Methods for downstream river flow prediction can be categorized into physics-based and empirical approaches. Although based on well-studied physical relationships, physics-based models rely on numerous hydrologic variables characteristic of the specific river system that can be costly to acquire. Moreover, simulation is often computationally intensive. Conversely, empirical models require less information about the system being modeled and can capture a system’s interactions based on a smaller set of observed data. This article introduces two empirical methods to predict downstream hydraulic variables based on observed stream data: a linear programming (LP) model, and a convolutional neural network (CNN). We apply both empirical models within the Colorado River system to a site located on the Green River, downstream of the Yampa River confluence and Flaming Gorge Dam, and compare it to the physics-based model Streamflow Synthesis and Reservoir Regulation (SSARR) currently used by federal agencies. Results show that both proposed models significantly outperform the SSARR model. Moreover, the CNN model outperforms the LP model for hourly predictions whereas both perform similarly for daily predictions. Although less accurate than the CNN model at finer temporal resolution, the LP model is ideal for linear water scheduling tools.

In this article, we present our recent study on computational methodology for predicting the toxicity of PFAS known as “forever chemicals” based on chemical structures through evaluation of multiple machine learning methods. To address the scarcity of PFAS toxicity data, a deep “transfer learning” method has been investigated by leveraging toxicity information over the entire organic chemical domain and an uncertainty-informed workflow by incorporating SelectiveNet architecture, which can support future guidance of high throughput screening with knowledge of chemical structures, has been developed.

Extreme precipitation and flooding events are are becoming more common in many places worldwide necessitating developing a thorough understanding of this phenomena and implementing sustainable management scenarios of these added water resources. One of those settings is the source areas of the Nile River. Twice (1998-2003; 2019-2022) in the last two decades, high precipitation over the Nile River source areas has led to flooding in northern and central Sudan, the filling of Lake Nasser (LN), and the diversion of the overflow to depressions in the plateau bounding the Nile River from the west, where it is lost to evaporation. We adopted an integrated approach using temporal GRACE and GRACE-FO data, precipitation, and continuous rainfall-runoff models (Soil Water Assessment Tool; SWAT) and global circulation model (GCMs; CCSM4, HadGEM3, and GFDL-CM4.0) projections to investigate the primary source area contributing to the observed increased in runoff that reached LN in the past two decades, and assessed the impact of climate change on the LN’s runoff throughout the 21st century. Findings include: (1) the primary contributor to increased downstream runoff reaching LN is the Blue Nile subbasin (BNS), (2) we simulated the BNS runoff in the 21st century using a calibrated (calibration period: 1965-1992) rainfall-runoff model with CCSM4, HadGEM3, and GFDL-CM4.0 projections as model inputs, (3) the extreme value analysis that we conducted for the projected runoff driven by GCMs’ output indicates extreme floods are more severe in the 21st-century, (4) based on the predicted median values of stream flows for the 21st century, the stream flows are expected to increase by 2%, 5%, 9%, and 13%, corresponding to 25-, 50-, 100-, and 200-year events, respectively, under RCP 4.5 and 2%, 7%, 11%, 15% and 20% for 10-, 25-, 50-, 100-, and 200-year events, respectively, under RCP 8.5, and (5) one adaptation for the projected 21st-century increase in precipitation (25%-39%) and flood (2%-20%) extremes is to recharge Egypt’s fossil aquifers during high flood years that are being depleted (2002-2023) by 0.98 km3/yr.

The Grand Ethiopian Renaissance Dam (GERD) has become a key focus for hydrological research due to its influence on the Nile Basin. Previous studies have revealed that seepage losses from the GERD reservoir, facilitated by fault networks in the fractured terrain, are underestimated. This study builds on earlier findings by analyzing seepage from the GERD's fourth filling, completed on September 28, 2023, reaching 39.8 km³. By June 18, 2024, the reservoir's volume decreased to 30.2 km³, with a total seepage of 9.6 km³. The analysis suggests that as the reservoir area expands, more fractures and faults are accessed, increasing seepage. These results highlight the necessity of integrating seepage analysis into hydrological models to support sustainable water management and informed decision-making in the Nile Basin. Failure to address these factors could lead to inaccurate hydrological predictions and impede regional cooperation.