Journal of The American Water Resources Association最新文献

Evaporation is a major water-loss process that significantly disrupts the hydrological cycle; therefore, reliable and continuous evaporation monitoring is essential for decision-makers in water resource management. However, hyper-arid climates exhibit accelerated evaporation rates, complicating hydrological modeling. This study represents the first attempt to integrate the RReliefF algorithm for meteorological feature selection with machine learning models for pan evaporation prediction in hyper-arid climates. This approach overcomes the arbitrary selection of features for ML model input. Daily average pan evaporation rates at the examined stations exceed 8 mm/day. Such extremely high evaporative losses have been shown to hinder ML model performance. Extreme gradient boosting (XGBoost), random forest model, and k-nearest neighbors were used. Meteorological datasets were preprocessed using the RReliefF algorithm to rank their influence on pan evaporation variability. Depending on the weather station, shortwave radiation, wind speed, and average diurnal temperature emerged as the best predictors of pan evaporation rates. During the validation period, the Nash–Sutcliffe efficiency coefficient (NS), root mean squared error (RMSE), and mean absolute error (MAE) were 0.85–0.94, 1.152–1.833, and 0.863–1.147 mm/day, respectively. The findings of this study offer a robust and efficient computational approach for forecasting evaporative losses in hyper-arid environments.

Resolving groundwater overuse is an ongoing challenge that will require irrigation to cease on some land, leading to questions about what to do with land no longer irrigated. At the same time, the world is undergoing a green-energy transition, with new renewable energy infrastructure needed to meet renewable energy targets. Transitioning previously irrigated land to solar energy production with agriculture (i.e., agrivoltaics) can provide simultaneous benefits of reducing water use while increasing renewable energy generation on already disturbed land. We share a study of the viability, benefits, and tradeoffs of transitioning previously irrigated land to agrivoltaics in Diamond Valley, Nevada, where a mandated groundwater management plan will inevitably result in land coming out of irrigated production. Nevada is committed to 50% of electricity sold being from renewable energy sources by 2030 and 100% by 2050, so there is strong interest in developing renewable energy infrastructure, including on previously disturbed lands such as those used for agriculture near transmission lines. We found that while soils and sunlight in arid places like Diamond Valley are suitable for incorporating agrivoltaics with little to no irrigation, transmission capacity is a limiting factor. Additional studies are needed to determine system upgrades required for solar, and effective solar leases are important to protect landowners from financial risks.

This study examines the effect of alfalfa (Medicago sativa L.) on nitrogen (N) and phosphorus (P) loads in subsurface (tile) drainage across storm events using edge-of-field monitoring data from two paired-field sites (A and B) with a before-after-control-impact (BACI) experimental design, located in the northwest region of Ohio, United States. A k-medians cluster analysis was used to classify 462 storm events at Site A and 684 storm events at Site B based on precipitation amount and antecedent moisture conditions (AMC), defined as the cumulative 7-day precipitation prior to a storm event. Patterns of nutrient loss in tile drainage were compared between fields with alfalfa and fields with cash and cover crops using a difference-in-differences analysis across three identified storm event types: Dry storm events, wet storm events, and large storm events. Compared to the cash and cover crop rotations, alfalfa had the following effects on discharge and water quality: little to no reduction in subsurface discharge across all storm events at both sites; significant reduction of subsurface nitrate and total N loads across all storm event types at Site A (~200%–800% lower), but not at Site B; ~45% reduction of subsurface dissolved reactive P during large events at both sites; and 11% and 110% reduction of total P loads during large events at Sites A and B, respectively. The impact of alfalfa during large storm events is important given that most nutrient export occurs during these events.

The Las Vegas Wash (the Wash) provides a mechanism for delivering recycled indoor water used in the Las Vegas Valley, NV to the most downstream basin of Lake Mead. The Wash introduces different water quality to Lake Mead, including higher nitrogen and phosphorus concentrations, and it may contain constituents common to urban runoff including microbial organisms and trace contaminants. A strong link has been established between the residential population of the Las Vegas Valley and the mean annual flowrate in the Wash; however, with the onset of the SARS-CoV-2 pandemic in 2020, significantly reduced tourism led to reduced flow in the Wash. This work expands previous modeling efforts which project Wash flowrates based on population projections by incorporating tourism numbers, using the individuals employed in hospitality as a surrogate. The resulting model suggests that mean yearly Wash flowrates could increase 20.6%–23.2% (between 1.95 and 2.19 $��m^3�/�s�)�$ by 2060, compared to 2022 levels. Numerical simulations of Lake Mead show that these increased Wash flowrates are not expected to have a significant thermal influence on either the drinking water intake (Intake) or Hoover Dam outflows. The Wash delivers about $��{10}^3�$ MW of heat into Lake Mead, while heat transfer at the Intake and Hoover Dam water columns was two orders of magnitude less. Wash water concentration and salinity increased in the simulations at the Intake and Hoover Dam outflows, respectively, by at most 0.91% and 1.3%.

The majority of the discharge in the Colorado River originates as snowmelt in headwater catchments. Flow from these streams exhibits significant year-to-year variability (Coefficient of Variation: 0.2–1.18), challenging the management of a critical water and energy resource for millions. Understanding the factors driving this variability will become even more important in a rapidly changing climate. To address this need, we examine regional patterns in winter baseflow, a metric for catchment groundwater storage, in 52 USGS-gauged watersheds in the Upper Colorado River Basin. Coherent 12- to 14-year patterns emerge in baseflow in 5 headwater regions, alongside an apparent 8-year periodicity in 2 regions, suggesting groundwater recharge is responding to a regional climate forcing. Subsequent analyses identified a statistically significant, positive relationship between antecedent winter baseflow and runoff efficiency (RE) in 22 of 24 headwaters with concurrent precipitation and streamflow data available since 1980. These relationships indicate that winter baseflow can be used to reduce uncertainty in RE and streamflow predictions months before snowmelt begins.

This study aims to characterize single-family household water consumption utilizing smart-metered subdaily water use data from more than 700 single-family households across the state of Arizona in the United States for the water year 2022. Using statistical evidence, we identify factors that drive household water consumption such as the number of occupants, appliance efficiency, and the presence of a swimming pool. Furthermore, climate and other regional drivers of water use are investigated. The analysis encompasses mixed-effects regression models to assess water use patterns on daily, weekly, monthly, and seasonal time-steps. The findings show that approximately 64% of water consumption in Arizona was used for outdoor purposes. Households with a swimming pool use approximately 56% more water overall than those without a pool. Even indoor water use is nearly 26% greater in households with a swimming pool. Deep analysis of smart-metered water use data offers greater insights into the efficiency levels of appliances in a household. Households with high-efficiency appliances use about 18.5% less water than households without high-efficiency appliances. Analysis indicates that log-linear mixed-effects regression models provide the most robust assessments for relating water consumption with household and regional factors. This study helps water managers identify and implement water conservation and demand reduction strategies in single-family neighborhoods.