SEARCH · Search NASA
Results for “Climate Model Data”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
A study of the kinetic energy generation with general circulation models
The history data of winter simulation by the GLAS climate model and the NCAR community climate model are used to examine the generation of atmospheric kinetic energy. The contrast between the geographic distributions of the generation of kinetic energy and divergence of kinetic energy flux shows that kinetic energy is generated in the upstream side of jets, transported to the downstream side and destroyed there. The contributions from the time-mean and transient modes to the counterbalance between generation of kinetic energy and divergence of kinetic energy flux are also investigated. It is observed that the kinetic energy generated by the time-mean mode is essentially redistributed by the time-mean flow, while that generated by the transient flow is mainly responsible for the maintenance of the kinetic energy of the entire atmospheric flow.
Prototype energy models for data centers
Data centers in the United States consume about two percent of the nation’s electricity. Because heat gains from IT equipment drive cooling demand, data centers offer unique opportunities for energy savings. However, no prototype energy model for data centers is available in the suite of existing U.S. Department of Energy’s Commercial Prototype Building Models. Here we present the development of two new data center prototype models and their implementation in OpenStudio and EnergyPlus. The small-size data center model represents a computer room in a building served by computer room air conditioners (CRACs); while the large-sized model represents stand-alone data centers served by computer room air handlers (CRAHs) with a central chiller plant. For each data center model, two levels of IT equipment (ITE) load density were considered, to cover the wide range of IT power density of data centers: 40 and 100 W/ft 2 (430 and 1076 W/m 2 ) for the computer room, and 100 and 500 W/ft 2 (1076 and 5382 W/m 2 ) for the stand-alone data center. All other assumptions, such as building envelope, lighting, HVAC efficiencies and schedules, were based on the minimal requirements of ASHRAE Standard 90.1 at various vintages. We introduced a novel concept of supply and return air approach temperatures to capture the essential effects of non-uniform airflow and temperature distribution in data centers. The approach temperatures were pre-computed by computational fluid dynamics (CFD) simulations for various configurations of ITE loads and airflow containment management in data centers. A new feature was developed in EnergyPlus to implement the approach temperature method. A case study was conducted to demonstrate the use of the data center models. The two data center models cover all U.S. climate zones and can be used to evaluate energy saving measures for data centers, as well as to support development of data center energy efficiency codes and standards.
Climate-Water Adaptation for Future US Electricity Infrastructure
Future climate-water conditions are anticipated to increase electricity demand, reduce transmission capacity, and limit power production. Yet, typical electricity capacity expansion planning does not consider climate-water constraints. We project four alternative U.S. power system configurations using an iterative modeling and data exchange platform that integrates climate-driven hydrological, thermal power plant, and capacity expansion models. Through a comparison with traditional modeling approaches, we show that this novel approach provides greater confidence in electricity capacity projections by incorporating feasibility checks that adjust infrastructure development to reach grid reliability thresholds under climate-water constraints. Initial projections without climate-water impacts on electricity generation show future power systems become less vulnerable, independent of climate-water adaptation, as economic drivers increase renewable and natural gas-based capacity, while water-intensive coal and nuclear plants retire. However, power systems may face reliability challenges without climate-water adaptation, revealing the significance of incorporating climate-water impacts into power system planning. Climate-adjusted (Iterative approach) projections require a 5.3-12.0% increase in national-level capacity, relative to Initial projections, leading to an additional $125-143 billion (5.0-7.0%) in infrastructure costs. Variable renewable and natural gas technologies account for nearly all the additional capacity and, together with regional trade-offs in electricity generation, enhance grid performance to reach reliability thresholds. These adaptation transitions also lower water use and emissions, contributing to climate change mitigation, and highlight the trade-offs and impacts of both near and long-term electricity generation planning decisions.
Assessing the performance of solar radiation management geoengineering simulations
Offsetting the global warming caused by anthropogenic increases in atmospheric greenhouse gases by deliberate injection of aerosols into the stratosphere is the most studied of solar radiation management geoengineering schemes. The long-term success or failure of such schemes in achieving their stated goals is assessed by comparing simulated geoengineered temperature, precipitation and tropical cyclones metrics to equivalent fields in the simulated targeted climate simulations. Results using available data sets from three single model stabilized climate target experiments and three multimodel climate change reduction experiments are presented and compared against a measure of internal variability. While all but one experimental scheme is successful in achieving their targeted global mean annual surface temperature, their success at regional scales varies significantly and is often larger than the internal variability metric used here.
Climatically Diverse Data Set for Flat-Plate PV Module Model Validations (Presentation)
Photovoltaic (PV) module I-V curves were measured at Florida, Colorado, and Oregon locations to provide data for the validation and development of models used for predicting the performance of PV modules.
Studies of 21st-Century Precipitation Trends Over West Africa
West Africa includes a semi-arid zone between the Sahara Desert and the humid Gulf of Guinea coast, approximately between 10 N and 20 N, which is irrigated by summer monsoon rains. This article refers to the region as the Sahel. Rain-fed agriculture is the primary sustenance for Sahel populations, and severe droughts (in the 1970s and 1980s), therefore, have devastating negative societal impacts. The future frequency of Sahel droughts and the evolution of its hydrological balance are therefore of great interest. The article reviews 10 recent research studies that attempt to discover how climate changes will affect the hydrology of the Sahel throughout the 21st century. All 10 studies rely on atmosphere ocean global climate model (AOGCM) simulations based on a range of greenhouse gas emissions scenarios. Many of the simulations are contained in the Intergovernmental Panel on Climate Change archives for Assessment Reports #3 and #4. Two of the studies use AOGCM data to drive regional climate models. Seven studies make projections for the first half of the 21st century and eight studies make projections for the second half. Some studies make projections of wetter conditions and some predict more frequent droughts, and each describes the atmospheric processes associated with its prediction. Only one study projects more frequent droughts before 2050, and that is only for continent-wide degradation in vegetation cover. The challenge to correctly simulate Sahel rainfall decadal trends is particularly daunting because multiple physical mechanisms compete to drive the trend upwards or downwards. A variety of model deficiencies, regarding the simulation of one or more of these physical processes, taints models climate change projections. Consequently, no consensus emerges regarding the impact of anticipated greenhouse gas forcing on the hydrology of the Sahel in the second half of the 21st century.
Bird Migration Under Climate Change - A Mechanistic Approach Using Remote Sensing
The broad-scale reductions and shifts that may be expected under climate change in the availability and quality of stopover habitat for long-distance migrants is an area of increasing concern for conservation biologists. Researchers generally have taken two broad approaches to the modeling of migration behaviour to understand the impact of these changes on migratory bird populations. These include models based on causal processes and their response to environmental stimulation, "mechanistic models", or models that primarily are based on observed animal distribution patterns and the correlation of these patterns with environmental variables, i.e. "data driven" models. Investigators have applied the latter technique to forecast changes in migration patterns with changes in the environment, for example, as might be expected under climate change, by forecasting how the underlying environmental data layers upon which the relationships are built will change over time. The learned geostatstical correlations are then applied to the modified data layers.. However, this is problematic. Even if the projections of how the underlying data layers will change are correct, it is not evident that the statistical relationships will remain the same, i.e. that the animal organism may not adapt its' behaviour to the changing conditions. Mechanistic models that explicitly take into account the physical, biological, and behaviour responses of an organism as well as the underlying changes in the landscape offer an alternative to address these shortcomings. The availability of satellite remote sensing observations at multiple spatial and temporal scales, coupled with advances in climate modeling and information technologies enable the application of the mechanistic models to predict how continental bird migration patterns may change in response to environmental change. In earlier work, we simulated the impact of effects of wetland loss and inter-annual variability on the fitness of migratory shorebirds in the central fly ways of North America. We demonstrated the phenotypic plasticity of a migratory population of Pectoral sandpipers consisting of an ensemble of 10,000 individual birds in response to changes in stopover locations using an individual based migration model driven by remotely sensed land surface data, climate data and biological field data. With the advent of new computing capabilities enabled hy recent GPU-GP computing paradigms and commodity hardware, it now is possible to simulate both larger ensemble populations and to incorporate more realistic mechanistic factors into migration models. Here, we take our first steps use these tools to study the impact of long-term drought variability on shorebird survival.
Uncertainties in Predicting Rice Yield by Current Crop Models Under a Wide Range of Climatic Conditions
Predicting rice (Oryza sativa) productivity under future climates is important for global food security. Ecophysiological crop models in combination with climate model outputs are commonly used in yield prediction, but uncertainties associated with crop models remain largely unquantified. We evaluated 13 rice models against multi-year experimental yield data at four sites with diverse climatic conditions in Asia and examined whether different modeling approaches on major physiological processes attribute to the uncertainties of prediction to field measured yields and to the uncertainties of sensitivity to changes in temperature and CO2 concentration [CO2]. We also examined whether a use of an ensemble of crop models can reduce the uncertainties. Individual models did not consistently reproduce both experimental and regional yields well, and uncertainty was larger at the warmest and coolest sites. The variation in yield projections was larger among crop models than variation resulting from 16 global climate model-based scenarios. However, the mean of predictions of all crop models reproduced experimental data, with an uncertainty of less than 10 percent of measured yields. Using an ensemble of eight models calibrated only for phenology or five models calibrated in detail resulted in the uncertainty equivalent to that of the measured yield in well-controlled agronomic field experiments. Sensitivity analysis indicates the necessity to improve the accuracy in predicting both biomass and harvest index in response to increasing [CO2] and temperature.
Laboratory Experiments and Instrument Intercomparison Studies of Carbonaceous Aerosol Particles
Aerosols containing black carbon (and some specific types of organic particulate matter) directly absorb incoming light, heating the atmosphere. In addition, all aerosol particles backscatter solar light, leading to a net-cooling effect. Indirect effects involve hydrophilic aerosols, which serve as cloud condensation nuclei (CCN) that affect cloud cover and cloud stability, impacting both atmospheric radiation balance and precipitation patterns. At night, all clouds produce local warming, but overall clouds exert a net-cooling effect on the Earth. The effect of aerosol radiative forcing on climate may be as large as that of the greenhouse gases, but predominantly opposite in sign and much more uncertain. The uncertainties in the representation of aerosol interactions in climate models makes it problematic to use model projections to guide energy policy. The objective of our program is to reduce the uncertainties in the aerosol radiative forcing in the two areas highlighted in the ASR Science and Program Plan. That is, (1) addressing the direct effect by correlating particle chemistry and morphology with particle optical properties (i.e. absorption, scattering, extinction), and (2) addressing the indirect effect by correlating particle hygroscopicity and CCN activity with particle size, chemistry, and morphology. In this connection we are systematically studying particle formation, oxidation, and the effects of particle coating. The work is specifically focused on carbonaceous particles where the uncertainties in the climate relevant properties are the highest. The ongoing work consists of laboratory experiments and related instrument inter-comparison studies both coordinated with field and modeling studies, with the aim of providing reliable data to represent aerosol processes in climate models. The work is performed in the aerosol laboratory at Boston College. At the center of our laboratory setup are two main sources for the production of aerosol particles: (a) two well-characterized source of soot particles and (b) a flow reactor for controlled OH and/or O3 oxidation of relevant gas phase species to produce well-characterized SOA particles. After formation, the aerosol particles are subjected to physical and chemical processes that simulate aerosol growth and aging. A suite of instruments in our laboratory is used to characterize the physical and chemical properties of aerosol particles before and after processing. The Time of Flight Aerosol Mass Spectrometer (ToF-AMS) together with a Scanning Mobility Particle Sizer (SMPS) measures particle mass, volume, density, composition (including black carbon content), dynamic shape factor, and fractal dimension. The–ToF-AMS was developed at ARI with Boston College participation. About 120 AMS instruments are now in service (including 5 built for DOE laboratories) performing field and laboratory studies world-wide. Other major instruments include a thermal denuder, two Differential Mobility Analyzers (DMA), a Cloud Condensation Nuclei Counter (CCN), a Thermal desorption Aerosol GC/MS (TAG) and the new Soot Particle Aerosol Mass Spectrometer (SP-AMS). Optical instrumentation required for the studies have been brought to our laboratory as part of ongoing and planned collaborative projects with colleagues from DOE, NOAA and university laboratories. Optical instruments that will be utilized include a Photoacoustic Spectrometer (PAS), a Cavity Ring Down Aerosol Extinction Spectrometer (CRD-AES), a Photo Thermal Interferometer (PTI), a new 7-wavelength Aethalometer and a Cavity Attenuated Phase Shift Extinction Monitor (CAPS). These instruments are providing aerosol absorption, extinction and scattering coefficients at a range of atmospherically relevant wavelengths. During the past two years our work has continued along the lines of our original proposal. We report on 12 completed and/or continuing projects conducted during the period 08/14 to 0814/2015. These projects are described in 17 manuscripts published in refereed journals.
Identifying Climate Patterns Using Clustering Autoencoder Techniques
Abstract The complexity of growing spatiotemporal resolution of climate simulations produces a variety of climate patterns under different projection scenarios. This paper proposes a new data-driven climate classification workflow via an unsupervised deep learning technique that can dimensionally reduce the vast volume of spatiotemporal numerical climate projection data into a compact representation. We aim to identify distinct zones that capture multiple climate variables as well as their future changes under different climate change scenarios. Our approach leverages convolutional autoencoders combined with k -means clustering (standard autoencoder) and online clustering based on the Sinkhorn–Knopp algorithm (clustering autoencoder) across the conterminous United States (CONUS) to capture unique climate patterns in a data-driven fashion from the Geophysical Fluid Dynamics Laboratory Earth System Model with GOLD component (GFDL-ESM2G). The developed approach compresses 70 years of GFDL-ESM2G simulation at 0.125° spatial resolution across the CONUS under multiple warming scenarios to a lower-dimensional space by a factor of 660 000 and then tested on 150 years of GFDL-ESM2G simulation data. The results show that five climate clusters capture physically reasonable and spatially stable climatological patterns matched to known climate classes defined by human experts. Results also show that using a clustering autoencoder can reduce the computational time for clustering by up to 9.2 times when compared to using a standard autoencoder. Our five unique climate patterns resulting from the deep learning–based clustering of the lower-dimensional space thereby enable us to provide insights on hydrometeorology and its spatial heterogeneity across the conterminous United States immediately without downloading large climate datasets. Significance Statement This paper presents a data-driven climate classification approach using unsupervised deep learning to dimensionally reduce climate model outputs and to identify distinct climate regions for their future changes. Our approach compresses climate information for 70 years of Geophysical Fluid Dynamics Laboratory Earth System Model data across the conterminous United States (CONUS) at 0.125° spatial resolution. The results reveal that five climate clusters capture reasonable and stable climatological patterns matched to known climate patterns. The embedded clustering process in deep learning provides ×9.2 times faster execution than the k -means clustering technique. These results give us insight about climate spatial patterns and heterogeneity of hydrological patterns across the conterminous United States without downloading large climate datasets.
A Multi-Model Framework for Assessing Long- and Short-Term Climate Influences on the Electric Grid
Climate change influences many aspects of the electric grid, but prior work and industry practices often ignore the potential effects of changing climate, or they only consider a single effect or individual effects in isolation. Challenges vary with each grid and include adapting to long-term trends such as changing temperature and precipitation or shorter-term events such as drought or storms that could increase in frequency or intensity. Here we present a multi-model framework designed to analyze the effects of long and short-term climate impacts in combination. This framework couples capacity expansion and production cost models with hydrologic models and future climate scenario data to analyze alternative climate and energy futures at high spatial, temporal, and process resolutions. Furthermore, we constructed and evaluated the results of a suite of simulated scenarios exploring climate impacts on capacity investment and stress-tested the resulting future infrastructures using hourly dispatch modeling under alternative drought and load conditions. We demonstrate the approach through a case study of the U.S. Western Interconnection, where climate impacts depend on interactions between temperature-induced load, water availability for hydropower, technology competitiveness, and demand flexibility. Changes in 2038 generating capacity range from -8.5-16.6 GW, and changes in 2038 transmission capacity range from -1-2 GW. Capacity increases are driven by higher load from higher temperatures, while capacity reductions can be achieved in scenarios with higher future hydropower availability and increased demand flexibility. Scenarios requiring additional capacity cost an additional $\$5$-$\$17$ billion (discounted) from 2018 to 2038; however, scenarios with capacity reductions cost $\$1$-$\$18$ billion less. Stress tests on four 2038 infrastructures demonstrated that the identified systems were able to serve at least 99.999% of load and 99.96% of reserves. However, drought and unexpected high-load conditions can result in reduced capacity to respond to contingency events we did not model. Although these results are system and scenario specific, they highlight the importance of considering multiple climate change impacts simultaneously in long-term planning efforts and demonstrate a multi-model, multiscale approach that can be flexibly applied to any system and set of climate change concerns.
Surface Melting Over the Greenland Ice Sheet Derived From Enhanced Resolution Passive Microwave Brightness Temperatures (1979–2019)
Surface melting is a major component of the Greenland ice sheet surface mass balance, and it affects sea level rise through direct runoff and the modulation of ice dynamics and hydrological processes, supraglacially, englacially and subglacially. Passive microwave (PMW) brightness temperature observations are of paramount importance in studying the spatial and temporal evolution of surface melting due to their long temporal coverage (1979–present) and high temporal resolution (daily). However, a major limitation of PMW datasets has been the relatively coarse spatial resolution, which has historically been of the order of tens of kilometers. Here, we use a newly released PMW dataset (37 GHz, horizontal polarization) made available through a NASA “Making Earth System Data Records for Use in Research Environments” (MeASUREs) program to study the spatiotemporal evolution of surface melting over the Greenland ice sheet at an enhanced spatial resolution of 3.125 km. We assess the outputs of different detection algorithms using data collected by automatic weather stations (AWSs) and the outputs of the Modèle Atmosphérique Régional (MAR) regional climate model. We found that sporadic melting is well captured using a dynamic algorithm based on the outputs of the Microwave Emission Model of Layered Snowpack (MEMLS), whereas a fixed threshold of 245 K is capable of detecting persistent melt. Our results indicate that, during the reference period from 1979 to 2019 (from 1988 to 2019), surface melting over the ice sheet increased in terms of both duration, up to 4.5 (2.9) d per decade, and extension, up to 6.9 % (3.6 %) of the entire ice sheet surface extent per decade, according to the MEMLS algorithm. Furthermore, the melting season started up to 4.0 (2.5) d earlier and ended 7.0 (3.9) d later per decade. We also explored the information content of the enhanced-resolution dataset with respect to the one at 25 km and MAR outputs using a semi-variogram approach. We found that the enhanced product is more sensitive to local-scale processes, thereby confirming the potential of this new enhanced product for monitoring surface melting over Greenland at a higher spatial resolution than the historical products and for monitoring its impact on sea level rise. This offers the opportunity to improve our understanding of the processes driving melting, to validate modeled melt extent at high resolution and, potentially, to assimilate these data in climate models.
Reconstructed Historical Land Cover and Biophysical Parameters for Studies of Land-Atmosphere Interactions within the Eastern United States
The local environment where we live within the Earth's biosphere is often taken for granted. This environment can vary depending on whether the land cover is a forest, grassland, wetland, water body, bare soil, pastureland, agricultural field, village, residential suburb, or an urban complex with concrete, asphalt, and large buildings. In general, the type and characteristics of land cover influence surface temperatures, sunlight exposure and duration, relative humidity, wind speed and direction, soil moisture amount, plant life, birds, and other wildlife in our backyards. The physical and biological properties (biophysical characteristics) of land cover help to determine our surface environment because they directly affect surface radiation, heat, and soil moisture processes, and also feedback to regional weather and climate. Depending on the spatial scale and land use intensity, land cover changes can have profound impacts on our local and regional environment. Over the past 350 years, the eastern half of the United States, an area extending from the grassland prairies of the Great Plains to the Gulf and Atlantic coasts, has experienced extensive land cover and land use changes that began with land clearing in the 1600s, led to extensive deforestation and intensive land use practices by 1920, and then evolved to the present-day landscape. Determining the consequences of such land cover changes on regional and global climate is a major research issue. Such research requires detailed historical land cover data and modeling experiments simulating historical climates. Given the need to understand the effects of historical land cover changes in the eastern United States, some questions include: - What were the most important land cover transformations and how did they alter biophysical characteristics of the land cover at key points in time since the mid-1600s? - How have land cover and land use changes over the past 350 years affected the land surface environment including surface weather, hydrologic, and climatic variability? - How do the potential effects of regional human-induced land cover change on the environment compare to similar changes that are caused by the natural variations of the Earth's climate system? To help answer these questions, we reconstructed a fractional land cover and biophysical parameter dataset for the eastern United States at 1650, 1850, 1920, and 1992 time-slices. Each land cover fraction is associated with a biophysical parameter class, a suite of parameters defining the biophysical characteristics of that kind of land cover. This new dataset is designed for use in computer models of land-atmosphere interactions, to understand and quantify the effects of historical land cover changes on the water, energy, and carbon cycles
Probabilistic Modeling of Climate Change Impacts on Renewable Energy and Storage Requirements for NM's Energy Transition Act (SAND Report)
This report provides a study of the potential impacts of climate change on intermittent renewable energy resources, battery storage, and resource adequacy in Public Service Company of New Mexico’s Integrated Resource Plan for 2020 – 2040. Climate change models and available data were first evaluated to determine uncertainty and potential changes in solar irradiance, temperature, and wind speed in NM in the coming decades. These changes were then implemented in solar and wind energy models to determine impacts on renewable energy resources in NM. Results for the extreme climate-change scenario show that the projected wind power may decrease by ~13% due to projected decreases in wind speed. Projected solar power may decrease by ~4% due to decreases in irradiance and increases in temperature in NM. Uncertainty in these climate-induced changes in wind and solar resources was accommodated in probabilistic models assuming uniform distributions in the annual reductions in solar and wind resources. Uncertainty in battery storage performance was also evaluated based on increased temperature, capacity fade, and degradation in round-trip efficiency. The hourly energy balance was determined throughout the year given uncertainties in the renewable energy resources and energy storage. The loss of load expectation (LOLE) was evaluated for the 2040 No New Combustion portfolio and found to increase from 0 days/year to a median value of ~2 days/year due to potential reductions in renewable energy resources and battery storage performance and capacity. A rank-regression analyses revealed that battery round-trip efficiency was the most significant parameter that impacted LOLE, followed by solar resource, wind resource, and battery fade. An increase in battery storage capacity to ~25,000 – 30,000 MWh from a baseline value of ~14,000 MWh was required to reduce the median value of LOLE to ~0.2 days/year with consideration of potential climate impacts and battery degradation.
Feature Tracking and Visualization of Madden-Julian Osciallation in Climate Simulation
Madden-Julian Oscillation (MJO) is one of the less understood aspects of tropical meteorology, which plays a significant role in tropical intra-seasonal variations in rain, temperature and winds over the Indian and Pacific Oceans. In this paper, we present an integrated analysis and visualization framework for MJO episodes simulated by a high resolution regional model. To distinguish MJOs from other weather phenomena, our framework utilizes domain knowledge to track MJOs as finding the globally optimized properties in the data. In addition to enhancing the animation with feature tracking, our visualization system also integrates different visualization components such as Virtual Globe and Hovmoller Diagrams to visualize large scale events both in space and time. By linking all of these visualization components on a web-based interface, scientists can identify cloud and environmental processes associated with the initiation and eastward propagation of MJO more easily.
The Changing Character of the California Sierra Nevada as a Natural Reservoir
The mountains of the Western United States provide a vital natural service through the storage and release of mountain snowpack, lessening impacts of seasonal aridity and satiating summer water demand. Yet, climate change continues to undermine these important processes. To understand how snowpack may change in the headwaters of California's major reservoirs, the North American Coordinated Regional Climate Downscaling Experiment is analyzed to assess peak water volume, peak timing, accumulation rate, melt rate, and snow season length across both latitudinal and elevational gradients. Under a high-emissions scenario, end-of-century peak snowpack timing occurs 4 weeks earlier and peak water volume is 79.3% lower. The greatest reductions are above Shasta, Oroville, and Folsom and between 0- and 2,000-m elevations. Regional climate model and global forcing data set choice is important in determining historical snowpack character, yet by end century all models show a significant and similar decline in mountain snowpack.
Temporal sequence transformer to advance long-term streamflow prediction
Accurate streamflow prediction is crucial for understanding climate change impacts on water resources and for effective management of extreme hydrological events. While Long Short-Term Memory (LSTM) networks have been the dominant data-driven approach for streamflow forecasting, recent advancements in transformer architectures for time series tasks have shown promise in outperforming traditional LSTM models. This study introduces a transformer-based model that integrates historical streamflow data with climatic variables to enhance streamflow prediction accuracy. We evaluated our transformer model against a benchmark LSTM across five diverse basins in the United States. Results demonstrate that the transformer architecture consistently outperforms the LSTM model across all evaluation metrics, highlighting its potential as a more effective tool for hydrological forecasting. This research contributes to the ongoing development of advanced AI techniques for improved water resource management and climate change adaptation strategies.