Snowpack levels across a wide swath of western US states are among the lowest seen in decades, even as regulators struggle to negotiate water rights in the region. States across the western US are facing record low snowpack levels in the middle of the winter season. The snowpack crisis, which could mean a drier, more wildfire -prone summer, is coming as states are racing unsuccessfully against a deadline to agree on terms to share water in the Colorado River Basin, the source of water for 40 million people across seven states in the West. "Barring a genuinely miraculous turnaround" in the remainder of the winter, says Daniel Swain, a climate scientist at the University of California Agriculture and Natural Resources, the low snowpack "has the potential to worsen both the ecological and political crisis on the Colorado Basin, and then also produce really adverse wildfire conditions in some parts of the West." Data provided by the US Department of Agriculture show that as of February 12, snowpack was at less than half its normal level in areas across nine Western states--some of the lowest levels seen in decades.

This paper presents a physics-constrained neural differential equation framework for parameterization, and employs it to model the time evolution of seasonal snow depth given hydrometeorological forcings. When trained on data from multiple SNOTEL sites, the parameterization predicts daily snow depth with under 9% median error and Nash Sutcliffe Efficiencies over 0.94 across a wide variety of snow climates. The parameterization also generalizes to new sites not seen during training, which is not often true for calibrated snow models. Requiring the parameterization to predict snow water equivalent in addition to snow depth only increases error to ~12%. The structure of the approach guarantees the satisfaction of physical constraints, enables these constraints during model training, and allows modeling at different temporal resolutions without additional retraining of the parameterization. These benefits hold potential in climate modeling, and could extend to other dynamical systems with physical constraints.

Streamflow, vital for water resource management, is governed by complex hydrological systems involving intermediate processes driven by meteorological forces. While deep learning models have achieved state-of-the-art results of streamflow prediction, their end-to-end single-task learning approach often fails to capture the causal relationships within these systems. To address this, we propose Hierarchical Conditional Multi-Task Learning (HCMTL), a hierarchical approach that jointly models soil water and snowpack processes based on their causal connections to streamflow. HCMTL utilizes task embeddings to connect network modules, enhancing flexibility and expressiveness while capturing unobserved processes beyond soil water and snowpack. It also incorporates the Conditional Mini-Batch strategy to improve long time series modeling. We compare HCMTL with five baselines on a global dataset. HCMTL's superior performance across hundreds of drainage basins over extended periods shows that integrating domain-specific causal knowledge into deep learning enhances both prediction accuracy and interpretability. This is essential for advancing our understanding of complex hydrological systems and supporting efficient water resource management to mitigate natural disasters like droughts and floods.

Accurately estimating the snowpack in key mountainous basins is critical for water resource managers to make decisions that impact local and global economies, wildlife, and public policy. Currently, this estimation requires multiple LiDAR-equipped plane flights or in situ measurements, both of which are expensive, sparse, and biased towards accessible regions. In this paper, we demonstrate that fusing spatial and temporal information from multiple, openly-available satellite and weather data sources enables estimation of snowpack in key mountainous regions. Our multisource model outperforms single-source estimation by 5.0 inches RMSE, as well as outperforms sparse in situ measurements by 1.2 inches RMSE.

Ants, as a group, are creatures of habit. While an individual's path isn't certain, biologists who have spent a lot of time watching the behavior of entire colonies can predict the average time any one ant might wander around underground before resurfacing. That got NASA physicist Yongxiang Hu wondering if the same predictability might be true of photons--particles of light--traveling through the snowpack. If so, that would let scientists use a laser pulsed from an orbiting satellite to estimate snow depth--potentially a powerful new way to monitor water supplies and the health of sea ice in the Arctic. NASA's ICESat-2 satellite is equipped with lidar, the same variety of laser system that self-driving cars use to build 3D maps of their surroundings.

Artificial Neural Network (ANN) based model is a computational approach commonly used for modeling the complex relationships between input and output parameters. Prediction of the flow rate of a river is a requisite for any successful water resource management and river basin planning. In the current survey, the effectiveness of an Artificial Neural Network was examined to predict the Colorado River discharge. In this modeling process, an ANN model was used to relate the discharge of the Colorado River to such parameters as the amount of precipitation, ambient temperature and snowpack level at a specific time of the year. The model was able to precisely study the impact of climatic parameters on the flow rate of the Colorado River. Keywords: Artificial Neural Network, Discharge, Colorado River, River basin planning 1. Introduction The volumetric flow rate of a river, also called its discharge, at a particular point, is the volume of water passing through the cross section of the river at that point in a unit of time. As aforementioned, forecasting the flow rate of a river could be very useful in water resources management. Any seasonal river basin planning for designation of water between different consumers can not succeed without knowing/predicting the amount of water (i.e.