Evapotranspiration Estimation Methods Research Articles

Sensitivity of global hydrological models to potential evapotranspiration estimation methods in the Senegal River Basin (West Africa)

Study regionSenegal River Basin in West Africa Study focusThis paper aims to evaluate the sensitivity of global hydrological models to potential evapotranspiration (PET) methods in the Senegal River Basin. Potential evapotranspiration is estimated using 21 methods and its influence on the performance of three GR models (GR4J, GR5J and GR6J) is investigated in five catchments. The data used are mean rainfall, discharge and observed and calculated PET over the period 1984–1995. PET is calculated based on observed climate data and those from NASA POWER reanalysis data. The methodology consists in: (i) comparing the consistency of reanalysis data with respect to the observed PET, (ii) assessing the robustness of GR models and their sensitivity to different PET estimation methods. The evaluation criteria used to assess the performance of the hydrological model are KGE and PBIAS. New hydrological insights for the regionGood consistency is obtained between PET calculated with observed and reanalysis data. The GR4J and GR5J are more efficient to simulate flows in the Senegal River sub-basins. The aerodynamic PET methods perform well with the three hydrological models. However, in this context where data are scarce, temperature methods such as Droogers and Allen are a good choice for hydrological modeling. The results also show that the GR models have the ability to adapt to poorly estimated PET.

A review on models, products and techniques for evapotranspiration measurement, estimation, and validation

AbstractIn this review study, the major available methods for measurement and estimation of evapotranspiration (ET) are discussed briefly while explaining the latest developments. The best available validation methods are also reviewed and explained. It highlights the importance of accurate ET quantification in managing water resources, evaluating climate change impacts, and supporting crop water requirement management. Measurement methods such as scintillometry, lysimetry, and the eddy covariance (EC) flux method are presented. Additionally, hydrological models are discussed as estimation approaches for actual and potential ET. The paper explores various ET estimation products, particularly those based on remote sensing techniques. Specifically, methods like Mapping EvapoTranspiration at high Resolution with Internalized Calibration (METRIC), Simplified Surface Energy Balance Operational (SSEBop ET), Moderate Resolution Imaging Spectroradiometer (MOD16), Surface Energy Balance Algorithm for Land (SEBAL), Global Land Surface Evaporation: Amsterdam Methodology (GLEAM), Satellite Application Facility on Land Surface Analysis (LSA‐SAF), and Global Land Data Assimilation System (GLDAS) are described. The integration of machine learning (ML) with EC and remote sensing is investigated, with a comprehensive discussion of different ML approaches. Validation methods including the EC method, water balance method‐derived ET (WBET), and statistical techniques are explained. Overall, this review paper provides a comprehensive overview of ET quantification, covering measurement techniques, estimation approaches, remote sensing methods, and the integration of ML. The insights gained from this review contribute to a profound knowledge of ET dynamics and helps those sectors dealing with drought monitoring, water resource management and climate change assessments.

Observed interannual variability and projected scenarios of drought in the Chorotega region, Costa Rica

The observation-based analysis of drought development in the Chorotega region showed that, despite the area being relatively small, agricultural drought exhibits high spatial variability across the region. However, the lack of net radiation data hinders the capacity to provide reliable estimates of evapotranspiration (ET), affecting the assessment of drought occurrence, since its propagation across the hydrological system is very sensitive to the ET estimation method. The coarse resolution of satellite-derived Normalized Difference Vegetation Index (NDVI) products and the lack of information on irrigation in agricultural areas limits the ability to properly establish a relationship between drought and vegetation response. Based on the observations, the most prominent precipitation deficits occur between September and October (–100 mm on average), showing that changes in the large-scale circulation are responsible for the impact of severe drought in the region. In agreement with previous studies, El Niño-Southern Oscillation (ENSO) is the main modulator of the drought severity, with the warm ENSO phase favoring an enhanced drought development and its influence being more significant between August and October, displaying correlations greater than –0.6. The climate change projections under RCP4.5 and RCP8.5 scenarios suggest the intensification of drought events in the Chorotega region at mid-century, with the Tempisque-Bebedero basin being the most affected area in terms of precipitation decrease and warming. The projected scenarios correspond to an increase of 1 oC for mean temperature and more of 2 oC for minimum and maximum temperature in the 2050 horizon, as well as a decrease of 400 to 800 mm for annual precipitation under both RCPs.

Assessment of crop water requirement and irrigation scheduling of potato using weather models under sub temperate climatic condition of North-Western Himalaya

To optimize irrigation water and enhance yield, appropriate amount of irrigation water and its scheduling according to real time weather data can help in reducing crop water foot prints. The evapotranspiration (ET) based irrigation scheduling using present weather conditions reflecting changed climatic scenarios in the study regions in the past few decades has proved to be sagacious use of water. To study the ET based irrigation scheduling, field experiment conducted with the treatments comprising four ET based weather models irrigation scheduling methods viz. no irrigation or rainfed, Thornthwaite, Hargreaves method, Temperature and modified Penman Monteith under sowing windows of 20th December and 20th January were conducted at research farm, of Department of Agronomy, CSK HPKV, Palampur, Himachal Pradesh during Rabi season of 2015-16. The estimated crop evapo-transpiration (ETc) with different ET estimation methods under North Western Himalayas showed variation during crop season. The highest estimated ET was observed in December as compared to January sown potato crop with Hargreaves method (376.8 mm) followed by Temperature (354 mm), Thornthwaite (329.5 mm) and the lowest in modified Penman Monteith methods (241.9 mm). The estimated ETc from pan evaporation observed to be the highest to the tune of (382.3 mm) compared to other methods of ETc estimation. The crop coefficients modeled at different pheno-phases were used for estimating water requirement of potato crop. It was found that irrigation scheduling based on modified Penman Monteith of ET estimation saved one irrigation to the tune of 60 mm during crop period without affecting significantly growth, and tuber yield of the potato as compared to other methods, viz., However, the irrigation schedule based on Thornthwaite method has recorded highest total potato tuber yield (171.76 q ha-1) being at par with irrigation schedules worked out from other ET estimation methods. The total water footprints were the lowest in modified Penman Monteith to the tune of 300 liters per kg of potato yield compared to other methods.

Irrigation Scheduling Based on Evapotranspiration of Tomato (Solanumm Lycopersicum) Using Atmometer in The Upland Rolling Production Area

Objective : The goal is to provide the right amount of water to the crop, at the right time, to meet the crop's needs. It was conducted to estimate evapotranspiration using atmometers for irrigation scheduling that may be highly beneficial in soil water balance, water allocation and budgeting, irrigation management, and improving crop yield and production for greater income. Methods: This study was conducted at the Organic Agricultural Research and Development Innovation Center at the Bataan Peninsula State University – Abucay Campus, Bangkal, Abucay, Bataan, Philippines (North 14°44'28” East 120°27'04”). The effects of using different evapotranspiration estimation methods on the timing and amounts of water application were then evaluated by using a computed irrigation scheduling model. Findings: The atmometer ( and Penman-Monteith ( values were statistically analyzed using linear regression (y=0.8573x + 1.586), coefficients of determination (R² = 0.7236), root mean – squared error (RMSD = 0.73), mean bias error (MBE = 0.03), and the t-statistics (significant at 5% level). The study revealed that the ETo-A data strongly correlated with the ETo-PM data. Using an atmometer for scheduling irrigation on tomatoes would have resulted in an equally similar distribution of irrigation events, more water would have been provided throughout the season compared to using ETo-PM data. Novelty : The calibrated evapotranspiration data from the atmometer may guide appropriate irrigation depth and schedule for tomatoes. By aligning irrigation intervals with other crops’ water requirements, farmers can optimize water usage, minimize water stress, and achieve higher crop yields. Keywords: Atmometer, Bataan Philippines, Evapotranspiration, Tomato, Penman-­Montheith Equation

A COMPARATIVE RELATION OF DISTINCT REFERENCE CROP EVAPOTRANSPIRATION MODELS FOR SOUTHERN BRAZIL

Water is one of the main limiting factors for achieving high productivity in agriculture. The hydric requirement of plants is fundamental for the dimensioning of the irrigation system and contributes to the better use of hydric resources. Moreover, the accurate computation of this element is essential for water management in agricultural systems. Nonetheless, due to the heterogeneity of different evapotranspiration estimation methods, the performance of its calculation can be considerably compromised. Accordingly, the aim of this study was to compare the methods for estimating reference evapotranspiration (ETo) by Benevides & Lopes, Camargo, Hargreaves & Samani, Jensen & Haise, Linacre, Makkink, Penman, Priestley & Taylor, Tanner & Pelton, and Turc, with the FAO-56 Penman-Monteith standard method, to evaluate the performance and accuracy of equational models. Furthermore, data from an automatic weather station belonging to the Brazilian National Institute of Meteorology (INMET), located in Palmeira das Missões, Rio Grande do Sul, Brazil, from January 1, 2020, to January 1, 2021, were used. Comparative statistical methods were utilized to express the accuracy of the models and indicate the most appropriate equations for the conditions of the selected location. Cluster analysis and Principal Component Analysis (PCA) were applied. For Palmeira das Missões, the model proposed by Hargreaves & Samani indicated the best results and was characterized as the most appropriate alternative to estimate the ETo more accurately. The method indicated the most favorable results for R2 (0.9890), d (0.9253), and r (0.9944). Furthermore, cluster and PCA analyses expressed the behavior of relationships between different mathematical models and meteorological parameters in relation to the ETo determination.

A thermodynamics-based versatile evapotranspiration estimation method of minimum data requirement for water resources investigations

A recent, two-parameter version of the thermodynamically derived complementary relationship (CR) of evaporation has been tested on a monthly basis at 124 FLUXNET stations around the globe. Local, station-by-station calibration explained 91% (R2) of the variance in eddy-covariance (EC) obtained latent-heat fluxes with the same Nash-Sutcliffe efficiency (NSE) value. When the dimensionless Priestley-Taylor parameter (α) was expressed as a universal function (f) of the estimated wet-environment air temperature (Tw), station-by-station calibration of the single dimensionless parameter, b (accounting for moisture advection), yielded an R2 value of 87% and NSE of 86%. Global calibration (all stations at once) of the two-parameter CR version yielded R2 = 82% and NSE = 81%, while the single-parameter version produced R2 = 81% and NSE = 79%. With a representative value (between the locally calibrated mode and mean) of b set equal to two, the thus, calibration-free CR still maintained an R2 of 80% and NSE of 78%, which is significantly better than Morton’s calibration-free WREVAP model (i.e., 71% and 55%, respectively). The advantage of the current CR model is that it can be employed in a fully calibration-free mode, similar to WREVAP, yet with available EC measurements or water-balance derived latent-heat fluxes the single [b, when α = f(Tw) is chosen] or two parameters (α and b) of the model can be easily calibrated within the respective 1 – 1.32 and 1 – 10 intervals, for further improved performance.

Evaluation of machine learning models for prediction of daily reference evapotranspiration in semi-arid India

Reference evapotranspiration (ET ) is controlled by climatic factors; hence, its estimation provides an idea 0 about the atmospheric demand of water. Machine learning techniques like elastic net (ELNET), K-nearest neighbours (KNN), multivariate adaptive regression splines (MARS), partial least squares regression (PSLR), random forest (RF), support vector regression (SVR), XGBoost and cubist were employed to predict daily reference evapotranspiration based on daily weather parameters of twenty years. Penman-Monteith method was used as the reference method for ET estimation. All models performed well during calibration showing 0 higher coefficient of determination (R2) which ranged from 0.97 (for PLSR) to 1 (for cubist models). Mean absolute error during calibration ranges from 0.027 mm d-1 for cubist to 0.607 mm d-1 for ELNET. Cubist model (R2 = 1, MAE = 0.017 mm d-1, RMSE = 0.027 mm d-1) outperformed other models during the calibration. During validation, the coefficient of determination (R2) for the machine learning models varied from 0.819 to 1, RMSE varied from 0.06 to 0.60 mm d-1 and MAE varied from 0.031 to 0.38 mm d-1. Based on statistical parameters, best performance was observed for cubist model (R2 = 1, RMSE = 0.06 mm d-1, MAE = 0.031 mm d-1) among the studied machine learning models for the prediction of reference evapotranspiration. Hence, the cubist model may be used to estimate daily reference evapotranspiration for the studied region.

Sensitive analysis of meteorological data and selecting appropriate machine learning model for estimation of reference evapotranspiration

This study applies three methods, Gene Expression Programming (GEP), M5 tree (M5T) model and optimized Artificial Neural Network by Genetic Algorithm (ANN-GA) for estimation of reference evapotranspiration in Ahvaz and Dezful in the southwest of Iran. Comparison between results of the FAO Penman-Monteith (FPM) method and the mentioned three methods shows that ANN-GA with the Levenberg-Marquardt training method is the best method and the M5T model is the second appropriate method for estimation of reference evapotranspiration. In Ahvaz, R2 and RMSE of ANN-GA method are 0.996, 0.184 mm/day. For M5T method, these values are 0.997 and 0259 mm/day, and for GEP method, they are 0.979 and 0.521 mm/day. In Dezful, R2 and RMSE of ANN-GA method are 0.994, 0.235 mm/day. For M5T method, these values are 0.992 and 0265 mm/day, and for GEP method, they are 0.963 and 0.544 mm/day. In addition, sensitivity analysis shows that the maximum temperature is the most effective parameter, and the wind speed is second effective parameter. In Dezful, the effect of the maximum temperature is more than those of Ahvaz but the effect of wind speed is less than those of Ahvaz. Because Ahvaz is more flatter than Dezful (the movement of wind in Ahvaz is freer than those of Dezful). The third effective meteorological parameter is the average relative humidity in Ahvaz and the sunny hours in Dezful. The reason for this subject is the less distant of Ahvaz from the Persian Gulf (it is source of moisture).

Estimation of Urban Evapotranspiration at High Spatiotemporal Resolution and Considering Flux Footprints

Evapotranspiration (ET) estimations at high spatiotemporal resolutions in urban areas are crucial for extreme weather forecasting and water management. However, urban ET estimation remains a major challenge in current urban hydrology and regional climate research due to highly heterogeneous environments, human interference, and a lack of observations. In this study, an urban ET model, called the PT-Urban model, was proposed for half-hourly ET estimations at a 10 m resolution. The PT-Urban model was validated using observations from the Hotel Torni urban flux site during the 2018 growing season. The results showed that the PT-Urban model performed satisfactorily, with an R2 and root-mean-square error of 0.59 and 14.67 W m−2, respectively. Further analysis demonstrated that urban canopy heat storage and shading effects are essential for the half-hourly urban energy balance. Ignoring the shading effects led to a 38.7% urban ET overestimation. Modeling experiments further proved that flux footprint variations were critical for the accurate estimation of urban ET. The setting source areas either as an invariant 70% historical footprint or as a circle with a 1 km radius both resulted in poor performances. This study presents a practical method for the accurate estimation of urban ET with high spatiotemporal resolution and highlights the importance of real-time footprints in urban ET estimations.

Effect of different methods for estimation of evapotranspiration and water deficit in soil on the growth rates of conilon clonal coffee

The water deficit is considered one of the main problems in coffee culture around the world. As a way to get around this problem, producers all over the world adopt the practice of irrigation. However, in the most cultivated areas, a rational irrigation management is not carried out, resulting in excessive or insufficient applications. Based on this, the present work aimed to evaluate the effect of different methods of evapotranspiration estimative and water deficit levels on the growth rates of the clonal conilon coffee. The experiment was conducted in a greenhouse, in a 2 x 3 factorial scheme with the factor evapotranspiration in 2 levels (Penman-Monteith (PM) and Hargreaves & Samani (HS), and the factor water deficit in 3 levels (80%, 50% and 30% of Evapotranspiration of Reference (ETo)) in a completely randomized design. From the results obtained, it was observed that the evapotranspiration estimation method and the water deficit levels had an effect on the growth rates and free assimilation rate of the clonal conilon coffee tree, with the highest growth values in plants irrigated with 80% of ETo by the HS method, and higher assimilation values in plants irrigated with 80% of ETo by the PM method.

Development and Application of ETCalc, a Unique Online Tool for Estimation of Daily Evapotranspiration

ABSTRACT Various empirical methods that use meteorological data have been developed for estimating evapotranspiration. However, there are currently no online tools available for the estimation of daily evapotranspiration based on user-provided daily data. Here, we introduce ETCalc (https://etcalc.hydrotools.tech), a free, unique online tool that integrates eight methods (i.e. Penman-Monteith, Thornthwaite, Blaney – Criddle, Turc, Priestley – Taylor, Hargreaves, Jensen – Haise and Abtew) for estimation of daily potential evapotranspiration, reference evapotranspiration and, by employing user-defined crop (or cover) coefficients, daily actual evapotranspiration, based on user-provided daily meteorological data. ETCalc has been developed in response to the effort of the Canadian federal government to encourage easier and open access to science and is applicable to any area for which basic meteorological data are available and hence, its suitability is not restricted to particular geographical areas. Through a streamlined interface, ETCalc allows for uploading of user-provided data, tabular and graphical inspection of the input and output data, as well as export of the output data and of the associated metadata. The use of ETCalc is exemplified using 10-year daily meteorological data from Charlottetown, Prince Edward Island, Canada for comparing the output from each of the ETCalc methods and for the calculation of the precipitation deficit.

Development of an evapotranspiration estimation method for lettuce via mobile phones using machine vision: Proof of concept

This study aims to elucidate the process of crop evapotranspiration (ETc) and develop models to estimate ETc as a crucial value to efficiently schedule irrigation. Accordingly, we propose a new idea for non-destructive online acquisition of ETc data via mobile phones. The greenhouse environmental data were collected by sensors and uploaded to a mobile phone for the Penman-Monteith (PM) calculations of reference crop evapotranspiration (ET0). The canopy coverage (PGC) was accurately extracted from mobile phone images using the super-green algorithm, and four machine learning algorithms (XGBoost, CatBoost, SVR, and RF) were used to construct an online computing model for crop coefficient (Kc), ETc=Kc*ET0 was used to dynamically monitor the change of water requirement of different types of lettuce. The results showed that the correlation coefficients for the four monitoring networks were in the range of 0.74–0.93, which could be used to estimate the Kc at different stages. The integrated models showed high accuracy and similar performance. Compared with that of the SVR model, the root mean square error (RMSE), mean absolute error (MAE), and mean square error (MSE) of the three algorithms were reduced. Regarding the generalisation and computational speed of the proposed approach, the estimation accuracy of the XGBoost model was improved by 17.31%, and the convergence speed was also increased by 12.57% relative to other models on different types of lettuce. The XGBoost model can be recommended as the Kc model for estimation. The daily mean values of leaf lettuce and head lettuce as measured in ETc were 2.38 mm d−1 and 6.22 mm d−1, respectively. The average errors were 3.48% and 6.36%, respectively. We further explored models of the mathematical relationship between Kc and the plant height (H), plant width (D), and normalised vegetation index (NDVI), to establish a new approach for reasonable and scientific irrigation management.

Performance evaluation of evapotranspiration estimation methods in Sultanpur, Uttar Pradesh, India

Accurate estimation of evapotranspiration plays a vital role in judicious irrigation scheduling. Keeping this in view, a study was undertaken to compare the performance of five methods, viz. Penman method (PM), FAO Penman Monteith (FAO-PM), Priestley-Taylor (PT), FAO Radiation method (FAO-RM) and Hargreaves methods (HM) for Sultanpur District of Uttar Pradesh, India using the daily weather data acquired from automatic weather station during 2016-17. The performance evaluation of selected methods was carried out using linear regression and simple statistical analysis to suggest a substitute of FAO-PM for estimation of reference evapotranspiration using minimal climatic parameters available at regional scales. It was observed that the PM method performed the best and was in line with estimated ET0 by FAO-PM method with coefficient of determination (R2), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Modified Nash-Sutcliffe Coefficient (EMNS), Percent Bias (PBIAS) estimated as 0.94, 0.54, 0.40, 0.73, 3.74 for the year 2016 and 0.95, 0.44, 0.36, 0.73, 6.01 in 2017, respectively. Nonetheless, it was observed form this study that the PM method which requires four parameters, viz. daily solar radiation, maximum and minimum air temperature, and wind speed can be a substitute to FAO-PM that require more parameters for ET0 estimation in data scarce situations.

Comparing Reference Evapotranspiration Calculated in ETo Calculator (Ukraine) Mobile App with the Estimated by Standard FAO-Based Approach

Reference evapotranspiration (ETo) is a key agrometeorological index for rational irrigation management. The standard method for ETo estimation, proposed by the FAO, is based on a complicated Penman–Monteith equation and requires many meteorological inputs, making it difficult for practical use by farmers. At present, there are many alternative simplified approaches for ETo estimation; most of them are directed at cutting the number of required meteorological inputs for calculation. Among them, special attention should be paid to the various temperature-based methods of ETo assessment. One of the temperature-based models for ETo computation was realized in the free mobile app ETo Calculator (Ukraine). The app gives Ukrainian farmers an opportunity to assess ETo values on a daily or monthly scale using mean air temperature, obtained through free online meteorological forecasts and archive services, as the only input. The objective of the study was to test the app’s accuracy compared to FAO-based calculations in five key regions of Ukraine, each representing a particular climatic zone of the country. It was established that the app provides relatively good accuracy of ETo estimation even in raw (not adjusted to wind speed and relative air humidity) runs. The results of the statistical comparison with the FAO-calculated values on the daily scale are as follows: R2 within 0.82–0.87, RMSE within 0.74–0.81 mm, MAE within 0.60–0.70 mm, MAPE within 18.07–25.50%, depending on the region. The results of the statistical comparison with the FAO-calculated values on the monthly scale are: R2 within 0.88–0.95, RMSE within 0.50–0.72 mm, MAE within 0.33–0.59 mm, MAPE within 8.96–24.08% depending on the region. The ETo Calculator (Ukraine) is a good alternative to the complicated Penman–Monteith method and could be recommended for Ukrainian farmers to be used for irrigation management.

Evaluation of Reference Evapotranspiration Estimation Methods for the Assessment of Hydrological Impacts of Photovoltaic Power Plants in Mediterranean Climates

Large-scale photovoltaic (PV) power plants may affect the hydrological cycle in all its components. Among the various components, evapotranspiration is one of the most important. As a preliminary step for assessing the impacts of PV plants on evapotranspiration, in this study, we performed an evaluation study of methods for estimating reference evapotranspiration (ETo). FAO and ASCE recommend the Penman–Monteith (PM) method for the estimation of ETo when the data for all involved variables are available. However, this is often not the case, and different empirical methods to estimate ETo, requiring mainly temperature data, need to be used. This study aimed at assessing the performance of different temperature- and radiation-based empirical ETo estimation methods against the standardized PM ETo method in an experimental photovoltaic power plant in Piazza Armerina, Sicily, Italy, where a meteorological station and a set of sensors for soil moisture were installed. The meteorological data were obtained from the Lab from July 2019 to end of January 2022. By taking the ETo estimations from the PM method as a benchmark, the study assessed the performance of various empirical methods. In particular, the following methods were considered: Hargreaves and Samani (HS), Baier and Robertson (BR), Priestley and Taylor (PT), Makkink (MKK), Turc (TUR), Thornthwaite (THN), Blaney and Criddle (BG), Ritchie (RT), and Jensen and Haise (JH) methods, using several performance metrics. The result showed that the PT is the best method, with a Nash–Sutcliffe efficiency (NSE) of 0.91. The second method in order of performance is HS, which, however, performs significantly worse than PT (NSE = 0.51); nevertheless, this is the best among methods using only temperature data. BG, TUR, and THN underestimate ETo, while MKK, BG, RT, and JH showed overestimation of ETo against the PM ETo estimation method. The PT and HS methods are thus the most reliable in the studied site.

The Global LAnd Surface Satellite (GLASS) evapotranspiration product Version 5.0: Algorithm development and preliminary validation

• GLASS ET Algorithm v5.0 (DNN) outperforms GLASS ET Algorithm v4.0 (BMA). • GLASS ET Algorithm v5.0 is a robust method for global ET estimation. • DNN has enhanced the performance of ET estimation for extreme cases. An accurate estimation of spatially and temporally continuous global terrestrial evapotranspiration (ET) is essential in the assessment of surface energy, water and carbon cycles. The Global LAnd Surface Satellite (GLASS) ET product Version 4.0 (v4.0) based on the Bayesian model averaging (BMA) method was generated to estimate global terrestrial ET. However, certain uncertainty for the GLASS ET product v4.0 limits its application. In this study, we introduced the deep neural networks (DNN) merging framework to improve terrestrial ET estimation for GLASS ET product Version 5.0 (v5.0) generation by integrating five satellite-derived ET products [Moderate Resolution Imaging Spectroradiometer (MODIS) ET product (MOD16), Shuttleworth–Wallace dual-source ET product (SW), Priestley–Taylor-based ET product (PT-JPL), modified satellite-based Priestley–Taylor ET product (MS-PT) and simple hybrid ET product (SIM)]. We compared the performance of DNN method against other merging methods, including GLASS ET algorithm v4.0 (BMA), the gradient boosting regression tree (GBRT) method and the random forest (RF) method, based on 195 global eddy covariance (EC) flux towers covering observations from 2000 through 2015. Validations indicated that the DNN had the highest accuracy among four merging methods across different land cover types, yielding the highest average determination coefficients (R 2 , 0.62), root-mean-squared-error (RMSE, 24.1 W/m 2 ) and Kling–Gupta efficiency (KGE, 0.77) with a of 99% confidence interval. Compared with GLASS ET algorithm v4.0, the DNN improved on the R 2 by approximately 7% ( p < 0.01) and the KGE by 10%. Based on the DNN, we then generated 8-day GLASS ET product v5.0 globally with a 1 km spatial resolution from 2001 to 2015 driven by GLASS vegetation and surface net radiation (R n ) datasets and Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA2) datasets. This global terrestrial ET product provides a valuable dataset for monitoring regional and global water resources and environmental changes.

Application of a Machine Learning Algorithm in Generating an Evapotranspiration Data Product From Coupled Thermal Infrared and Microwave Satellite Observations.

Land surface evapotranspiration (ET) is one of the main energy sources for atmospheric dynamics and a critical component of the local, regional, and global water cycles. Consequently, accurate measurement or estimation of ET is one of the most active topics in hydro-climatology research. With massive and spatially distributed observational data sets of land surface properties and environmental conditions being collected from the ground, airborne or space-borne platforms daily over the past few decades, many research teams have started to use big data science to advance the ET estimation methods. The Geostationary satellite Evapotranspiration and Drought (GET-D) product system was developed at the National Oceanic and Atmospheric Administration (NOAA) in 2016 to generate daily ET and drought maps operationally. The primary inputs of the current GET-D system are the thermal infrared (TIR) observations from NOAA GOES satellite series. Because of the cloud contamination to the TIR observations, the spatial coverage of the daily GET-D ET product has been severely impacted. Based on the most recent advances, we have tested a machine learning algorithm to estimate all-weather land surface temperature (LST) from TIR and microwave (MW) combined satellite observations. With the regression tree machine learning approach, we can combine the high accuracy and high spatial resolution of GOES TIR data with the better spatial coverage of passive microwave observations and LST simulations from a land surface model (LSM). The regression tree model combines the three LST data sources for both clear and cloudy days, which enables the GET-D system to derive an all-weather ET product. This paper reports how the all-weather LST and ET are generated in the upgraded GET-D system and provides an evaluation of these LST and ET estimates with ground measurements. The results demonstrate that the regression tree machine learning method is feasible and effective for generating daily ET under all weather conditions with satisfactory accuracy from the big volume of satellite observations.

An atmospheric and soil thermal-based wheat crop coefficient method using additive crop growth models

Accurate crop evapotranspiration simulation is vital for reasonable irrigation planning. The two-step method for crop evapotranspiration estimation has been widely used for irrigation water demand management. However, in the two-step method, the crop coefficient (Kc) is typically taken as a constant in crop growth stages, which may result in a certain error in evapotranspiration estimation. It is of significance to estimate Kc accurately for effective water resources management. Based on three-year lysimeter experimental data collected from the Wudaogou Station for Hydrology and Water Resources, this study proposes a thermal-based Kc method for actual crop evapotranspiration estimation. The proposed thermal-based Kc method uses air temperature and soil heat flux as input variables, which could mainly characterise the thermal effect on crop growth from the atmosphere and soil. Results showed that the proposed method is effective in estimating Kc. The estimated Kc highly correlates with actual Kc values, which are derived from lysimeter observation, with correlation coefficient above 0.87 and Nash-Sutcliffe efficiency above 0.70. The proposed method could estimate Kc daily, which can be used to estimate crop evapotranspiration accurately.

Spatial-Temporal Interpolation of Reference Evapotranspiration for Pakistan

The unequal water distribution in the universe has resulted in more than 2 billion people living in water-stressed areas. Globally, withdrawals of water resources are still below the critical level. Pakistan is also affected by this serious problem. In the context of environmental evaluation, irrigation scheduling, and water resource management, evapotranspiration predictions are crucial. The evapotranspiration estimation methods may need to be evaluated daily or monthly to understand the climate change effects in the local areas. This paper investigates spatial-temporal interpolation of evapotranspiration data from 41 Pakistani meteorological stations. To estimate evapotranspiration, we have used the average time series data between 2006 and 2015 on temperature, relative humidity, wind speed, and solar radiation. We developed a new modified Hargreaves equation to estimate evapotranspiration and evaluate the performance of different models. We found that our modified Hargreaves model could perform better than the original Hargreaves model. To analyze trends in different seasons, such as rainy, dry, and annual, FAO-56 Penman-Monteith, Blaney-Criddle, and Hargreaves-Samani models are used. We used Ordinary Kriging for Functional Data to map and predict evapotranspiration spatially and temporally in various locations.