Estimation Of Daily Evapotranspiration Research Articles

Global Terrestrial Evapotranspiration Estimation from Visible Infrared Imaging Radiometer Suite (VIIRS) Data

It is a difficult undertaking to reliably estimate global terrestrial evapotranspiration (ET) using the Visible Infrared Imaging Radiometer Suite (VIIRS) at high spatial and temporal scales. We employ deep neural networks (DNN) to enhance the estimation of terrestrial ET on a global scale using satellite data. We accomplish this by merging five algorithms that are process-based and that make use of VIIRS data. These include the Shuttleworth–Wallace dual-source ET method (SW), the Priestley–Taylor-based ET algorithm (PT-JPL), the MOD16 ET product algorithm (MOD16), the modified satellite-based Priestley–Taylor ET algorithm (MS-PT), and the simple hybrid ET algorithm (SIM). We used 278 eddy covariance (EC) tower sites from 2012 to 2022 to validate the DNN approach, comparing it to Bayesian model averaging (BMA), gradient boosting regression tree (GBRT) and random forest (RF). The validation results demonstrate that the DNN significantly improves the accuracy of daily ET estimates when compared to three other merging methods, resulting in the highest average determination coefficients (R2, 0.71), RMSE (21.9 W/m2) and Kling–Gupta efficiency (KGE, 0.83). Utilizing the DNN, we generated a VIIRS ET product with a 500 m spatial resolution for the years 2012–2020. The DNN method serves as a foundational approach in the development of a sustained and comprehensive global terrestrial ET dataset. The basis for characterizing and analyzing global hydrological dynamics and carbon cycling is provided by this dataset.

Evaluating the ability of deep learning on actual daily evapotranspiration estimation over the heterogeneous surfaces

Evapotranspiration (ET) is one of the most critical components in hydrological processes and is of great importance to water resource management. Data-oriented deep learning (DL) has been increasingly utilized to forecast hydrological variables over recent years. In this study, the ability of the widely-used DL methods, including long short-term memory (LSTM), bi-directional LSTM (BiLSTM), deep neural network (DNN), and deep belief network (DBN), on the estimation of actual daily ET over the heterogeneous surfaces, was investigated using four groups of experiments. Firstly, the influence of the different input variables on the above DL-based ET model over the various land cover types was explored and analyzed. The results showed that the DL-based ET model can accurately estimate the actual daily ET over the heterogeneous surfaces using a few key conventional observations, i.e., net radiation (Rn), relative humidity (RH), air temperature (Ta), and wind speed (u), as well as soil water content (SWC). However, the performance of the DL-based ET model varied from the different combinations of input variables. SWC was crucial to the estimation of ET, with RMSD decreased from 0.94 to 0.76 mm d−1 when SWC was added to the DL model. Then, the comparison of the four DL-based ET model performances was done. The minor difference in ET estimates was caused by the algorithm differences between LSTM, BiLSTM, DNN, and DBN. Based on the above work, a unified DL-based ET model over the heterogeneous surfaces was developed. The unified DL-based model improved the applicability of DL in ET estimation although it underperformed the separate DL-based model. Finally, the comparison of the performance of the TSEB model and DL method was done. Evaluated with the results of heterogeneous surface, the DL-based model had a better accuracy with a MAPE of 16 – 50%; while the TSEB model had a larger MAPE of 20 – 55%. The results of this research suggest that the DL-based ET model is a promising alternative for the simulation of the daily ET over the heterogeneous surfaces.

Performance of potential evapotranspiration models across different climatic stations in New South Wales, Australia

Study regionNew South Wales, southeast Australia Study focusEstimating potential evapotranspiration (ETp) rates, detecting its temporal trends and analysing its interannual oscillation are critical for long-term assessment of water availability and regional drought. This study aimed to evaluate the comprehensive performance of 12 simplified models in characterising ETp against the benchmark Penman model across different climate sites in southeast Australia. This study used Taylor skill score (S), normalised root mean square error (nRMSE) and relative mean bias error (rMBE) to estimate models’ capability in estimating ETp rates. Then, this study adopted Mann-Kendall test and continuous wavelet transform (CWT) to test temporal trends and periodicity of ETp estimated by all models. New hydrology insights for the regionJensen-Haise model was capable to produce fair (nRMSE ≤ 30%) estimates of daily ETp across all stations. Models except Mak1 were generally able to produce reasonable estimates of ETp at larger time scale. In addition, we found that the 12 alternative ETp models generally agreed with the Penman model on the primary (9.6–12.4-year) and quasi (2.6–3.9-year) periods of ETp, but they did not show matchable ability in detecting ETp temporal trends. The comprehensive investigation on models’ performance will shed light on models’ selection in estimation of drought and hydrological cycle.

Application of a two source energy balance model coupled with satellite based soil moisture and thermal infrared data

Satellite retrievals of thermal infrared data from the Earth’s surface when combined with the surface energy balance equation are widely used to monitor land surface evapotranspiration (ET). Unfortunately thermal infrared based models using this approach have a systematic positive bias with ET during soil water stress conditions. The Two-source Energy Balance model using information on Soil Moisture as another key constraint (TSEB-SM) improves model performance and utility. Here, high resolution satellite soil moisture data from SMAP (Soil Moisture Active Passive) were used to provide soil water boundary conditions for the TSEB-SM model. This SMAP driven TSEB-SM approach was applied in the Heihe river basin in China across a wide range of soil moisture conditions using MODIS thermal infrared land surface temperature (LST) data. The ET estimates from TSEB-SM were validated using flux tower measurements over the main land cover types in the study area and compared with the results from the original TSEB model. The comparison showed that the TSEB-SM approach reduced the RMSE error in latent heat flux estimates by 5 % to 20 % at satellite overpass time, and TSEB-SM outperformed the original TSEB model in daily ET estimates (RMSE values of 1.0 mm/d versus 1.5 mm/d, respectively) across all validation sites. The magnitude of the TSEB-SM improvements were larger early and late in the growing season and in arid areas, consistent with soil moisture improving estimates of ET particularly under times of higher water stress. The use of SMAP satellite soil moisture data with the TSEB-SM model increased the reliability of ET estimation and provides an improved capacity for continuos monitoring of drought effects on the global water cycle.

Estimation of daily evapotranspiration in gully area scrub ecosystems on Loess Plateau of China based on multisource observation data

Evapotranspiration is an important process in the hydrologic cycle. In this study, different types of daily evapotranspiration were accurately estimated and comparatively analyzed for 2020–2021 in the gully area of Loess Plateau, China with multisource observation data. The daily actual evapotranspiration was directly observed by the eddy covariance system and then compared with the daily evapotranspiration simulated by HYDRUS-1D, and they showed reasonable agreement and similar variability characteristics it with a correlation coefficient of 0.71. The FAO-56 Penman-Monteith equation was used to calculate the reference evapotranspiration and compared with the pan evaporation in the study area. The reference evapotranspiration has a single-peak pattern and seasonal variation, similar to the actual evapotranspiration process during the annual period. The crop coefficient had seasonal-variation characteristics and remained 0.8 in the mid-growth period. The correlation analysis between evapotranspiration and environmental factors at different time scales suggested that energy was the dominant factor affecting evapotranspiration in the study area. The validation results of the complementary relationship of evaporation indicated a significant asymmetric complementary relationship in the study area. Both generalized nonlinear complementary relationship functions have good applicability, but the performance of the sigmoid function was slightly better than that of the polynomial function. The results provided a reference for local water resource management and agricultural production.

Remote sensing of daily evapotranspiration and gross primary productivity of four forest ecosystems in East Asia using satellite multi-channel passive microwave measurements

Satellite multi-channel passive microwave observations provide valuable information of canopy vertical structure and vegetation water content, while their potential in the simultaneous estimation of forest evapotranspiration (ET) and gross primary productivity (GPP) has not been explored. This study combined satellite multi-channel microwave Emissivity Difference Vegetation Indices (EDVIs) from 6.9 to 36.5 GHz with the Penman-Monteith (PM), Priestley-Taylor (PT) and light use efficiency (LUE) models to estimate daily ET and GPP over four forest ecosystems in East Asia. In-situ measurements of daily ET, GPP and meteorological variables from ChinaFLUX network (2003 to 2010) were used to calculate and validate the method. An exponential relationship was found between EDVIs and canopy conductance across the forests. By inverting EDVIs-based LUE model coupled with PM and PT models when canopy carbon uptake and water content reached a maximum in growing seasons, the maximum LUE (EDVI-LUEmax) was derived and found to agree with other independent studies. Site validations showed that the EDVI-based method performed a good simultaneous estimation of daily ET and GPP at a tropical forest and three sub-tropical and temperate forests (R2 of 0.64 to 0.94 and RMSE of −11.7% to 19.6% for ET, R2 of 0.39 to 0.73 and RMSE of 31.2% to 50.4% for GPP). The modeled relationship between GPP and ET also agreed that from in-situ measurements. The results were also found to correlate with microwave frequencies. In-situ ET and GPP showed a stronger correlation with EDVI at 18.7 and 36.5 GHz than that at 10.7 and 18.7 GHz across the forests. EDVI-LUEmax of a temperate mixed forest increased as EDVI incorporated lower frequency microwave emissions from deeper canopy layers, which reduced the negative bias in GPP estimation. Overall, this study demonstrated the potential of multi-channel microwave measurements in estimating ET and GPP with the potential effects of vegetation vertical structure.

Coupling physical constraints with machine learning for satellite-derived evapotranspiration of the Tibetan Plateau

More accurate and process-based satellite evapotranspiration (ET) estimation for the Tibetan Plateau (TP)—the Third Pole of the world—have long been of major interest in hydrometeorology. Combining recent advances in satellite-based ET mechanistic algorithms and data-oriented methods allows ET hybrid modeling by coupling physical constraints with machine learning (ML). Specifically, we developed two hybrid models, a surface conductance-based ML model (ML-Gs) and a soil evaporation-based ML model (ML-Es), to estimate regional ET on the TP. These hybrid models have biophysical framework, under which one of the parameters or components is modeled using ML. Hybrid models make ML complementary to the process-based ET framework, which to find an optimal junction between well physical mechanism and high model performance. The daily ET estimates were evaluated at 28 eddy covariance flux tower sites, as well as by comparison with two process-based ET algorithms (a Penman–Monteith-based ET-PM algorithm and a Priestley-Taylor-based ET-PT algorithm) and a data-oriented pure ML method. The hybrid models decreased the root-mean-square-error (RMSE) of two physical algorithms (1.11 mm/day for ET-PM, 1.09 mm/day for ET-PT) to 0.50 mm/day, and increased the Kling-Gupta efficiency (KGE) (0.35 for ET-PM, 0.36 for ET-PT) to 0.92. Our hybrid models also showed improved performance (KGE of 0.65) than pure ML (KGE of 0.62) at data-sparse regions as well as for the responses to extreme weather events. It indicates that our approach does not only boost the ET simulation accuracy, but also improve the physical understanding of ML-based ET estimation. More importantly, ML-Es focuses on the ET components on the TP and is more well-defined than ML-Gs. An innovation of our approach is that for data-sparse regions and extreme cases, the more robust physical mechanism was coupled, the better generalization performance of hybrid model could achieve. The spatiotemporal ET patterns based on our hybrid models were consistent with the variations in local climatic regions and could provide critical information on the understanding of hydrological processes under the global and regional climate changes.

Parameter variability across different timescales in the energy balance-based model and its effect on evapotranspiration estimation

Evapotranspiration is a key consideration for addressing a number of scientific and engineering issues. There are considerable errors in current evapotranspiration models due to the high uncertainty in model parameters. Considering that evapotranspiration models maintain the same mathematical form when run on different timescales, we argue that the uncertainty in model parameters can be reduced by considering the parameter variability across different timescales. Here, the four key parameters in the energy balance-based evapotranspiration model, including aerodynamic roughness length, thermodynamic roughness length, surface conductance, and energy balance ratio, are retrieved and evaluated on instantaneous and daily timescales based on the observations from 113 sites in the FLUXNET2015 dataset. Then data-driven instantaneous and daily parameter models are built to estimate evapotranspiration. The results show that strong multi-timescale variability occurs in all four parameters. The coefficients of variation of the four instantaneous parameters range from 0.32 to 1.70. The links of parameters on different timescales are weak. The correlation coefficients of the daily mean value of instantaneous parameter values and daily parameter values vary from 0.44 to 0.83. By considering the multi-timescale variability of the parameters, the accuracy of evapotranspiration estimation can be largely improved, with RMSE of the instantaneous and daily evapotranspiration estimation decreasing from 35.76 to 9.52 W m−2 and from 12.01 to 3.01 W m−2, respectively. We also find that the parameter models perform well on their inherent timescales but degrade significantly when transferring to other timescales. This study proves the necessity of defining parameter variability across different timescales in evapotranspiration models and provides new insight into the model parameters.

A comparative study on daily evapotranspiration estimation by using various artificial intelligence techniques and traditional regression calculations.

Evapotranspiration is an important parameter to be considered in hydrology. In the design of water structures, accurate estimation of the amount of evapotranspiration allows for safer designs. Thus, maximum efficiency can be obtained from the structure. In order to accurately estimate evapotranspiration, the parameters affecting evapotranspiration should be well known. There are many factors that affect evapotranspiration. Some of these can be listed as temperature, humidity in the atmosphere, wind speed, pressure and water depth. In this study, models were created for the estimation of the daily evapotranspiration amount by using the simple membership functions and fuzzy rules generation technique (fuzzy-SMRGT), multivariate regression (MR), artificial neural networks (ANNs), adaptive neuro-fuzzy inference system (ANFIS) and support vector regression (SMOReg) methods. Model results were compared with each other and traditional regression calculations. The ET amount was calculated empirically using the Penman-Monteith (PM) method which was taken as a reference equation. In the created models, daily air temperature (T), wind speed (WS), solar radiation (SR), relative humidity (H) and evapotranspiration (ET) data were obtained from the station near Lake Lewisville (Texas, USA). The coefficient of determination (R2), root mean square error (RMSE) and average percentage error (APE) were used to compare the model results. According to the performance criteria, the best model was obtained by Q-MR (quadratic-MR), ANFIS and ANN methods. The R2, RMSE, APE values of the best models were 0,991, 0,213, 18,881% for Q-MR; 0,996; 0,103; 4,340% for ANFIS and 0,998; 0,075; 3,361% for ANN, respectively. The Q-MR, ANFIS and ANN models had slightly better performance than the MLR, P-MR and SMOReg models.

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.

Estimation of evapotranspiration using all-weather land surface temperature and variational trends with warming temperatures for the River Source Region in Southwest China

• All-weather LST is used to overcome satellite TIR LST missing in cloudy conditions. • Estimation scheme for all-weather evapotranspiration (AWET) is proposed. • Response of AWET to the warming temperatures exhibits a spatiotemporal heterogeneity. Land surface temperature (LST) derived from satellite thermal infrared (TIR) remote sensing is widely used in estimating land surface evapotranspiration (ET) through energy balance theory. However, as satellite TIR remote sensing is frequently affected by clouds, the derived LST with spatial missing makes it impossible to estimate seamless ET for large areas. In this study, based on the all-weather LST (AWLST) generated through the synergistic use of TIR and passive microwave (PMW) remote sensing, we propose an estimation scheme of all-weather ET (AWET) for the River Source Region of Southwest China (RSR-SC) with complex environmental characteristics of frequent clouds and fog. Specifically, the parameters of the Surface Energy Balance System (SEBS), which have high impact on the model results, are first accurately determined by a global sensitivity analysis. Second, the parametric calculation schemes of turbulent exchange, such as surface roughness heights for momentum ( z 0m ) and heat ( z 0h ) transfer, and soil heat flux ( G 0 ) of the SEBS model are refined. Third, the daily ET temporal upscaling method is developed. Then a long-term AWET product for the RSR-SC is generated. Comparison against ground measurements from 12 eddy covariance (EC) sites indicates a good accuracy of the AWET product: the mean absolute percent error (MAPE) and root mean square error (RMSE) of the daily ET estimates are 36 % and 0.88 mm/d, respectively. From aspects of temporal variation of daily ET, monthly spatiotemporal distribution patterns, interannual variation of different land covers and spatial distribution comparison with other ET products, the generated AWET exhibits to be realistic, reflecting relatively subtle variations. Further investigation indicates that the response of surface ET to the warming temperatures exhibits the spatial heterogeneity of the change trend. This study highlights the importance of estimating daily AWET at a MODIS-like scale using AWLST data to significantly benefit water resource monitoring and conduct runoff prediction over large areas and watershed scales.

Managing spatial irrigation using remote-sensing-based evapotranspiration and soil water adaptive control model

Irrigation has traditionally been managed as uniform applications where an entire field receives the same depth of water. Motivation to improve current irrigation practices has led to different approaches utilizing remotely-sensed images to inform variable rate irrigation management. This study conducted in 2019 and 2020 implemented the Spatial EvapoTranspiration Modeling Interface (SETMI), a remote-sensing-based evapotranspiration (ET) and water balance model, for managing variable rate irrigation of a maize and soybean field. This model tracked soil water content through the estimation of daily ET and tracking of various water fluxes entering and leaving a field. SETMI was used in two different irrigation treatments informed using Planet satellite (SETMI-SAT) and unmanned aerial system (UAS, SETMI-UAS) remotely-sensed images. A uniform irrigation approach managed by a professional crop consultant and a non-irrigated approach were used as the baseline in comparing irrigation management approaches. The irrigation treatments were evaluated on dry grain yield, gross irrigation, actual ET, deep percolation, change in soil water content, and water productivity. The uniform irrigation approach managed by the crop consultant applied the highest irrigation in 2019 and 2020 for maize (2019: 155 mm, 2020: 213 mm) and soybean (2019: 124 mm; 2020: 183 mm) while the SETMI irrigation treatments applied less irrigation for maize (2019: 131, 132 mm; 2020: 154, 140 mm) and soybean (2019: 116, 94 mm; 2020: 154, 175 mm). Maize yield was highest for the uniform irrigation approach in 2019 (14.9 Mg ha−1) and 2020 (13.3 Mg ha−1). The highest soybean yield was produced by the SETMI-SAT irrigation management approach in 2019 (5.0 Mg ha−1) and 2020 (4.8 Mg ha−1). Significant differences (p-value < 0.05) in applied irrigation between the uniform and SETMI irrigation management approaches were observed while there were no significant differences in dry grain yield for both maize and soybean in 2019 and 2020. At least one of the SETMI irrigation treatments produced higher crop, irrigation, and ET water productivity values in comparison to those produced by the uniform irrigation treatment for all crop-years. A post-season analysis was completed using the SETMI-UAS and SETMI-SAT treatments to evaluate the accuracy of estimated rootzone soil water depletion provided by SETMI. Rootzone depletion calculated from neutron probe volumetric soil water content measurements were compared to the modeled depletion from the SETMI-UAS and SETMI-SAT treatments. The 2020 modeled and measured depletion comparison produced better agreement resulting in a root mean squared error (RMSE) < 17 mm compared to 2019 (RMSE < 27 mm). The VRI center pivot malfunctioned during the 2019 season which caused unresolved discrepancies between actually applied irrigation and what the system was programmed to apply. The VRI system was fixed before the 2020 season.

Sensitivity of METRIC-based tree crop evapotranspiration estimation to meteorology, land surface parameters and domain size

Evapotranspiration (ET), a key component of crop water balance, is essential for growers to manage agricultural irrigation. METRIC model is increasingly used to estimate and map ET with satellite observations, especially at daily and field scales. However, the analysis of its uncertainties on tree crop ET estimation is very limited. We used METRIC model to estimate daily ET over the almond and pistachio orchards throughout the entire growing season in the San Joaquin Valley, California, and evaluated the impacts of variations in meteorological variables, land surface parameters and different study sizes on ET estimation. The results showed that the METRIC-based daily ET estimates of the almond and pistachio trees were most sensitive to net radiation (Rn), followed by instantaneous reference ET (ETo), land surface temperature (LST), and near-surface temperature difference (dT). When a 5% error was added to Rn, the uncertainty in daily ET estimates was up to 8.44 ± 1.24%, corresponding to a mean increase of 0.45 ± 0.02 mm/day throughout the entire growing season. A 5% or 0.5 K error in instantaneous ETo, LST, and dT also caused changes in daily ET estimates by 4.15 ± 3.75%, 2.97 ± 0.14%, and 2.31 ± 1.20%, respectively. ET estimates with METRIC were relatively much less sensitive to other variables. The results from different study sizes indicated that daily ET estimates in Size-2 were 0.33 ± 0.08 mm/day overestimation compared with those from Size-1 in the almond site. Similarly, a mean overestimation of 0.30 ± 0.14 mm/day was found between Size-1 and Size-3 in the pistachio sites throughout the entire growing season. This study demonstrates the potential to further improve the accuracy of remote sensing based ET estimation for complex tree crops with METRIC model, by optimizing key parameters and choosing a proper domain size for study area.

Improved Daily Evapotranspiration Estimation Using Remotely Sensed Data in a Data Fusion System

Evapotranspiration (ET) represents crop water use and is a key indicator of crop health. Accurate estimation of ET is critical for agricultural irrigation and water resource management. ET retrieval using energy balance methods with remotely sensed thermal infrared data as the key input has been widely applied for irrigation scheduling, yield prediction, drought monitoring and so on. However, limitations on the spatial and temporal resolution of available thermal satellite data combined with the effects of cloud contamination constrain the amount of detail that a single satellite can provide. Fusing satellite data from different satellites with varying spatial and temporal resolutions can provide a more continuous estimation of daily ET at field scale. In this study, we applied an ET fusion modeling system, which uses a surface energy balance model to retrieve ET using both Landsat and Moderate Resolution Imaging Spectroradiometer (MODIS) data and then fuses the Landsat and MODIS ET retrieval timeseries using the Spatial-Temporal Adaptive Reflectance Fusion Model (STARFM). In this paper, we compared different STARFM ET fusion implementation strategies over various crop lands in the central California. In particular, the use of single versus two Landsat-MODIS pair images to constrain the fusion is explored in cases of rapidly changing crop conditions, as in frequently harvested alfalfa fields, as well as an improved dual-pair method. The daily 30 m ET retrievals are evaluated with flux tower observations and analyzed based on land cover type. This study demonstrates improvement using the new dual-pair STARFM method compared with the standard one-pair STARFM method in estimating daily field scale ET for all the major crop types in the study area.

Evapotranspiration uncertainty at micrometeorological scales: the impact of the eddy covariance energy imbalance and correction methods

Under ideal conditions, evapotranspiration (ET) fluxes derived through the eddy covariance (EC) technique are considered a direct measure of actual ET. Eddy covariance flux measurements provide estimates at a temporal frequency that allows examining sub-daily, daily, and seasonal scale processes and relationships between different surface fluxes. The Grape Remote Sensing Atmospheric Profile and Evapotranspiration eXperiment (GRAPEX) project has collected micrometeorological and biophysical data to ground-truth new remote sensing tools for fine-tuning vineyard irrigation management across numerous sites since 2013. This rich dataset allows us to quantify the impact of different approaches to estimate daily ET fluxes, while accounting for energy imbalance. This imbalance results from the lack of agreement between the total available energy and turbulent fluxes derived by the EC technique. We found that different approaches to deal with this energy imbalance can lead to uncertainty in daily ET estimates of up to 50%. Over the growing season, this uncertainty can lead to considerable biases in crop water use estimates, which in some cases were equivalent to ~ 1/3rd of the total growing season applied irrigation We analyzed ET uncertainty relative to atmospheric meteorological, stability, and advective conditions, and highlight the importance of recognizing limitations of micrometeorological observational techniques, considered state of the art, to quantify ET for model validation and field-scale monitoring. This study provides a framework to quantify daily ET estimates’ uncertainty and expected reliability when using the eddy covariance technique for ground-truthing or model validation purposes.

Modeling Reference Crop Evapotranspiration Using Support Vector Machine (SVM) and Extreme Learning Machine (ELM) in Region IV-A, Philippines

The need for accurate estimates of reference crop evapotranspiration (ETo) is important in irrigation planning and design, irrigation scheduling, reservoir management among other applications. ETo can be accurately determined using the internationally accepted FAO Penman–Monteith (FAO-56 PM) equation. However, this requires numerous observed data, including solar radiation, air temperature, relative humidity, and wind speed, which in most cases are unavailable, particularly in developing countries such as the Philippines. This study developed models based on Support Vector Machines (SVMs) and Extreme Learning Machines (ELMs) for the estimation of daily ETo using different input combinations of meteorological data in Region IV-A, Philippines. The performance of machine learning models was compared with the different established alternative empirical models for ETo. The results show that the SVM and ELM models, with at least Tmax, Tmin, and Rs as inputs, provide the best daily ETo estimates. The accuracy of machine learning models was also found to be superior compared to the empirical models given with same input requirements. In general, SVM and ELM models showed similar modeling performance, although the former showed lower run time than the latter.

Operational daily evapotranspiration mapping at field scale based on SSEBop model and spatiotemporal fusion of multi-source remote sensing data.

Accurate understanding of daily evapotranspiration (ET) at field scale is of great significance for agricultural water resources management. The operational simplified surface energy balance (SSEBop) model has been applied to estimate field scale ET with Landsat satellite imagery. However, there is still uncertainty in the ET time reconstruction for cloudy days based on limited clear days’ Landsat ET fraction (ETf) computed by SSEBop. The Moderate Resolution Imaging Spectroradiometer (MODIS) remote sensing data can provide daily surface observation over clear-sky areas. This paper presented an enhanced gap-filling scheme for the SSEBop ET model, which improved the temporal resolution of Landsat ETf through the spatio-temporal fusion with SSEBop MODIS ETf on clear days and increased the time reconstruction accuracy of field-scale ET. The results were validated with the eddy covariance (EC) measurements over cropland in northwestern China. It indicated that the improved scheme performed better than the original SSEBop Landsat approach in daily ET estimation, with higher Nash-Sutcliffe efficiency (NSE, 0.75 vs. 0.70), lower root mean square error (RMSE, 0.95 mm·d-1 vs. 1.05 mm·d-1), and percent bias (PBias, 16.5% vs. 25.0%). This fusion method reduced the proportion of deviation (13.3% vs. 25.5%) in the total errors and made the random error the main proportion, which can be reduced over time and space in regional ET estimation. It also evidently improved the underestimation of crop ET by the SSEBop Landsat scheme during irrigation before sowing and could more accurately describe the synergistic changes of soil moisture and cropland ET. The proposed MODIS and Landsat ETf fusion can significantly improve the accuracy of SSEBop in estimating field-scale ET.

Assessing uncertainties in estimating surface energy fluxes from remote sensing over natural grasslands in Brazil

Evapotranspiration (ET) is one of the main fluxes in the global water cycle. As the Brazilian Pampa biome carries a rich biodiversity, accurate information on the ET dynamics is essential to support its proper monitoring and establish conservation strategies. In this context, we assessed an operational methodology based on the Simplified Surface Energy Balance Index (S-SEBI) model to estimate energy fluxes over the natural grasslands of the Pampa between 2014 and 2019. The S-SEBI is an ET model that requires a minimum of meteorological inputs and has demonstrated reasonable accuracy worldwide. Therefore, we investigated the model performance considering radiation data from both ERA5 reanalysis and Eddy Covariance measurements from a flux tower. Furthermore, comparisons from satellite-based estimates with in situ measurements were performed with and without energy balance closure (EBC). Results indicated that the meteorological inputs have low sensitivity on daily ET estimates from the S-SEBI model. In contrast, the instantaneous energy balance components are more affected. The strong seasonality impacts the evaporative fraction, which is more evident in late summer and autumn and may compromise the performance of the model in the biome. The effects in the daily ET are lower when in situ data without EBC are considered as ground truth. However, they are less correlated with the remote sensing-based estimates. These insights are useful to monitor water and energy fluxes from local to regional scale and provide the opportunity to capture ET trends over the natural grasslands of the Pampa.

Estimation of Daily Evapotranspiration in Gully Area on Loess Plateau of China Based on Multisource Observation Data

Estimation of Daily Evapotranspiration in Gully Area on Loess Plateau of China Based on Multisource Observation Data

Estimating Evapotranspiration of Mediterranean Oak Savanna at Multiple Temporal and Spatial Resolutions. Implications for Water Resources Management

Mediterranean oak savanna is composed of a mixture of scattered oak trees, crops, pasture, and shrubs. It is the most widespread agroforestry landscape in Europe, and its conservation faces multiple threats including water scarcity, which has been exacerbated by global warming and greater climate variability. Evapotranspiration (ET) can be used as a proxy of the vegetation water status and response to water shortage conditions, providing relevant information about the ecosystem stability and its hydrological dynamics. This study evaluates a framework to estimate ET at multiple spatial and temporal scales and applies it to the monitoring of the oak savanna vegetation water consumption for the years 2013–2015. We used a remote sensing-based energy balance model (ALEXI/DisALEXI approach), and the STARFM data fusion technique to provide daily ET estimates at 30 m resolution. The results showed that modeled energy balance components compared well to ground measurements collected by an eddy covariance system, with root mean square error (RMSE) values ranging between 0.60 and 2.18 MJ m−2 d−1, depending on the sensor dataset (MODIS or Landsat) and the flux. The daily 30 m ET series generated by STARFM presented an RMSE value of 0.67 mm d−1, which yielded a slight improvement compared to using MODIS resolution or more simple interpolation approaches with Landsat. However, the major advantage of the high spatio-temporal resolution was found in the analysis of ET dynamics over different vegetation patches that shape the landscape structure and create different microclimates. Fine-scale ET maps (30 m, daily) provide key information difficult to detect at a coarser spatial resolution over heterogeneous landscapes and may assist management decisions at the field and farm scale.