht, Steven R., advisor; Elder, Kelly J., committee member; Liston, Glen E., committee member; Painter, Thomas H., committee member
Spatial distributions of snow in mountain environments represent the time integration of accumulation and ablation processes, and are strongly and dynamically linked to mountain hydrologic, ecologic, and climatic systems. Accurate measurement and modeling of the spatial distribution and variability of the seasonal mountain snowpack at different scales are imperative for water supply and hydropower decision-making, for investigations of land-atmosphere interaction or biogeochemical cycling, and for accurate simulation of earth system processes and feedbacks. Assessment and prediction of snow distributions in complex terrain are heavily dependent on scale effects, as the pattern and magnitude of variability in snow distributions depends on the scale of observation. Measurement and model scales are usually different from process scales, and thereby introduce a scale bias to the estimate or prediction. To quantify this bias, or to properly design measurement schemes and model applications, the process scale must be known or estimated. Airborne Light Detection And Ranging (lidar) products provide high-resolution, broad-extent altimetry data for terrain and snowpack mapping, and allow an application of variogram fractal analysis techniques to characterize snow depth scaling properties over lag distances from 1 to 1000 meters. Snow depth patterns as measured by lidar at three Colorado mountain sites exhibit fractal (power law) scaling patterns over two distinct scale ranges, separated by a distinct break at the 15-40 m lag distance, depending on the site. Each fractal range represents a range of separation distances over which snow depth processes remain consistent. The scale break between fractal regions is a characteristic scale at which snow depth process relationships change fundamentally. Similar scale break distances in vegetation topography datasets suggest that the snow depth scale break represents a change in wind redistribution processes from wind/vegetation interactions at small lags to wind/terrain interactions at larger lags. These snow depth scale characteristics are interannually consistent, directly describe the scales of action of snow accumulation, redistribution, and ablation processes, and inform scale considerations for measurement and modeling. Snow process models are designed to represent processes acting over specific scale ranges. However, since the incorporated processes vary with scale, the model performance cannot be scale-independent. Thus, distributed snow models must represent the appropriate process interactions at each scale in order to produce reasonable simulations of snow depth or snow water equivalent (SWE) variability. By comparing fractal dimensions and scale break lengths of modeled snow depth patterns to those derived from lidar observations, the model process representations can be evaluated and subsequently refined. Snow depth simulations from the SnowModel seasonal snow process model exhibit fractal patterns, and a scale break can be produced by including a sub-model that simulates fine-scale wind drifting patterns. The fractal dimensions provide important spatial scaling information that can inform refinement of process representations. This collection of work provides a new application of methods developed in other geophysical fields for quantifying scale and variability relationships.
Embargo
Recent and future Colorado water: snow drought, streamflow, and winter recreation
(Colorado State University. Libraries, 2023) Pfohl, Anna K. D., author; Fassnacht, Steven R., advisor; Barnard, David M., committee member; Kampf, Stephanie K., committee member; Rasmussen, Kristen L., committee member
Water in the western United States is a crucial resource for ecosystems, the abiotic environment, and people (for industrial, agricultural, and residential purposes). A majority of this water originates in the seasonal snowpack in the mountains. The snowpack is responsible for maintaining the water supply, and changes to this system have broad and severe implications. Various metrics have been used to quantify these patterns when snow is less than normal, often referred to as a snow drought or a low snow year. In recent decades, the number of years with low snow have increased, and this will continue and intensify into the future. With observed decreases in long-term snow and modeled decreases for the future, high snow years become more critical to support the water supply. Beyond supplying water for downstream use, the seasonal snowpack also sustains the winter recreation industry, which is a large component of many local and state economies. The Weather Research and Forecasting Model (WRF) is a 4-km mesoscale model that can capture orography and convective processes over complex terrain. WRF includes two time periods: the control (CTL) based on historic conditions and the future under pseudo-global warming (PGW) conditions. This dataset was used to drive SnowModel (WRF-SM) to produce 100-m, daily snow water equivalent (SWE), total precipitation, solid precipitation, snowmelt, runoff, and air temperature. Using these datasets, this research examines past and future snow and streamflow in Colorado. We evaluated 1) common metrics and trends for snow drought; 2) used WRF data to drive the Ages hydrologic model to examine changes (snow, streamflow, and flow partitioning) in two high snow years; and 3) ski opportunities at nine different resorts. To evaluate methods of defining snow drought, we used SWE and winter precipitation data from Snow Telemetry stations and the WRF-SM dataset described above. Classifying drought with the ratio of SWE to winter precipitation resulted in drought occurrence for more than 50% of station-years from 1981 to 2020. Using percentiles of long-term peak SWE indicated that occurrence of low or very low years increased from 2001 to 2020 compared with the previous 20 years. Under PGW conditions, elevations between 1800 and 2400 m shifted drought classification towards low or very low, with higher elevations (3200 m and above) remaining relatively unchanged. To examine changes in snow, streamflow, and flow partitioning under a PGW scenario for two high snow years (2008 and 2011), we used Ages, a spatially distributed watershed model, in the Upper Blue River watershed in central Colorado. Changes in snow (snowmelt and solid precipitation) were greatest in magnitude at high elevations. Timing of peak streamflow shifted to nearly two months earlier under a PGW scenario. To examine ski opportunities, we developed metrics to quantify ski conditions. The number of opportunities for snowmaking in the future will decrease throughout the season, but especially in October and November. Ski days (snow depth greater than 50 cm) will decrease in early and late season and increase at lower elevations from January through March. Powder days (fresh depth greater than 15 cm and fresh density greater than 125 kg/m3) follow a similar pattern. Ski resorts at low elevations will generally be more susceptible to changes under a PGW scenario. Additionally, using a fine-resolution dataset allowed investigation of smaller study areas to understand the changes that are not captured with coarser resolutions.
Open Access
Regional patterns of snow water equivalent in the Colorado River Basin using snowpack telemetry (SNOTEL) data
(Colorado State University. Libraries, 2008) Derry, Jeffrey Edward, author; Fassnacht, Steven R., advisor; Doesken, Nolan J., committee member; Stednick, John D., committee member
Identifying regions of homogeneity of precipitation data is often a crucial preliminary step in natural resource investigations. Previous clustering of station based snow water equivalent (SWE) data has typically grouped stations based on spatial proximity, political boundaries, or watershed boundaries, and has been restricted due to the temporal resolution of snow course data. This investigation utilized daily data from 216 snowpack telemetry (SNOTEL) stations located in and around the Colorado River Basin over a 15-year period (1991-2005) to cluster stations, i.e., identify regions of homogeneity, based on the patterns and variability of SWE. To achieve this, data were submitted to a selforganizing map (SOM), a particular application of artificial neural networks. This methodology represents a learning algorithm that is non-linear, non-parametric, unsupervised, and learns through an iterative training process.
Open Access
Snow depth variability in sagebrush drifts in high altitude rangelands, North Park, Colorado
(Colorado State University. Libraries, 2010) Tedesche, Molly Elizabeth, author; Fassnacht, Steven R., advisor; Meiman, Paul J., committee member; Knapp, Alan K., 1956-, committee member
In high altitude rangelands, such as those in Colorado, sagebrush and other shrubs can affect transport and deposition of wind-blown snow, thus enabling the formation of snowdrifts. Sagebrush management techniques could have significant effects on snow accumulation patterns. Snow that potentially could have been trapped by the plants may return to the atmosphere through sublimation. Soil moisture and subsequent plant growth may be affected by this sublimation. Measurement of snow depth and the spatial variability of these measurements might be important information for understanding snowdrift formation processes. Determination of the most effective measurement scale for understanding important ecologic and hydrologic processes in this environment is therefore essential. Directional variogram analyses and Moran's I statistics are two efficient methods for representing the spatial variability of snow depth at different measurement scales in shallow rangeland snow packs. The three following hypotheses are tested to determine the nature of snow depth spatial variability in the high altitude plateau rangeland of North Park, Colorado, using directional variogram analyses and Moran's I statistical methods: (1) Sagebrush plant dimensions (microtopography) are less spatially autocorrelated than the variations in snow depth measurements in resultant snowdrifts around an individual plant; (2) As winter progresses and the voids within sagebrush plants are filled with wind-distributed snow, the resultant surface evolves into a progressively less spatially variable microtopography; (3) When measuring a shallow rangeland snow pack, smaller scale measurements produce a progressively more spatially variable dataset of snow depths and a therefore less spatially autocorrelated snow surface texture. Results of both the variogram and Moran's I analyses indicate that the first hypothesis may be supported. Variogram gamma values and fractal dimensions for the sagebrush canopy microtopography tend to be larger than for the corresponding snow depth measurements. This specifies more spatial variability in the sagebrush surface than in snow depths. The Moran's I values also indicate that there is less spatial autocorrelation within sagebrush plant geometry than there is among snow depth measurements in resultant snowdrifts. The second hypothesis is also supported by the results of both variogram analyses and Moran's I statistics. Variogram analyses indicate that snow depth becomes less spatially variable (with lower sill values) as the winter progresses. There is also evidence of a "leveling-off" of the spatial variability occurring later in the season. Variogram coefficients of variation and fractal dimensions are also very close in value. The Moran's I values also indicate more positive spatial autocorrelation among snow depths throughout the winter season. Results of the variogram analyses for the multiple scale snow depth datasets do not support the third hypothesis. The results actually indicate that smaller scale snow depth measurements produce a more spatially autocorrelated dataset in shallow rangeland snow packs. As scale in snow depth measurements increases, both lag distances and gamma values increase slightly, as well. The Moran's I values are more supportive of the third hypothesis, indicating that mid-range small-scale snow depth measurements may be the least spatially variable.
Open Access
Snow sublimation and seasonal snowpack variability
(Colorado State University. Libraries, 2016) Sexstone, Graham A., author; Fassnacht, Steven R., advisor; Clow, David W., committee member; Hiemstra, Christopher A., committee member; Butters, Gregory L., committee member
In the western United States, seasonal melt from snow in mountainous regions serves as an essential water resource for ecological and anthropological needs, and improving our abilities to quantify the amount of water stored in the seasonal snowpack and provide short-term forecasts of snowmelt inputs into river systems is a critical science endeavor. Two important uncertainties in characterizing the seasonal evolution of snow in mountainous environments are related to the inherent spatial variability of snow in complex terrain and the magnitude and variability of snow sublimation fluxes between snow and the atmosphere; these uncertainties motivate this collection of research which includes three studies conducted in the north-central Colorado Rocky Mountains. The first study uses fine resolution airborne lidar snow depth datasets to evaluate the spatial variability of snow within areas comparable to coarse scale model grids (i.e. subgrid variability at 500 m resolution). Snow depth coefficient of variation, which was used as a metric for evaluating subgrid snow variability, exhibited substantial variability in mountainous terrain and was well correlated with mean snow depth, land cover type, as well as canopy and topography characteristics. Results highlight that simple statistical models for predicting subgrid snow depth coefficient of variation in alpine and subalpine areas can provide useful parameterizations of subgrid snow distributions. Given that snow sublimation fluxes are expected to exert important influences on snow distributions, the second and third studies focus on measuring and modeling the variability and importance of snow sublimation. To evaluate the relative merits and measurement uncertainty of methods for quantifying snow sublimation in mountainous environments, a comparison was made between the eddy covariance, Bowen ratio-energy balance, bulk aerodynamic flux, and aerodynamic profile methods within two forested openings. Biases between methods are evaluated over a range of environmental conditions, which highlight limitations and uncertainties of each method as well as the challenges related to measuring surface sublimation in snow-covered regions. Results provide guidance for future investigations seeking to quantify snow sublimation through station measurements and suggest that the eddy covariance and/or bulk aerodynamic flux methods are superior for estimating surface sublimation in snow-covered forested openings. To evaluate the spatial variability and importance of snow sublimation, a process-based snow model is applied across a 3600 km2 domain over five water years. In-situ eddy covariance observations of snow sublimation compare well with modeled snow sublimation at sites dominated by surface and canopy sublimation, but highlight challenges with model evaluation at sites where blowing sublimation is prominent. Modeled snow sublimation shows considerable spatial variability at the hillslope scale that is evident across elevation gradients and between land cover types. Snow sublimation from forested areas (canopy plus surface sublimation) accounted for the majority of modeled sublimation losses across the study domain and highlights the importance of sublimation from snow stored in the forest canopy in this region. Model simulations suggest that snow sublimation is a significant component of the winter water balance, accounting for losses equivalent to 43 percent of total snowfall, and strongly influences snow distributions in this region. Results from this study have important implications for future water management and decision making.
Open Access
Snowpack depletion modeling using fast all-season soil strength (FASST) and SnowModel in a high-elevation, high relief catchment in the central Rocky Mountains
(Colorado State University. Libraries, 2007) Sawyer, Anne Elizabeth, author; Elder, Kelly, advisor; Fassnacht, Steven R., advisor
In the western United States, snowmelt from mountain basins has historically provided 70-90% of annual runoff and the winter snowpack acts as a reservoir to store water for spring and summer soil moisture and stream recharge. Modeling the timing and magnitude of snowpack depletion and runoff in mountainous basins is an essential tool for forecasting water supply for irrigation, drinking and industrial uses. Modeled point estimates of snow depth depletion at two forested, sub-alpine sites (using Fast All-Season Soil STrength (FASST) and SnowModel) were compared to observed seasonal snow depths from an acoustic snow depth sensor. Meteorological forcing data for each model were collected at both sites between March and June of 2003 and included air temperature, relative humidity, air pressure, wind speed and direction, incoming and outgoing shortwave radiation and upwelling and downwelling longwave radiation. Precipitation was measured using precipitation gauges near each site. SnowModel was also used to simulate distributed snow cover depletion and runoff in a mountain catchment, St. Louis Creek (82.5 km2), at varying spatial resolutions of Hydrologic Response Units (HRUs). HRUs were created based on physiographic characteristics of the basin including elevation, slope, aspect and vegetation cover. The number of HRUs in five simulations ranged from one (basin average) to 3726. Snow covered area (SCA) and basin-average snow water equivalent (SWE) depletion curves were generated for each simulation. Depletion curves were compared to modeled and observed St. Louis Creek discharge. Diversions above the basin outlet necessitated the reconstruction of 2003 St. Louis Creek discharge using statistical relationships between discharge from St. Louis Creek and two smaller gauged streams within the basin using pre-diversion discharge data (1943 – 1955). Both FASST and SnowModel successfully simulated one-dimensional snow depth depletion at both sites when compared to observed snow depth using standard statistical metrics for evaluation. SnowModel produced realistic SCA and SWE depletion curves for St. Louis Creek basin, and the finest spatial resolution simulation best represented the spatial variability within the basin and produced the most realistic results. However, as anticipated, the timing and magnitude of runoff was incorrect due to a lack of a runoff routing module within SnowModel.
Open Access
Spatial accumulation patterns of snow water equivalent in the southern Rocky Mountains
(Colorado State University. Libraries, 2016) Von Thaden, Benjamin C., author; Fassnacht, Steven R., advisor; Stednick, John D., committee member; Butters, Gregory, committee member
Only several point measurements may be taken within a given watershed to estimate snow water equivalent (SWE) due to cost limitations, which necessitates basin-scale estimation of SWE. Modeling often assumes consistency in the spatial distribution of SWE, which may not be correct. Identifying patterns and variability in the spatial distribution of SWE can improve snow hydrology models and result in more accurate modeling. Most previous snow distribution studies focused on small domains, less than 10 km. This study examined SWE distribution at a domain of 757 km. This study used variogram analysis for SWE data from 90 long-term SNOTEL stations to determine if a physical distance exists at which snow accumulation patterns across the southern Rocky Mountains vary abruptly. The concurrent accumulation period from SNOTEL stations were paired one-by-one until all 90 stations were compared among each other for all years on record. This comparison generated a relative accumulation slope (relative to the accumulation slope of all other 89 SNOTEL stations from the period of record) and along with physical distance between station pairs, variograms were computed using the semi-variance of the relative accumulation slopes. A physical divide (a break in high-elevation terrain) exists in the topography of the study region that runs East-West about the parallel 38°45’N. Two subset variograms were computed, one by dividing station pairs by their location relative the parallel 38°45’N into a north zone and a south zone, and the second by the pair’s land cover type, specifically evergreen, non-evergreen, or mixed. From the variogram analyses two physical distances were determined (100 and 340 km) at which snow accumulation patterns in the southern Rocky Mountains vary abruptly. There was more variance in snow accumulation south of the 38°45’N parallel, as the zone north of the 38°45’N parallel experiences storm tracks different from the storm tracks that dominate the zone south of this dividing parallel. Land cover was shown to have little effect on snow accumulation patterns. The amount of variability in individual day SWE was found to be correlated to the magnitude of the average SWE among all SNOTEL stations, such that the greater the average SWE, the larger the variability in SWE across the southern Rock Mountains.
Open Access
The effects of input data degradation on hydrological model performance for a snowmelt dominated watershed
(Colorado State University. Libraries, 2006) McKim, Scott D., author; Fassnacht, Steven R., advisor
The quality and quantity of hydrometeorological data used as input to a hydrologic model is varied and the output compared to observed historical flows. Temperature and precipitation data were used to feed the National Weather Service River Forecast System (NWSRFS); this hydrologic model outputs streamflow and is used daily throughout the country to forecast streamflows. NWSRFS is a lumped empirical model developed in the 1970s for the NWS and is calibrated in this study to model a portion of the snowmelt dominated Yampa River watershed in northwest Colorado. An analysis scheme is followed to capture the model's dependence on representative meteorological stations located in an around the modeled basin. Many regions in the United States experience meteorological and hydrological data scarcity issues. Operationally this becomes important when the available data is insufficient enough to produce reliable model outputs. Similar to Tsintikidis et al. (2002) concluding that the installation of additional rain gauges in a modeled basin would decrease the error of precipitation measurements in the model, we sought to find if increasing data input into a model, both the quantity and quality given by site representivity, will increase the accuracy of our model runs. The study basin was chosen for its snowmelt dominance characteristic. Mean areal precipitation and temperature values for the modeled zones are developed individually in each analysis scheme by the arrangement of stations used in each sensitivity analysis. A statistical analysis of the relative difference between model runs and archived observed values is performed in an effort to illustrate the effect of different model input data arrangements on model simulations. This study aimed at testing the tenable assertion that subtracting hydrometeorological data from a model's dataset would decrease the accuracy of forecasted stream flows from that model. Stream flows and snow water equivalence are analyzed to test the model's sensitivity to the amount of data used. Since the NWSRFS uses predetermined weights to determine MAPs, the number of stations used does not significantly affect model output. The usage of predetermined weights maintains a consistent year-to-year MAP. Varying the MAT station configuration showed a more sizeable effect than the MAP scheme illustrated. Though this procedure could and should be replicated for other hydroclimates and for basins with different sizes, the specific results are not transferable to other basins. The basin modeled is very heavily snowmelt dominated; this quality, as well as it size, climate, topography, and available hydrometeorological stations all influence model results; altering any of these would change the model performance.
Open Access
Trends and tree-rings: an investigation of the historical and paleo proxy hydroclimate record of the Khangai Mountain Region of Mongolia
(Colorado State University. Libraries, 2016) Venable, Niah B. H., author; Fassnacht, Steven R., advisor; Laituri, Melinda J., committee member; Sanford, William E., committee member; Brown, Peter M., committee member
The Khangai Mountain region of western central Mongolia is a diverse area of mountain, forest, steppe, and desert steppe landscapes reaching across and beyond the mountains. The tradition of nomadic pastoralism is strong in the region, with water for domestic and livestock needs supplied through lakes, springs, rivers, and wells. Herders of the region have felt impacts from the climatic extremes of the last few decades in terms of increasing temperatures and decreasing water supplies. The main objective of this dissertation is to quantify the changing climate of Mongolia through analysis of key hydrometeorological variables over space and through time. The assessments of trends in the data and the paleo proxy analyses herein address interdisciplinary research questions using multidisciplinary approaches. In closing, this work also examines how the data and analyses presented are used as objects that cross disciplinary boundaries, and can facilitate communication and collaboration between different groups. To provide context for this work, a countrywide view of changing maximum temperature, minimum temperature, and precipitation are examined using trend analyses of gridded datasets. Both minimum and maximum temperatures are significantly warming across the country. Significant decreases in precipitation are concentrated in the central and eastern parts of the country for the 50-year period of analysis. Local knowledge of hydroclimatic change provides another source of climatic information with herders of the Khangai Mountain region observing temperature increases, though the exact time period over which change has occurred varies depending upon memory. Therefore, temperature data were analyzed from five meteorological stations with varying lengths of record from 15 to 50 years and varying start periods based on the available length of record. The most highly significant changes occurred for the longest time periods and for annual average minimum temperatures. Issues of data availability, serial correlation, and homogeneity of climate records were explored using the Mann-Kendall test for trend significance and the Thiel-Sen method for determining trend slope or magnitude in precipitation and streamflow records. An additional step of prewhitening the data prior to testing was used to reduce the influence of autocorrelation on results. Homogeneity testing was also performed. Decreasing trends in annual, spring, and summer precipitation and/or streamflow were found at several Mongolian stations, particularly on the northern side of the mountains, with increasing winter precipitation trends at one site. Results were compared to analyses using Colorado data. Degradation of the Colorado hydroclimate records by shortening the time series and introducing gaps to simulate inconsistencies found in Mongolian datasets created significant trends where none previously existed. Tree-ring reconstructions of Mongolian hydroclimate variables have provided insight on multidecadal and muticentennial trends in climate variability over many other parts of the country, but that work has not been extended to contextualize the recent sharply decreasing streamflows of the Khangai Mountain region. Cores from two new sites collected in the summer of 2012 and records from eight other moisture-sensitive sites in the region were used to reconstruct streamflow for four gages. Missing streamflow data were filled by multiple imputation/predictive mean matching methods with data from six nearby meteorological stations prior to use in multiple linear regression models developed for the reconstructions. A quantitative evaluation of reconstructed and historical extremes of wet and dry conditions in each basin and qualitative analyses of event synchrony are discussed. The drought events of the last decade and a half, while extreme are not beyond the range of natural variability found over the last 300+ years in the four Khangai Mountain region rivers and could be considered plausible flow conditions for the future, particularly under a warming and possibly drying climate. Finally, this dissertation explores cross-boundary connections within each previous chapter and contributions of this work to selected goals of the Mongolian Rangelands and Resilience (MOR2) project, an interdisciplinary and cross-cultural collaboration investigating the resilience of Mongolian pastoral systems to climate change. Changes to the livelihoods of traditional nomadic pastoralists of Mongolia are not only attributable to climate, but also represent changes to socio-ecological, economic, and governmental/policy systems. The analyses of observational gridded, station-based, and paleo proxy data in this dissertation provide a quantitative foundation for continued investigations of the physical hydroclimate systems of the region and further themes developed in previous research from across Asia and within Mongolia. The results of this work will prove useful as a foundation for the development of water policy and infrastructure ideally favoring sustainable nomadic pastoral use of the region’s finite water resources under a changing climate.
Open Access
Trends in snow water equivalent in Rocky Mountain National Park and the northern Front Range of Colorado, USA
(Colorado State University. Libraries, 2016) Patterson, Glenn G., author; Fassnacht, Steven R., advisor; Laituri, Melinda J., committee member; Sanford, William E., committee member; Pritchett, James, committee member
The seasonal snowpack in Rocky Mountain National Park and the northern Front Range of Colorado, USA, within 50 km of the park, is undergoing changes that will pose challenges for water providers, natural resource managers, and winter recreation enthusiasts. Assessing long-term temporal trends in measures of the seasonal snowpack, and in the climatic factors that influence its annual accumulation and ablation, helps to characterize those challenges. In particular, evaluating the patterns of variation in those trends over different parts of the snow season provides new understanding as to their causes. This also helps to determine specific ramifications of the trends. In addition, placing the current 35-year trends in the longer context of longer-term observational records, and paleoclimate tree-ring reconstructions, provides useful comparisons of current and past trends. Finally, projections of future trends provided by linked climate and hydrologic models offer a sense of how these trends are likely to affect the snowpack of the future. Some factors such as the high elevation of the study area help to preserve conditions favorable to development of the seasonal snowpack, and hence to limit trends toward greater warming-induced melt and less precipitation falling as snow. Nevertheless, traditional snowpack measures such as April 1 snow water equivalent (SWE) show consistent declining trends over the 35-year period of record for automated snow monitoring stations in the study area. The trends are not uniform throughout the snow season, but vary significantly by month. As a result, November and March have warming and drying trends that delay the beginning of the winter snow season and reduce the traditional accumulation that formerly characterized the early spring. In contrast, the core winter months of December, January, and February have cooling and wetting trends that have been enhancing SWE during the heart of the winter. Mid-April to early May is another period during which cooling and wetting trends have been enhancing SWE, although these months also show more variability. This oscillating pattern helps to explain why there has not been a pervasive shift to earlier and lower annual peak SWE in the study area. Paleo SWE reconstructions based on tree-ring chronologies show that at least some of the recent 35-year trends in observed SWE described in this study have comparable precedents during the preceding five centuries, but we do not yet know how long the recent trends will continue. Linked climate and hydrologic models project that the observed trends are likely to continue, and that by 2050 measures such as April 1 SWE in the study area are likely to decrease by 25 percent.
Open Access
Water quality and survivability of Didymosphenia geminata
(Colorado State University. Libraries, 2012) Beeby, Johannes, author; Stednick, John D., advisor; Fassnacht, Steven R., advisor; Clements, William H., committee member
Didymosphenia geminata or Didymo has become a world-wide invasive aquatic species. During blooms, the algae can form thick mats covering entire reaches of stream bottom, which in turn creates negative aesthetic, ecologic, and economic impacts. Although Didymo is historically present in the United States, it is spreading quickly into areas that were previously free of it, and is even growing in waters that were thought not ideal habitat for Didymo. Previous research on how water quality affects Didymo growth and spreading appear to be influenced by streamflow rates and water pH levels. Other water quality parameters have not been fully tested on Didymo, which would contribute to a better understanding of what controls Didymo growth. The first goal of this study was to colonize Didymo in an artificial stream within a laboratory setting. The second goal was to evaluate the survivability of Didymo by exposing it to different water quality parameters. Artificial stream configurations with various light intensity and duration, water temperature and velocity, source water chemistry, and different growth media were used. In all attempts colonization of Didymo was unsuccessful as Didymo slowly deteriorated and became covered by other algae that were more successful in the artificial conditions. Didymo survivability as affected by a 60 minute exposure to different water quality parameters followed previously determined results in that known algaecides did affect cell viability, while other non-toxic parameters showed no effect on Didymo. Nitrate, nitrite, phosphate, chloride, calcium, and magnesium did not affect Didymo survivability. Ammonia also did not affect Didymo but signs of cells lysis were observed and possible mortality may occur with longer exposure times. Copper, zinc, chlorine, and pH affected Didymo survivability. Copper showed the greatest affect on Didymo survivability with the median lethal concentrations (LC50) for copper at 9.3°C and 13.0°C being 3.3 mg/L and 5.4 mg/L respectively at pH 7.7. For copper toxicity in waters with a lower pH (6.7) the resulting LC50 was 33 mg/L. Generally, both colder water temperature and higher pH increased copper toxicity on Didymo. The affect of temperature on copper toxicity was shown to be statistically significant (p-value 0.02). However, there was no statistically significant affect of pH on copper toxicity (p-value 0.07). The LC50 could also not be determined for all three zinc tests but the highest zinc concentration of 40 mg/L had on average 56% of Didymo cells surviving. No apparent trend on the affect of temperature to zinc toxicity on Didymo could be determined; however, the interaction of temperature on zinc toxicity was statistically significant (p-value 0.02). Chlorine at temperatures of 11.5°C and 17.3°C had LC50s of 5.67 and 8.46 mg/L respectively. The affect of temperature on chlorine toxicity was statistically significant (p-value <0.001). Didymo survivability was affected in water with pH 4.3 but not in water with pH 5.9 and 6.9. Cell lysis was occurring in water with pH 10.7 but no sign of any affect on Didymo survivability was found in water with pH 9.9.