Prediction Of Snow Water Equivalent Research Articles

Using Temporal Deep Learning Models to Estimate Daily Snow Water Equivalent Over the Rocky Mountains

AbstractIn this study we construct and compare three different deep learning (DL) models for estimating daily snow water equivalent (SWE) from high‐resolution gridded meteorological fields over the Rocky Mountain region. To train the DL models, Snow Telemetry (SNOTEL) station‐based SWE observations are used as the prediction target. All DL models produce higher median Nash‐Sutcliffe Efficiency (NSE) values than a conceptual SWE model and interpolated gridded data sets, although mean squared errors also tend to be higher. Sensitivity of the SWE prediction to the model's input variables is analyzed using an explainable artificial intelligence (XAI) method, yielding insight into the physical relationships learned by the models. This method reveals the dominant role precipitation and temperature play in snowpack dynamics. In applying our models to estimate SWE throughout the Rocky Mountains, an extrapolation problem arises since the statistical properties of SWE (e.g., annual maximum) and geographical properties of individual grid points (e.g., elevation) differ from the training data. This problem is solved by normalizing the SWE with its historical maximum value to alleviate extrapolation for all tested DL models. Our work shows that the DL models are promising tools for estimating SWE, and sufficiently capture relevant physical relationships to make them useful for spatial and temporal extrapolation of SWE values.

Attention-Based Models for Snow-Water Equivalent Prediction

Snow Water-Equivalent (SWE)—the amount of water available if snowpack is melted—is a key decision variable used by water management agencies to make irrigation, flood control, power generation, and drought management decisions. SWE values vary spatiotemporally—affected by weather, topography, and other environmental factors. While daily SWE can be measured by Snow Telemetry (SNOTEL) stations with requisite instrumentation, such stations are spatially sparse requiring interpolation techniques to create spatiotemporal complete data. While recent efforts have explored machine learning (ML) for SWE prediction, a number of recent ML advances have yet to be considered. The main contribution of this paper is to explore one such ML advance, attention mechanisms, for SWE prediction. Our hypothesis is that attention has a unique ability to capture and exploit correlations that may exist across locations or the temporal spectrum (or both). We present a generic attention-based modeling framework for SWE prediction and adapt it to capture spatial attention and temporal attention. Our experimental results on 323 SNOTEL stations in the Western U.S. demonstrate that our attention-based models outperform other machine-learning approaches. We also provide key results highlighting the differences between spatial and temporal attention in this context and a roadmap toward deployment for generating spatially-complete SWE maps.

LSTM-Based Data Integration to Improve Snow Water Equivalent Prediction and Diagnose Error Sources

Abstract Accurate prediction of snow water equivalent (SWE) can be valuable for water resource managers. Recently, deep learning methods such as long short-term memory (LSTM) have exhibited high accuracy in simulating hydrologic variables and can integrate lagged observations to improve prediction, but their benefits were not clear for SWE simulations. Here we tested an LSTM network with data integration (DI) for SWE in the western United States to integrate 30-day-lagged or 7-day-lagged observations of either SWE or satellite-observed snow cover fraction (SCF) to improve future predictions. SCF proved beneficial only for shallow-snow sites during snowmelt, while lagged SWE integration significantly improved prediction accuracy for both shallow- and deep-snow sites. The median Nash–Sutcliffe model efficiency coefficient (NSE) in temporal testing improved from 0.92 to 0.97 with 30-day-lagged SWE integration, and root-mean-square error (RMSE) and the difference between estimated and observed peak SWE values dmax were reduced by 41% and 57%, respectively. DI effectively mitigated accumulated model and forcing errors that would otherwise be persistent. Moreover, by applying DI to different observations (30-day-lagged, 7-day-lagged), we revealed the spatial distribution of errors with different persistent lengths. For example, integrating 30-day-lagged SWE was ineffective for ephemeral snow sites in the southwestern United States, but significantly reduced monthly-scale biases for regions with stable seasonal snowpack such as high-elevation sites in California. These biases are likely attributable to large interannual variability in snowfall or site-specific snow redistribution patterns that can accumulate to impactful levels over time for nonephemeral sites. These results set up benchmark levels and provide guidance for future model improvement strategies.

Snow water equivalent prediction in a mountainous area using hybrid bagging machine learning approaches

Snow water equivalent prediction in a mountainous area using hybrid bagging machine learning approaches

Spatial variability of snow density and its estimation in different periods of snow season in the middle Tianshan Mountains, China

AbstractSnow density is an essential property of snowpack. To obtain the spatial variability of snow density and estimate it in different periods of the snow season remain challenging, particularly in the mountainous area. This study analysed the spatial variability of snow density with in‐situ measurements in three different periods (i.e., accumulation, stable and melt periods) of the snow seasons of 2017/2018 and 2018/2019 in the middle Tianshan Mountains, China. The simulation performances of the multiple linear regression (MLR) model and three machine learning (random forest [RF], extreme gradient boosting [XGB] and light gradient boosting machine [LGBM]) models were evaluated. Results showed that snow density in the melt period (0.27 g cm−3) was generally greater than that in the stable (0.20 g cm−3) and accumulation periods (0.18 g cm−3), and the spatial variability of snow density in the melt period was slightly smaller compared to that in other two periods. The snow density in the mountainous areas was generally higher than that in the plain or oasis areas. It increased significantly (p < 0.05) with elevation during the accumulation and stable periods. In addition to elevation, latitude and ground surface temperature also had critically impacted the spatial variability of snow density in the study area. In the current study, the machine learning models, especially RF, performed better than MLR for simulating snow density in the three periods. Based on the key environmental variables identified by the machine learning model and correlation analysis, this study also provides practical MLR equations to estimate the spatial variance of snow density during different snow periods in the middle Tianshan Mountains. This method can be used for regional snow mass and snow water equivalent prediction, leading to a better understanding of local snow resources.

Automated Cloud Based Long Short-Term Memory Neural Network Based SWE Prediction

Snow derived water is a critical component of a large part of the US water supply. Measurements of the Snow Water Equivalent (SWE) and associated predictions of peak SWE and snowmelt onset are essential inputs for water management efforts. This paper aims to develop a data-driven approach for estimating and predicting SWE dynamics using the Long Short-Term Memory neural network (LSTM) method. Our approach uses historical datasets (precipitation, air temperature, SWE and snow thickness) collected at NRCS Snow Telemetry (SNOTEL) stations to train the LSTM network and current year data to predict SWE behavior. The performance of our prediction was compared for different prediction dates and prediction training datasets. Our results suggest that the proposed LSTM network can be an efficient tool for forecasting the SWE timeseries, as well as Peak SWE and snowmelt timing. Results showed that the window size impacts the model performance (where the Nash Sutcliffe efficiency (NSE) ranged from 0.96 to 0.85 and the Rooted Mean Square Error (RMSE) ranged from 0.038 to 0.07) with an optimum number that should be calibrated for different stations and climate conditions. By implementing the LSTM prediction capability in a cloud based site-monitoring platform we automate model-data integration. By making the data accessible through a graphical web interface and an underlying API which exposes both training and prediction capabilities. the associated results can be made easily accessible to a broad range of stakeholders.

Enhancing Soil and Water Assessment Tool Snow Prediction Reliability with Remote-Sensing-Based Snow Water Equivalent Reconstruction Product for Upland Watersheds in a Multi-Objective Calibration Process

Precipitation occurs in two basic forms defined as liquid state and solid state. Different from rain-fed watershed, modeling snow processes is of vital importance in snow-dominated watersheds. The seasonal snowpack is a natural water reservoir, which stores snow water in winter and releases it in spring and summer. The warmer climate in recent decades has led to earlier snowmelt, a decline in snowpack, and change in the seasonality of river flows. The Soil and Water Assessment Tool (SWAT) could be applied in the snow-influenced watershed because of its ability to simultaneously predict the streamflow generated from rainfall and from the melting of snow. The choice of parameters, reference data, and calibration strategy could significantly affect the SWAT model calibration outcome and further affect the prediction accuracy. In this study, SWAT models are implemented in four upland watersheds in the Tulare Lake Basin (TLB) located across the Southern Sierra Nevada Mountains. Three calibration scenarios considering different calibration parameters and reference datasets are applied to investigate the impact of the Parallel Energy Balance Model (ParBal) snow reconstruction data and snow parameters on the streamflow and snow water-equivalent (SWE) prediction accuracy. In addition, the watershed parameters and lapse rate parameters-led equifinality is also evaluated. The results indicate that calibration of the SWAT model with respect to both streamflow and SWE reference data could improve the model SWE prediction reliability in general. Comparatively, the streamflow predictions are not significantly affected by differently lumped calibration schemes. The default snow parameter values capture the extreme high flows better than the other two calibration scenarios, whereas there is no remarkable difference among the three calibration schemes for capturing the extreme low flows. The watershed and lapse rate parameters-induced equifinality affects the flow prediction more (Nash-Sutcliffe Efficiency (NSE) varies between 0.2–0.3) than the SWE prediction (NSE varies less than 0.1). This study points out the remote-sensing-based SWE reconstruction product as a promising alternative choice for model calibration in ungauged snow-influenced watersheds. The streamflow-reconstructed SWE bi-objective calibrated model could improve the prediction reliability of surface water supply change for the downstream agricultural region under the changing climate.

Prediction of snow water equivalent using artificial neural network and adaptive neuro-fuzzy inference system with two sampling schemes in semi-arid region of Iran

Direct measurement of snow water equivalent (SWE) in snow-dominated mountainous areas is difficult, thus its prediction is essential for water resources management in such areas. In addition, because of nonlinear trend of snow spatial distribution and the multiple influencing factors concerning the SWE spatial distribution, statistical models are not usually able to present acceptable results. Therefore, applicable methods that are able to predict nonlinear trends are necessary. In this research, to predict SWE, the Sohrevard Watershed located in northwest of Iran was selected as the case study. Database was collected, and the required maps were derived. Snow depth (SD) at 150 points with two sampling patterns including systematic random sampling and Latin hypercube sampling (LHS), and snow density at 18 points were randomly measured, and then SWE was calculated. SWE was predicted using artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS) and regression methods. The results showed that the performance of ANN and ANFIS models with two sampling patterns were observed better than the regression method. Moreover, based on most of the efficiency criteria, the efficiency of ANN, ANFIS and regression methods under LHS pattern were observed higher than the systematic random sampling pattern. However, there were no significant differences between the two methods of ANN and ANFIS in SWE prediction. Data of both two sampling patterns had the highest sensitivity to the elevation. In addition, the LHS and the systematic random sampling patterns had the least sensitivity to the profile curvature and plan curvature, respectively.

A Finite Volume Blowing Snow Model for Use With Variable Resolution Meshes

AbstractBlowing snow is ubiquitous in cold, windswept environments. In some regions, blowing snow sublimation losses can ablate a notable fraction of the seasonal snowfall. It is advantageous to predict alpine snow regimes at the spatial scale of snowdrifts (≈1 to 100 m) because of the role of snow redistribution in governing the duration and volume of snowmelt. However, blowing snow processes are often neglected due to computational costs. Here, a three‐dimensional blowing snow model is presented that is spatially discretized using a variable resolution unstructured mesh. This represents the heterogeneity of the surface explicitly yet, for the case study reported, gained a 62% reduction in computational elements versus a fixed‐resolution mesh and resulted in a 44% reduction in total runtime. The model was evaluated for a subarctic mountain basin using transects of measured snow water equivalent (SWE) in a tundra valley. Including blowing snow processes improved the prediction of SWE by capturing inner‐annual snowdrift formation, more than halved the total mean bias error, and increased the coefficient of variation of SWE from 0.04 to 0.31 better matching the observed CV (0.41). The use of a variable resolution mesh did not dramatically degrade the model performance. Comparison with a constant resolution mesh showed a similar CV and RMSE as the variable resolution mesh. The constant resolution mesh had a smaller mean bias error. A sensitivity analysis showed that snowdrift locations and immediate up‐wind sources of blowing snow are the most sensitive areas of the landscape to wind speed variations.

Snow water equivalent prediction using Bayesian data assimilation methods

Using the U.S. National Weather Service’s SNOW-17 model, this study compares common sequential data assimilation methods, the ensemble Kalman filter (EnKF), the ensemble square root filter (EnSRF), and four variants of the particle filter (PF), to predict seasonal snow water equivalent (SWE) within a small watershed near Lake Tahoe, California. In addition to SWE estimation, the various data assimilation methods are used to estimate five of the most sensitive parameters of SNOW-17 by allowing them to evolve with the dynamical system. Unlike Kalman filters, particle filters do not require Gaussian assumptions for the posterior distribution of the state variables. However, the likelihood function used to scale particle weights is often assumed to be Gaussian. This study evaluates the use of an empirical cumulative distribution function (ECDF) based on the Kaplan–Meier survival probability method to compute particle weights. These weights are then used in different particle filter resampling schemes. Detailed analyses are conducted for synthetic and real data assimilation and an assessment of the procedures is made. The results suggest that the particle filter, especially the empirical likelihood variant, is superior to the ensemble Kalman filter based methods for predicting model states, as well as model parameters.

Mapping snow water equivalent by combining a spatially distributed snow hydrology model with passive microwave remote-sensing data

An algorithm to incorporate passive microwave remote-sensing measurements within a spatially distributed snow hydrology model to provide estimates of the spatial distribution of snow water equivalent (SWE) as a function of time is implemented. A priori information provided by the snow hydrology model is used to provide initial estimates of snow parameters from brightness temperature measurements. The algorithm is illustrated by applying it to a mountainous region. The passive microwave remote-sensing measurements are 25-km resolution (grid). However, in mountain regions, the spatial variability of SWE over a 25-km grid is large due to topographic influences. On the other hand, the snow hydrology model has built-in topographic information and the capability to estimate SWE at a 1-km resolution. In their work, the snow hydrology SWE estimates are updated and corrected using Special Sensor Microwave/Imager (SSM/I) passive microwave remote-sensing measurements. The method is applied to the Upper Rio Grande River Basin in the mountains of Colorado. The change in prediction of SWE from hydrology modeling with and without updating is compared with measurements from two SNOwpack TELemetry (SNOTEL) sites in and near the basin. The results indicate that the method incorporating the remote-sensing measurements into the hydrology model is able to more closely estimate the temporal evolution of the measured values of SWE as a function of time.