Changes in Snow Water Storage and Hydrologic Partitioning Across Western North America
Bibliographic record
Abstract
Seasonal snowpack is an essential component in the Earth’s hydrological cycle. About one-sixth of the global population relies on seasonal snowpack and glacier-derived runoff as a primary water resource. Snowmelt contributes to regional water supply, partially dictating the timing and volume of downstream water resources. Mountain snowpacks act as a natural ‘water tower,’ storing winter precipitation until spring and summer months when downstream water demand is greatest. The magnitude and duration of regional snow water storage at the Earth’s surface is thus a function of precipitation phase (as rainfall or snowfall) and the subsequent timing of water release, is unevenly distributed across regions, and is highly sensitive to climate changes. In mountainous western North America, hydrologic partitioning of catchment water inputs is likely sensitive to snow water storage, greatly influencing the volume and timing of downstream water resources. While previous works have studied the distribution of snow water equivalent (SWE) and trends in SWE, previous works have not evaluated the magnitude and duration of snow water storage. As a result, our understanding of how future changes in snowpacks will impact land surface hydrology is poorly understood. Hence, by evaluating trends in the magnitude and duration of snow water storage, and its impact on land surface hydrology, this dissertation adds substantively to the current literature. In this dissertation, I developed a snow water storage metric with a focus on surface water (i.e., above the soil layer), investigated historical and future changes in snow water storage, and related this metric to hydrologic partitioning, or the allocation of water inputs to streamflow (or evapotranspiration) across multiple spatial scales. After an overall introduction of the work (Chapter One), the second chapter of this dissertation is an overview of a newly developed snow water storage metric, which quantifies the differences in volume and timing between precipitation and surface water inputs (SWI, the daily summation of rainfall and snowmelt). Using precipitation forcings and modeled SWE outputs from the Variable Infiltration Capacity (VIC) model, I produced a Snow Storage Index (SSI) to quantify snow water storage volume and duration across western North America. I found that the average annual SSI has decreased (<em>p</em> < 0.01) from 1950-2013. By evaluating precipitation and SWI trends, I showed that the decrease in SSI was a result of significantly earlier SWI in spring months and comparable decreases in SWI later in the year. In mountainous regions where the SSI is declining, which includes > 25% of the western North America study domain, snowmelt and rainfall have begun occurring earlier in the year, reducing the duration and magnitude of snow water storage. This is particularly evident in the Cascades and Southern Rockies. Additional declines in winter precipitation have reduced snow water storage in the Canadian and Northern Rockies. The sensitivity of the SSI depends on annual and seasonal temperature and precipitation variability and varies across different regional mountain ranges. As opposed to trends in SWE or snow fraction, the SSI represents the degree to which snow is delaying the timing (and magnitude) of SWI relative to precipitation. This lag between precipitation inputs and water availability is a fundamental component of the hydrologic cycle in snow-affected regions, offering a more hydrologically relevant perspective (than SWE trends, for example) on changes in water delivery and related climatic sensitivities for hydrologic and ecologic cycles and water resource management. In Chapter Three of this work, I related the SSI to hydrologic partitioning across the United States mountainous west. I discovered that the relationship between SSI and partitioning of incoming precipitation to streamflow is strongly and positively correlated within many ecoregions in the study domain. The ecoregions showing the strongest, positive correlations included: Cascades (r<sup>2</sup> = 0.62), North Cascades (r<sup>2</sup> = 0.61), Blue Mountains (r<sup>2</sup> = 0.56), Canadian Rockies (r<sup>2</sup> = 0.55), Idaho Batholith (r<sup>2</sup> = 0.48), and Columbia Mountains / Northern Rockies (r<sup>2</sup> = 0.45). The ratio of weekly SWI to weekly precipitation (SWI:P) was an equally strong predictor for hydrologic partitioning, particularly in mid-spring (e.g., March / April) and early summer (e.g., June / July) in the same mountainous ecoregions. When less water enters the soil system in spring months, and more in summer months, indicating a longer duration of water storage in the snowpack, more annual water inputs are partitioned to streamflow (maximum r<sup>2</sup> across the same ecoregions = 0.62-0.74). Secondarily, when clustering ecoregions by climate and energy- vs. water-limitations, there was a strong and positive correlation between the SSI and hydrologic partitioning to streamflow in regions with greater energy-limitations, in both maritime (r<sup>2</sup> = 0.57) and inter-mountain / continental (r<sup>2</sup> = 0.42) climates. Relatively water-limited ecoregions, such as the Sierra Nevada, Middle Rockies, Wasatch / Uinta Mountains, and Southern Rockies, showed less sensitivity of hydrologic partitioning to the SSI, potentially due to relatively high aridity. As snow water storage decreases with warming, the timing of water delivery will change to varying degrees across the western United States, with large implications for hydrological and ecological processes and for water resource management across Earth’s snow-influenced regions. In Chapter Four of this work, I used similar methodology to represent historical (control) and future (warming) snow water storage and hydrologic partitioning behavior and relationships at a smaller, alpine watershed in the Front Range of Colorado. Using the Distributed Hydrology Soil Vegetation Model and Weather Research and Forecasting Model-based projections of future climatic conditions, I generated a control and end-of-century warming simulation to compare snow water storage in the past and the future. Similar to the larger scale analyses in Chapters Two and Three, I found that areas where SSI was high experienced a decrease in snow water storage magnitude and duration in the warming (future) simulation, compared to the control (historical) simulation, due to increased rainfall and earlier snowmelt. Within both simulations, areas annually storing water as snow in larger volumes and for longer durations (i.e., greater SSI) partitioned more water to streamflow compared to areas of lower snow water storage (i.e., lower SSI), particularly within bare ground (r<sup>2</sup> = 0.82 (control), 0.76 (warming)), alpine meadow (r<sup>2</sup> = 0.71, 0.79) and closed shrub (r<sup>2</sup> = 0.80, 0.72) vegetation types.On average across the catchment, the warming simulation showed decreased snow water storage (SSI: -0.11) from the control simulation (SSI: -0.07), resulting in a -57% change in SSI. Spatially, SSI percent change across the catchment ranged from -100% to +27%, with increases occurring in wind-scoured areas of the catchment where summer-dominant precipitation seasonality became more uniform. As such, using the Budyko framework, there was an average -10.2% change in the expected amount of precipitation that was partitioned to streamflow under warming conditions. Decreases in partitioning to streamflow with warming suggests that, particularly in cold, alpine regions, future streamflow losses may stress ecological, biological, and sociological dependents downstream, even at small, sub-catchment scales.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".