Daily evapotranspiration (ET_d) plays a key role in irrigation water management and is particularly important in drought-stricken areas, such as California and high-value crops. Remote sensing allows for the cost-effective estimation of spatial evapotranspiration (ET), and the advent of small unmanned aerial systems (sUAS) technology has made it possible to estimate instantaneous high-resolution ET at the plant, row, and subfield scales. sUAS estimates ET using “instantaneous” remote sensing measurements with half-hourly/hourly forcing micrometeorological data, yielding hourly fluxes in W/m² that are then translated to a daily scale (mm/day) under two assumptions: (a) relative rates, such as the ratios of ET-to-net radiation (R_n) or ET-to-solar radiation (R_s), are assumed to be constant rather than absolute, and (b) nighttime evaporation (E) and transpiration (T) contributions are negligible. While assumption (a) may be reasonable for unstressed, full cover crops (no exposed soil), the E and T rates may significantly vary over the course of the day for partially vegetated cover conditions due to diurnal variations of soil and crop temperatures and interactions between soil and vegetation elements in agricultural environments, such as vineyards and orchards. In this study, five existing extrapolation approaches that compute the daily ET from the “instantaneous” remotely sensed sUAS ET estimates and the eddy covariance (EC) flux tower measurements were evaluated under different weather, grapevine variety, and trellis designs. Per assumption (b), the nighttime ET contribution was ignored. Each extrapolation technique (evaporative fraction (EF), solar radiation (R_s), net radiation-to-solar radiation (R_n/R_s) ratio, Gaussian (GA), and Sine) makes use of clear skies and quasi-sinusoidal diurnal variations of hourly ET and other meteorological parameters. The sUAS ET estimates and EC ET measurements were collected over multiple years and times from different vineyard sites in California as part of the USDA Agricultural Research Service Grape Remote Sensing Atmospheric Profile and Evapotranspiration eXperiment (GRAPEX). Optical and thermal sUAS imagery data at 10 cm and 60 cm, respectively, were collected by the Utah State University AggieAir sUAS Program and used in the Two-Source Energy Balance (TSEB) model to estimate the instantaneous or hourly sUAS ET at overpass time. The hourly ET from the EC measurements was also used to validate the extrapolation techniques. Overall, the analysis using EC measurements indicates that the R_s, EF, and GA approaches presented the best goodness-of-fit statistics for a window of time between 1030 and 1330 PST (Pacific Standard Time), with the R_s approach yielding better agreement with the EC measurements. Similar results were found using TSEB and sUAS data. The 1030–1330 time window also provided the greatest agreement between the actual daily EC ET and the extrapolated TSEB daily ET, with the R_s approach again yielding better agreement with the ground measurements. The expected accuracy of the upscaled TSEB daily ET estimates across all vineyard sites in California is below 0.5 mm/day, (EC extrapolation accuracy was found to be 0.34 mm/day), making the daily scale results from TSEB reliable and suitable for day-to-day water management applications.

Evapotranspiration (ET) is a key variable for hydrology and irrigation water management, with significant importance in drought-stricken regions of the western US. This is particularly true for California, which grows much of the high-value perennial crops in the US. The advent of small Unmanned Aerial System (sUAS) with sensor technology similar to satellite platforms allows for the estimation of high-resolution ET at plant spacing scale for individual fields. However, while multiple efforts have been made to estimate ET from sUAS products, the sensitivity of ET models to different model grid size/resolution in complex canopies, such as vineyards, is still unknown. The variability of row spacing, canopy structure, and distance between fields makes this information necessary because additional complexity processing individual fields. Therefore, processing the entire image at a fixed resolution that is potentially larger than the plant-row separation is more efficient. From a computational perspective, there would be an advantage to running models at much coarser resolutions than the very fine native pixel size from sUAS imagery for operational applications. In this study, the Two-Source Energy Balance with a dual temperature (TSEB2T) model, which uses remotely sensed soil/substrate and canopy temperature from sUAS imagery, was used to estimate ET and identify the impact of spatial domain scale under different vine phenological conditions. The analysis relies upon high-resolution imagery collected during multiple years and times by the Utah State University AggieAir^TM sUAS program over a commercial vineyard located near Lodi, California. This project is part of the USDA-Agricultural Research Service Grape Remote Sensing Atmospheric Profile and Evapotranspiration eXperiment (GRAPEX). Original spectral and thermal imagery data from sUAS were at 10 cm and 60 cm per pixel, respectively, and multiple spatial domain scales (3.6, 7.2, 14.4, and 30 m) were evaluated and compared against eddy covariance (EC) measurements. Results indicated that the TSEB2T model is only slightly affected in the estimation of the net radiation (R_n) and the soil heat flux (G) at different spatial resolutions, while the sensible and latent heat fluxes (H and LE, respectively) are significantly affected by coarse grid sizes. The results indicated overestimation of H and underestimation of LE values, particularly at Landsat scale (30 m). This refers to the non-linear relationship between the land surface temperature (LST) and the normalized difference vegetation index (NDVI) at coarse model resolution. Another predominant reason for LE reduction in TSEB2T was the decrease in the aerodynamic resistance (R_a), which is a function of the friction velocity ( $u_{*}$ ) that varies with mean canopy height and roughness length. While a small increase in grid size can be implemented, this increase should be limited to less than twice the smallest row spacing present in the sUAS imagery. The results also indicated that the mean LE at field scale is reduced by 10% to 20% at coarser resolutions, while the with-in field variability in LE values decreased significantly at the larger grid sizes and ranged between approximately 15% and 45%. This implies that, while the field-scale values of LE are fairly reliable at larger grid sizes, the with-in field variability limits its use for precision agriculture applications.

Streamflow prediction plays a vital role in water resources planning in order to understand the dramatic change of climatic and hydrologic variables over different time scales. In this study, we used machine learning (ML)-based prediction models, including Random Forest Regression (RFR), Long Short-Term Memory (LSTM), Seasonal Auto- Regressive Integrated Moving Average (SARIMA), and Facebook Prophet (PROPHET) to predict 24 months ahead of natural streamflow at the Lees Ferry site located at the bottom part of the Upper Colorado River Basin (UCRB) of the US. Firstly, we used only historic streamflow data to predict 24 months ahead. Secondly, we considered meteorological components such as temperature and precipitation as additional features. We tested the models on a monthly test dataset spanning 6 years, where 24-month predictions were repeated 50 times to ensure the consistency of the results. Moreover, we performed a sensitivity analysis to identify our best-performing model. Later, we analyzed the effects of considering different span window sizes on the quality of predictions made by our best model. Finally, we applied our best-performing model, RFR, on two more rivers in different states in the UCRB to test the model’s generalizability. We evaluated the performance of the predictive models using multiple evaluation measures. The predictions in multivariate time-series models were found to be more accurate, with RMSE less than 0.84 mm per month, R-squared more than 0.8, and MAPE less than 0.25. Therefore, we conclude that the temperature and precipitation of the UCRB increases the accuracy of the predictions. Ultimately, we found that multivariate RFR performs the best among four models and is generalizable to other rivers in the UCRB.

Streamflow prediction is vital for effective water resource management, enabling a better understanding of hydrological variability and its response to environmental factors. This study presents a spatio-temporal graph neural network (STGNN) model for streamflow prediction in the Upper Colorado River Basin (UCRB), integrating graph convolutional networks (GCNs) to model spatial connectivity and long short-term memory (LSTM) networks to capture temporal dynamics. Using 30 years of monthly streamflow data from 20 monitoring stations, the STGNN predicted streamflow over a 36-month horizon and was evaluated against traditional models, including random forest regression (RFR), LSTM, gated recurrent units (GRU), and seasonal auto-regressive integrated moving average (SARIMA). The STGNN outperformed these models across multiple metrics, achieving an R² of 0.78, an RMSE of 0.81 mm/month, and a KGE of 0.79 at critical locations like Lees Ferry. A sequential analysis of input–output configurations identified the (36, 36) setup as optimal for balancing historical context and forecasting accuracy. Additionally, the STGNN showed strong generalizability when applied to other locations within the UCRB. These results underscore the importance of integrating spatial dependencies and temporal dynamics in hydrological forecasting, offering a scalable and adaptable framework to improve predictive accuracy and support adaptive water resource management in river basins.

Understanding the spatial variability in highly heterogeneous natural environments such as savannas and river corridors is an important issue in characterizing and modeling energy fluxes, particularly for evapotranspiration (ET) estimates. Currently, remote-sensing-based surface energy balance (SEB) models are applied widely and routinely in agricultural settings to obtain ET information on an operational basis for use in water resources management. However, the application of these models in natural environments is challenging due to spatial heterogeneity in vegetation cover and complexity in the number of vegetation species existing within a biome. In this research effort, small unmanned aerial systems (sUAS) data were used to study the influence of land surface spatial heterogeneity on the modeling of ET using the Two-Source Energy Balance (TSEB) model. The study area is the San Rafael River corridor in Utah, which is a part of the Upper Colorado River Basin that is characterized by arid conditions and variations in soil moisture status and the type and height of vegetation. First, a spatial variability analysis was performed using a discrete wavelet transform (DWT) to identify a representative spatial resolution/model grid size for adequately solving energy balance components to derive ET. The results indicated a maximum wavelet energy between 6.4 m and 12.8 m for the river corridor area, while the non-river corridor area, which is characterized by different surface types and random vegetation, does not show a peak value. Next, to evaluate the effect of spatial resolution on latent heat flux (LE) estimation using the TSEB model, spatial scales of 6 m and 15 m instead of 6.4 m and 12.8 m, respectively, were used to simplify the derivation of model inputs. The results indicated small differences in the LE values between 6 m and 15 m resolutions, with a slight decrease in detail at 15 m due to losses in spatial variability. Lastly, the instantaneous (hourly) LE was extrapolated/upscaled to daily ET values using the incoming solar radiation (R_s) method. The results indicated that willow and cottonwood have the highest ET rates, followed by grass/shrubs and treated tamarisk. Although most of the treated tamarisk vegetation is in dead/dry condition, the green vegetation growing underneath resulted in a magnitude value of ET.

With increasing demands for precise water resource management, there is a growing need for advanced techniques in mapping water bodies. The currently deployed satellites provide complementary data that are either of high spatial or high temporal resolutions. As a result, there is a clear trade‐off between space and time when considering a single data source. For the efficient monitoring of multiple environmental resources, various Earth science applications need data at high spatial and temporal resolutions. To address this need, many data fusion methods have been described in the literature, that rely on combining data snapshots from multiple sources. Traditional methods face limitations due to sensitivity to atmospheric disturbances and other environmental factors, resulting in noise, outliers, and missing data. This paper introduces Hydrological Generative Adversarial Network (Hydro‐GAN), a novel machine learning‐based method that utilizes modified GANs to enhance boundary accuracy when mapping low‐resolution MODIS data to high‐resolution Landsat‐8 images. We propose a new non‐saturating loss function for the Hydro‐GAN generator, which maximizes the log of discriminator probabilities to promote stable updates and aid convergence. By focusing on reducing squared differences between real and synthetic images, our approach enhances training stability and overall performance. We specifically focus on mapping water bodies using MODIS and Landsat‐8 imagery due to their relevance in water resource management tasks. Our experimental results demonstrate the effectiveness of Hydro‐GAN in generating high‐resolution water body maps, outperforming traditional methods in terms of boundary accuracy and overall quality.