More Like this

1. Temperature, Precipitation, and Agro-Hydro-Meteorological Indicator Based Scenarios for Decision Making in Ogallala Aquifer Region

   https://doi.org/10.3390/w15030600  

   Anandhi, Aavudai; Deepa, Raveendranpillai; Bhardwaj, Amit; Misra, Vasubandhu (February 2023, Water)

   The Ogallala Aquifer is one of the most productive agricultural regions and is referred to as the “breadbasket of the world”. It covers approximately 225,000 square miles beneath the Great Plains region spanning the states of Texas, New Mexico, Oklahoma, Kansas, Nebraska, South Dakota, Wyoming, and Colorado. The aquifer is a major water source for the region, with its use exceeding recharge. Previous studies have documented climate changes and their impacts in the region. However, this is the first study to document temperature and precipitation changes over the entire Ogallala region from 35 General Circulation Models participating in Phase 5 of the Climate Model Intercomparison Project (CMIP5). The main study objectives were (1) to provide estimates of present and future climate change scenarios for the High Plains Aquifer, (2) to translate the temperature and precipitation changes to agro-ecosystem indicator changes for Kansas using scenario funnels, and (3) to make recommendations for water resource and ecosystem managers to enable effective planning for the future availability of ecosystem services. The temperature change ranged from −4 °C to 8 °C, while the precipitation changes were between −50% to +50% over the region. This study improves the understanding of climate change on water resources and agro-ecosystems. This knowledge can be used to evaluate similar resources where the replenishment rate is slow. 

   more » « less
2. Applying Deep Learning to the Cache Replacement Problem

   Zhan Shi, Xiangru Huang (October 2019, International Symposium on Microrchitecture)

   Despite its success in many areas, deep learning is a poor fit for use in hardware predictors because these models are impractically large and slow, but this paper shows how we can use deep learning to help design a new cache replacement policy. We first show that for cache replacement, a powerful LSTM learning model can in an offline setting provide better accuracy than current hardware predictors. We then perform analysis to interpret this LSTM model, deriving a key insight that allows us to design a simple online model that matches the offline model's accuracy with orders of magnitude lower cost. The result is the Glider cache replacement policy, which we evaluate on a set of 33 memory-intensive programs from the SPEC 2006, SPEC 2017, and GAP (graph-processing) benchmark suites. In a single-core setting, Glider outperforms top finishers from the 2nd Cache Replacement Championship, reducing the miss rate over LRU by 8.9%, compared to reductions of 7.1% for Hawkeye, 6.5% for MPPPB, and 7.5% for SHiP++. On a four-core system, Glider improves IPC over LRU by 14.7%, compared with improvements of 13.6% (Hawkeye), 13.2% (MPPPB), and 11.4% (SHiP++). 

   more » « less
3. Monthly Crop Water Consumption of Irrigated Crops in the United States From 1981 to 2019

   https://doi.org/10.1029/2024WR038334  

   Lamsal, Gambhir; Marston, Landon_T (February 2025, Water Resources Research)

   Abstract Irrigated agriculture depends on surface water and groundwater, but we do not have a clear picture of how much water is consumed from these sources by different crops across the US over time. Current estimates of crop water consumption are insufficient in providing the spatial granularity and temporal depth required for comprehensive long‐term analysis. To fill this data gap, we utilized crop growth models to quantify the monthly crop water consumption ‐ distinguishing between rainwater, surface water, and groundwater ‐ of the 30 most widely irrigated crops in the US from 1981 to 2019 at 2.5 arc min. These 30 crops represent approximately 95% of US irrigated cropland. We found that the average annual total crop water consumption for these 30 irrigated crops in the US was 154.2 km³, 70% of which was from irrigation. Corn and alfalfa accounted for approximately 16.7 and 24.8 km³of average annual blue crop water consumption, respectively, which is nearly two‐fifths of the blue crop water consumed in the US. Surface water consumption decreased by 41.2%, while groundwater consumption increased by 6.8%, resulting in a 17.3% decline in blue water consumption between 1981 and 2019. We find good agreement between our results and existing modeled evapotranspiration (ET) products, remotely sensed ET estimates (OpenET), and water use data from the US Geological Survey and US Department of Agriculture. Our data set and model can help assess the impact of irrigation practices and water scarcity on crop production and sustainability. 

   more » « less
4. Potential for Managed Aquifer Recharge to Enhance Fish Habitat in a Regulated River

   https://doi.org/10.3390/w12030673  

   Kirk, Robert W.; Contor, Bryce A.; Morrisett, Christina N.; Null, Sarah E.; Loibman, Ashly S. (March 2020, Water)

   null (Ed.)

   Managed aquifer recharge (MAR) is typically used to enhance the agricultural water supply but may also be promising to maintain summer streamflows and temperatures for cold-water fish. An existing aquifer model, water temperature data, and analysis of water administration were used to assess potential benefits of MAR to cold-water fisheries in Idaho’s Snake River. This highly-regulated river supports irrigated agriculture worth US $10 billion and recreational trout fisheries worth $100 million. The assessment focused on the Henry’s Fork Snake River, which receives groundwater from recharge incidental to irrigation and from MAR operations 8 km from the river, addressing (1) the quantity and timing of MAR-produced streamflow response, (2) the mechanism through which MAR increases streamflow, (3) whether groundwater inputs decrease the local stream temperature, and (4) the legal and administrative hurdles to using MAR for cold-water fisheries conservation in Idaho. The model estimated a long-term 4%–7% increase in summertime streamflow from annual MAR similar to that conducted in 2019. Water temperature observations confirmed that recharge increased streamflow via aquifer discharge rather than reduction in river losses to the aquifer. In addition, groundwater seeps created summer thermal refugia. Measured summer stream temperature at seeps was within the optimal temperature range for brown trout, averaging 14.4 °C, whereas ambient stream temperature exceeded 19 °C, the stress threshold for brown trout. Implementing MAR for fisheries conservation is challenged by administrative water rules and regulations. Well-developed and trusted water rights and water-transaction systems in Idaho and other western states enable MAR. However, in Idaho, conservation groups are unable to engage directly in water transactions, hampering MAR for fisheries protection. 

   more » « less
5. Using Deep Learning in Ensemble Streamflow Forecasting: Exploring the Predictive Value of Explicit Snowpack Information

   https://doi.org/10.1029/2024MS004582  

   Modi, Parthkumar; Jennings, Keith; Kasprzyk, Joseph; Small, Eric; Wobus, Cameron; Livneh, Ben (March 2025, Journal of Advances in Modeling Earth Systems)

   Abstract The Ensemble Streamflow Prediction (ESP) framework combines a probabilistic forecast structure with process‐based models for water supply predictions. However, process‐based models require computationally intensive parameter estimation, increasing uncertainties and limiting usability. Motivated by the strong performance of deep learning models, we seek to assess whether the Long Short‐Term Memory (LSTM) model can provide skillful forecasts and replace process‐based models within the ESP framework. Given challenges inimplicitlycapturing snowpack dynamics within LSTMs for streamflow prediction, we also evaluated the added skill ofexplicitlyincorporating snowpack information to improve hydrologic memory representation. LSTM‐ESPs were evaluated under four different scenarios: one excluding snow and three including snow with varied snowpack representations. The LSTM models were trained using information from 664 GAGES‐II basins during WY1983–2000. During a testing period, WY2001–2010, 80% of basins exhibited Nash‐Sutcliffe Efficiency (NSE) above 0.5 with a median NSE of around 0.70, indicating satisfactory utility in simulating seasonal water supply. LSTM‐ESP forecasts were then tested during WY2011–2020 over 76 western US basins with operational Natural Resources Conservation Services (NRCS) forecasts. A key finding is that in high snow regions, LSTM‐ESP forecasts using simplified ablation assumptions performed worse than those excluding snow, highlighting that snow data do not consistently improve LSTM‐ESP performance. However, LSTM‐ESP forecasts that explicitly incorporated past years' snow accumulation and ablation performed comparably to NRCS forecasts and better than forecasts excluding snow entirely. Overall, integrating deep learning within an ESP framework shows promise and highlights important considerations for including snowpack information in forecasting. 

   more » « less