Estimation Of Evapotranspiration Research Articles

Evaluating data-driven and hybrid modeling of terrestrial actual evapotranspiration based on an automatic machine learning approach

The performances of physics-based, data-driven, and hybrid models for estimating terrestrial actual evapotranspiration (ETa) is currently under debate, which requires thorough evaluations of those models particularly with recent developments in automatic machine learning (AML) techniques. In this study, six AML-based models were first constructed using the H2O-AML platform, from which an optimal (AML-OP) model was selected for estimating daily ETa at ecosystem scales. In addition, hybrid models were developed by combining the six AML models with surface conductance (Gs) inverted from the Penman-Monteith equation and an optimal (PM-OP) model was also selected. With 15 predictor variables as model inputs that were compiled from various data sources, the performances of those models for estimating daily ETa were evaluated using observed data from the FLUXNET2015 dataset. The results revealed that no models showed consistently low noise levels across different ecosystem types, making it necessary to use AML techniques for selecting ecosystem-specific models. Interestingly, the AML-OP models (root mean square error (RMSE) and symmetric mean absolute percentage error (SMAPE) were 0.16–0.31 mm d-1 and 9 %–36 % respectively) showed slightly better predictive results than the PM-OP models (RMSE and SMAPE were 0.23–0.36 mm d-1 and 15 %–68 % respectively), likely owing to model parameter uncertainties and tight constraints of physical models on application condition. Secondly, as ETa nonlinearly responds to environmental variables, model predictability under extreme weather (drought and heatwave) conditions was examined. The results showed that the prediction of the AML-OP and PM-OP models expectedly worsened (RMSE and SMAPE increased by 0.06–0.77 mm d-1 and −19 % to 79 %, respectively); however, the AML-OP model still outperformed the PM-OP model in most ecosystems, further underscoring the need to understand ETa regulation mechanisms under varying climatic conditions. Finally, the PM-OP models developed here provided better daily ETa estimates compared to other recently proposed hybrid models (RMSE reduced by 0.98–1.80 mm d-1). Both models can be better applied to wetlands that have been less frequently evaluated previously (RMSE reduction of 0.22 mm d-1 and 0.18 mm d-1 for the AML-OP and PM-OP models).

Comprehensive accuracy assessment of long-term geostationary SEVIRI-MSG evapotranspiration estimates across Europe

This study quantifies the accuracy of evapotranspiration (ET) estimates from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) geostationary sensor onboard the Meteosat Second Generation (MSG) satellites, along seven key dimensions, i.e., diurnal cycle, daily, intra-annual, inter-annual, ecosystem, climate zone, and products intercomparison. In situ measurements were collected at 54 eddy covariance (EC) sites to evaluate the accuracy of SEVIRI actual ET products (diurnal and daily SEVIRI-ETa) as well as reference ET (daily SEVIRI-ET0) covering the period from 2004 to 2018 across Europe. SEVIRI-ETa is produced by the Tiled ECMWF Surface Scheme of Exchange processes at the Land surface (TESSEL) model, while SEVIRI-ET0 is estimated by a combination of a thermodynamically-based and an atmospheric boundary layer model. This evaluation is further separated according to the land cover heterogeneity of the SEVIRI pixels across all 54 EC sites, using MODIS land cover data. The Root Mean Squared Error (RMSE), along with the Kling-Gupta efficiency (KGE) and their respective decompositions, were employed to quantify the errors.For diurnal SEVIRI-ETa estimates, we found that the KGE (RMSE [mm hour−1]) varied between −1.6 (0.04) to 0.8 (0.14), with a median value of 0.26 (0.07). Higher accuracy for diurnal SEVIRI-ETa was obtained in the summer and during the mid-day time. For daily SEVIRI-ETa, the KGE (RMSE [mm day−1]) varied between −0.88 (0.43) to 0.93 (1.79), with a median value of 0.6 (0.77) and for daily SEVIRI-ET0 the KGE (RMSE [mm day−1]) varied between 0.51 (0.40) to 0.94 (1.50), with a median value of 0.77 (0.57). For daily SEVIRI-ETa, intra-annual accuracy was low from January to March, increased in the mid-year, and then began to decline from November to December. Although accuracy remained relatively stable during the middle of the year, it varied considerably in the winter period. In the inter-annual dimension, the mid-year positive KGE values and distributions changed over time from 2004 to 2018. In spatial dimensions, the highest accuracy was in peat and grassland ecosystems, and the lowest in cropland ecosystem, with similar patterns observed in the boreal snow fully humid warm summer and warm temperate fully humid hot summer climate zones. Regarding SEVIRI-ET0 results, similar to SEVIRI-ETa, intra-annual accuracy was low in the first quarter of the year and the last one but high in the mid-year. In the inter-annual dimension, unlike SEVIRI-ETa, almost an identical pattern was observed for the mid-year positive KGE values, demonstrating only a slight change in SEVIRI-ET0 accuracy during 2004–2018. However, the highest accuracy was found in crop ecosystem, while the lowest was in forest ecosystem, reflecting similar trends in the warm temperate fully humid hot summer and warm temperate summer dry hot summer climate zones. The observed range of median RMSE changed between 0.4 and 1.5 mm day−1, also suggests a reasonable accuracy for SEVIRI-ET estimates in all spatial domains.Our results showed that the main trends in the accuracies (median KGEs) of SEVIRI-ET (both ETa and ET0) remained similar in separated homogeneous and heterogeneous sites and were comparable to combined sites among the dimensions. Through error decomposition, we discerned that SEVIRI-ET estimates performed particularly well in explaining inter-annual and spatial variabilities. Furthermore, the intercomparison of ET products revealed that SEVIRI satellite-derived ETa exhibited the strongest correlation with in situ ET measurements across all ecosystem types and climate zones, outperforming other products (such as MODIS, PML, GLEAM, and BESS). The ET estimates from other products exhibited lower standard deviation errors and were in closer agreement with the in situ measurements. This study provides the first comprehensive evaluation of the accuracy of SEVIRI diurnal and daily ET products across Europe, which may serve as a stimulus for further optimized selection of these products by potential users for various applications.

A review of recent advances and future prospects in calculation of reference evapotranspiration in Bangladesh using soft computing models

Evapotranspiration (ETo) is a complex and non-linear hydrological process with a significant impact on efficient water resource planning and long-term management. The Penman-Monteith (PM) equation method, developed by the Food and Agriculture Organization of the United Nations (FAO), represents an advancement over earlier approaches for estimating ETo. Eto though reliable, faces limitations due to the requirement for climatological data not always available at specific locations. To address this, researchers have explored soft computing (SC) models as alternatives to conventional methods, known for their exceptional accuracy across disciplines. This critical review aims to enhance understanding of cutting-edge SC frameworks for ETo estimation, highlighting advancements in evolutionary models, hybrid and ensemble approaches, and optimization strategies. Recent applications of SC in various climatic zones in Bangladesh are evaluated, with the order of preference being ANFIS > Bi-LSTM > RT > DENFIS > SVR-PSOGWO > PSO–HFS due to their consistently high accuracy (RMSE and R2). This review introduces a benchmark for incorporating evolutionary computation algorithms (EC) into ETo modeling. Each subsection addresses the strengths and weaknesses of known SC models, offering valuable insights. The review serves as a valuable resource for experienced water resource engineers and hydrologists, both domestically and internationally, providing comprehensive SC modeling studies for ETo forecasting. Furthermore, it provides an improved water resources monitoring and management plans.

Estimation of Daily Actual Evapotranspiration of Tea Plantations Using Ensemble Machine Learning Algorithms and Six Available Scenarios of Meteorological Data

The tea plant (Camellia sinensis), as a major, global cash crop providing beverages, is facing major challenges from droughts and water shortages due to climate change. The accurate estimation of the actual evapotranspiration (ETa) of tea plants is essential for improving the water management and crop health of tea plantations. However, an accurate quantification of tea plantations’ ETa is lacking due to the complex and non-linear process that is difficult to measure and estimate accurately. Ensemble learning (EL) is a promising potential algorithm for accurate evapotranspiration prediction, which solves this complexity through the new field of machine learning. In this study, we investigated the potential of three EL algorithms—random forest (RF), bagging, and adaptive boosting (Ad)—for predicting the daily ETa of tea plants, which were then compared with the commonly used k-nearest neighbor (KNN), support vector machine (SVM), and multilayer perceptron (MLP) algorithms, and the experimental model. We used 36 estimation models with six scenarios from available meteorological and evapotranspiration data collected from tea plantations over a period of 12 years (2010–2021). The results show that the combination of Rn (net radiation), Tmean (mean air temperature), and RH (relative humidity) achieved reasonable precision in assessing the daily ETa of tea plantations in the absence of climatic datasets. Compared with other advanced models, the RF model demonstrated superior performance (root mean square error (RMSE): 0.41–0.56 mm day−1, mean absolute error (MAE): 0.32–0.42 mm day−1, R2: 0.84–0.91) in predicting the daily ETa of tea plantations, except in Scenario 6, followed by the bagging, SVM, KNN, Ad, and MLP algorithms. In addition, the RF and bagging models exhibited the highest steadiness with low RMSE values increasing (−15.3~+18.5%) in the validation phase over the testing phase. Considering the high prediction accuracy and stability of the studied models, the RF and bagging models can be recommended for estimating the daily ETa estimation of tea plantations. The importance analysis from the studied models demonstrated that the Rn and Tmean are the most critical influential variables that affect the observed and predicted daily ETa dynamics of tea plantations.

Machine-learned actual evapotranspiration for an irrigated pecan orchard in Northwest Mexico

Accurate field-scale estimates of the actual evapotranspiration (ETact) based on readily available input data is indispensable to optimize irrigation in (semi-)arid regions. In this study, at an irrigated pecan orchard in Northwest Mexico we explored the potential for three different machine learning algorithms to improve upon the 30-min ETact estimates provided by the FAO Penman-Monteith method (FAO-PM) when trained and tested on multi-year eddy-covariance measurements. We found that all three machine learning algorithms performed well (R2>0.9), with the Random Forest (RF) giving the best overall performance. RF consistently outperformed FAO-PM on sub-daily scales in the growing season, showing its potential to further improve sub-daily ETact estimates. In addition, RF 1) shows a clear improvement (R2 increase up to 0.168) compared to a typical FAO-PM estimate based on the procedures described in FAO report 56, and 2) mostly performed similar as FAO-PM with a highly accurate locally fitted crop factor on the seasonal scale. The latter highlights its ability to infer the correct seasonal trend through input variables that reflect the underlying physical processes, giving potential for application at other similar fields in the region with limited local tuning. To assess whether incorporating existing (semi-)physical parameterizations could improve the performance, we additionally combined our machine learning models with FAO-PM both with and without locally fitted crop factor. Overall, we found this resulted in a modest improvement in performance in most cases (R2 increase up to 0.01). Notably, incorporating the locally calibrated crop factor and FAO-PM each reduced the errors made by the RF in the second half of the growing season. This implies that it is still useful to incorporate FAO-PM and a crop factor in the machine learning model, especially when the crop factor can be determined with high accuracy.

Evapotranspiration Partitioning and Estimation Based on Crop Coefficients of Winter Wheat Cropland in the Guanzhong Plain, China

Accurate estimation and effective portioning of actual evapotranspiration (ETa) into soil evaporation (E) and plant transpiration (T) are important for increasing water use efficiency (WUE) and optimizing irrigation schedules in croplands. In this study, E/T partitioning was performed on ETa rates measured using the eddy covariance (EC) technique in three winter wheat growing seasons from October 2020 to June 2023. The variation in the crop coefficients (Kc, α, and KHc) were quantified by combining the ETa and reference evapotranspiration rates using the Penman–Monteith, Priestley–Taylor, and Hargreaves equations. In addition, the application of models based on the modified crop coefficient (Kc, α, and KHc) was proposed to estimate the ETa rates. According to the obtained results, the average cumulative ETa, T, and E rates in the three winter wheat growth seasons were 471.4, 265.2, and 206.3 mm, respectively. The average T/ETa ratio ranged from 0.16 to 0.72 at the different winter wheat growth stages. Vapor pressure deficit (VPD) affected the ETa rates at a threshold of 1.27 KPa. The average Kc, α, and KHc values in the middle stage were 1.34, 1.54, and 1.21, respectively. The measured ETa rates and ETa rates estimated using the adjusted Kc, α, and KHc showed regression slope coefficients of 0.96, 0.99, and 0.96, and coefficients of determination (R2) of 0.92, 0.93, and 0.90, respectively. Therefore, the Priestley–Taylor-equation-based adjusted crop coefficient is recommended. The adjusted crop-coefficient-based models can be used as valuable tools for local policymakers to effectively improve water use.

Evapotranspiration Estimation Using Machine Learning Methods

Abstract The study examined the performance of four machine learning algorithms (regression trees, boosted trees, random forests, and artificial neural networks) for estimating evapotranspiration (ETo) based on incomplete meteorological data. Meteorological variables (mean and maximum air temperature, average air humidity, average level of solar radiation, vapor pressure deficit, extraterrestrial solar radiation, and day number of the year) were used as input. The simulation used two calculation scenarios: data with and without average solar radiation. The performance of the different machine learning models was evaluated using the mean square error, root mean square error, coefficient of determination, and slope of regression forced through the origin between the measured and simulated ETo. The results demonstrated that the applied models were able to describe nonlinear relationships between weather parameters and evapotranspiration. The accuracy of evapotranspiration estimation depended on the type of input variables and the machine learning model used. The highest level of evapotranspiration prediction was obtained using the artificial neural networks model. Including solar radiation data in the calculations improved the quality of evapotranspiration prediction in all four models. In the absence of data on the actual solar radiation reaching the Earth's surface, it is advisable to supplement the input data with data on extraterrestrial solar radiation and the day number of the year. Such an approach can be helpful in areas and situations with limited access to meteorological data.

Surface flux equilibrium estimates of evaporative fraction and evapotranspiration at global scale: Accuracy evaluation and performance comparison

Accurate estimation of terrestrial evapotranspiration (ET) is crucial to understand the water cycle and partitioning of turbulent energy fluxes at the land surface. However, large-scale ET estimation is always difficult to achieve due to the complexity of controlling factors and the heterogeneity of landscapes. A recent model, referred to as the surface flux equilibrium (SFE) model, proposes that evaporative fraction (EF, the ratio of ET to available energy at the surface) can be estimated accurately from near-surface specific humidity and air temperature, which makes it applicable to common weather stations and climate reanalysis datasets. However, previous validation only focused on the evaluation of this model at site and basin scale for ET estimation, making the real performance of the SFE model for EF estimation heretofore unknown. There also exists a gap in the comparison between the SFE model and other existing products in the representation of global ET patterns and variation trend. Against this background, a comprehensive validation was performed in this study to evaluate the accuracy of the SFE model for both EF and ET estimation. The validation against 136 flux towers worldwide shows that the ground-based SFE model achieved EF estimation with a root-mean-square-error (RMSE) of 0.22 and 0.16 at daily and monthly scale, respectively. The corresponding RMSE of ET estimation was 0.94 mm/day and 16.37 mm/month. The application of the SFE model on a global scale was achieved by using two reanalysis products (ERA5-Land and GLDAS-CLSM) as inputs respectively. The results indicate that the SFE model did hold potential to reduce the RMSE of these two products for ET estimation, but the cost was the decrease in correlation coefficient (r). The comparison with five existing global ET products (BESS, GLEAM, AVHRR, MOD16, and SSEBop) shows that the SFE model performed on the basis of ERA5-Land product has achieved moderate accuracy among these products. Judged from r, mean absolute error, RMSE, and bias, its accuracy was ranked as fourth, fourth, second, and second, respectively. However, a further evaluation over different land cover types suggests that the ERA-based SFE model has significant superiority over cropland, relative superiority over forest and savannas, and significant inferiority over shrubland and grassland. This analysis is the first global validation of the SFE model for EF estimation as well as its comparison with existing global ET products. The validation analysis proves that this simple model has generally achieved comparable accuracy to existing complex models in the representation of global ET patterns and variation trend.

A Stochastic Bayesian Artificial Intelligence Framework to Assess Climatological Water Balance under Missing Variables for Evapotranspiration Estimates

The sustainable use of water resources is of utmost importance given climatological changes and water scarcity, alongside the many socioeconomic factors that rely on clean water availability, such as food security. In this context, developing tools to minimize water waste in irrigation is paramount for sustainable food production. The evapotranspiration estimate is a tool to evaluate the water volume required to achieve optimal crop yield with the least amount of water waste. The Penman-Monteith equation is the gold standard for this task, despite it becoming inapplicable if any of its required climatological variables are missing. In this paper, we present a stochastic Bayesian framework to model the non-linear and non-stationary time series for the evapotranspiration estimate via Bayesian regression. We also leverage Bayesian networks and Bayesian inference to provide estimates for missing climatological data. Our obtained Bayesian regression equation achieves 0.087 mm · day−1 for the RMSE metric, compared to the expected time series, with wind speed and net incident solar radiation as the main components. Lastly, we show that the evapotranspiration time series, with missing climatological data inferred by the Bayesian network, achieves an RMSE metric ranging from 0.074 to 0.286 mm · day−1.

Evapotranspiration estimation using a satellite-based surface energy balance: a case study of Upper Bari Doab, Pakistan

Evapotranspiration estimation using a satellite-based surface energy balance: a case study of Upper Bari Doab, Pakistan

Estimation of crop evapotranspiration and crop coefficient for coriander using Portable Automatic Closed Canopy Chamber

An experiment was carried out to determine crop evapotranspiration and crop coefficient of Coriander crop by using Portable Automatic Closed Canopy Chamber (PACCC), Micro-Lysimeter (MLs) and field water balance (FWB) methods. The results revealed that there was no significant difference in the coriander crop evapotranspiration measured by the MLs inside and outside the PACCC and no significant difference among the crop evapotranspiration measured by the PACCC, MLs and FWB at 95 percent confidence level was found. It is indicating that, there are no effects of the change in micro-climate for a short period of 2 minutes in the chamber and on the plant physiological processes. During validation of PACCC, the average crop coefficients measured by MLs were varied from 0.66 to 1.26 for coriander crop. However, the stage wise crop coefficients of corianders measured by FWB were varied from 0.67 to 1.28 during field testing of PACCC. The result showed that the PACCC can be used for measurement of crop evapotranspiration in the field condition.

Improving field-scale crop actual evapotranspiration monitoring with Sentinel-3, Sentinel-2, and Landsat data fusion

One of the primary applications of satellite Land Surface Temperature (LST) observations lies in their utilization for modeling of actual evapotranspiration (ET) in agricultural crops, with the primary goals of monitoring and enhancing irrigation practices and improving crop water use productivity, as stipulated by Sustainable Development Goal (SDG) indicator 6.4.1. Evapotranspiration is a complex and dynamic process, both temporally and spatially, necessitating LST observations with high spatio-temporal resolution. Presently, none of the existing spaceborne thermal sensors can provide quasi-daily field-scale LST observations, prompting the development of methods for data fusion (thermal sharpening) of observations from various shortwave and thermal sensors to meet this spatio-temporal requirement. Previous research has demonstrated the effectiveness of combining shortwave-multispectral Sentinel-2 observations with thermal-infrared Sentinel-3 observations to derive daily, field-scale LST and ET estimates. However, these studies also highlighted limitations in capturing the distinct thermal contrast between cooler LST in irrigated agricultural areas and the hotter, adjacent dry regions. In this study, we aim to address this limitation by incorporating information on thermal spatial variability observed by Landsat satellites into the data fusion process, without being constrained by infrequent or cloudy Landsat thermal observations and while retaining the longwave radiance emission captured by the Sentinel-3 thermal sensor at its native resolution. Two approaches are evaluated, both individually and as a complementary combination, and validated against in situ LST measurements. The best performing approach, which leads to reduction in root mean square error of up to 1.5 K when compared to previous research, is subsequently used to estimate parcel-level actual evapotranspiration. The ET modeling process has also undergone various improvements regarding the gap-filling of input and output data, input datasets and code implementation. The resulting ET is validated using lysimeters and eddy covariance towers in Spain, Lebanon, Tunisia, and Senegal resulting in minimal overall bias (systematic underestimation of less than 0.07 mm/day) and a low root mean square error (down to 0.84 mm/day) when using fully global input datasets. The enhanced LST sharpening methodology is sensor agnostic and should remain relevant for the upcoming thermal missions while the accuracy of the modeled ET fluxes is encouraging for further utilization of observations from Sentinel satellites, and other Copernicus data, for monitoring SDG indicator 6.4.1.

Development of an enhanced bidirectional recurrent neural network combined with time-varying filter-based empirical mode decomposition to forecast weekly reference evapotranspiration

Evapotranspiration is one of agricultural water management's most significant and impactful hydrologic processes. A new multi-decomposition deep learning-based technique is proposed in this study to forecast weekly reference evapotranspiration (ETo) in western coastal regions of Australia (Redcliffe and Gold Coast). The time-varying filter-based empirical mode decomposition (TVF-EMD) technique was used to first break down the original meteorological variables/signals into intrinsic mode decomposition functions (IMFs), which included maximum and minimum temperature, relative humidity, wind speed, and solar radiation. Using a partial autocorrelation function (PACF), the significant lagged values were then calculated from the decomposed sub-sequences (i.e., IMFs). A novel Extra Tree- Boruta feature selection algorithm was used to extract important features from the decomposed IMFs. Four machine learning approaches, including bidirectional recurrent neural network (Bi-RNN), multi-layer perceptron neural network (MLP), random forest (RF), and extreme gradient boosting (XGBoost), were used to forecast weekly evapotranspiration using the TVF-EMD-based decomposed meteorological data. Different statistical metrics were applied to evaluate the model performances. The results showed that the decomposition of the input data by TVF-EMD significantly improved the accuracy compared with the non-decomposed inputs (single models without decomposition). The findings indicate that the TVF-BiRNN model, as presented, achieved the highest level of accuracy in simulating weekly ET0 at both the Redcliffe and Gold Coast stations (Redcliffe: R=0.9281, RMSE=3.8793 mm/week, MAPE = 9.2010%; Gold Coast: R=0.8717, RMSE=4.1169 mm/week, MAPE = 11.5408%). The novel hybrid modeling technique can potentially improve agricultural water management through its ability to generate more accurate ETo estimates weekly. The proposed methodology exhibits potential applicability to various other environmental and hydrological modeling issues.

Average monthly recharge, surface runoff, and actual evapotranspiration estimation using WetSpass-M model in Low Folded Zone, Iraq

Abstract The evaluation of the spatial and temporal distribution of groundwater recharge is required to develop the regional groundwater model for more precise simulations of various management scenarios. WetSpass-M (Water and Energy Transfer between Soil, Plants, and Atmosphere under Steady-State Conditions), which is a GIS-based spatially distributed water balance model, is deployed to evaluate monthly groundwater recharge, surface runoff, and actual evapotranspiration in the Low Folded Zone from 2000 to 2019. ArcGIS software prepares the essential relevant input data for the Wetspass-M model as grid maps. They include monthly climatological measurements (precipitation, temperature, and wind speed), land cover distribution, soil map, groundwater depth, topography, and slope. The mean annual groundwater recharge, evapotranspiration, and runoff were found to be 128, 131, and 72 mm, respectively. Accordingly, recharge accounts for 39% of the precipitation, while the rest, 40 and 21%, become evapotranspiration and surface runoff, respectively. WetSpass-M model results are meant to enable integrated groundwater modeling. The study of simulation data demonstrates that the WetSpass-M model accurately simulates the hydrological water budget components of the Low Folded Zone. In addition, a better understanding of the simulated spatial and temporal distribution of water balance components is beneficial for managing and planning the available water resources of the Low Folded Zone in Iraq, which faces water scarcity threats.

Evaluation of spatial downscaling for satellite retrieval of evapotranspiration from the nonparametric approach in an arid area

Evapotranspiration (ET) data is critical for monitoring the limited hydrological resources in arid areas. At present, the spatial resolution of ET derived by the thermal infrared images from satellite data is coarse (always more than 500 m). The low-resolution data cannot meet the needs of the high-precision mapping of ET. Therefore, it is important to undertake a downscaling procedure to improve the spatial resolution of ET. This paper selects the oasis region and surrounding deserts of Zhangye City of China as study area and compares with two downscaling procedures that include input and output downscaling procedure. The ET is estimated with the Remote Sensing-Non Parametric (RS-NP) model. The two downscaling procedures use Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM) to improve the spatial resolution of ET from 1-km to 100 m. Compared with the result derived from the output downscaling procedure, the high-resolution ET derived from the input downscaling procedure has a slightly better performance with R2 of 0.92 and RMSE of 41.8 W/m2, respectively. Neglectingthe downscaling of land surface temperature and broadband reflectance (BBR) result in larger errors of ET estimation with bias values of 73.0 W/m2 and 23.7 W/m2, respectively. Those values account for almost 1/3 and 1/10 of ET values, respectively. In non-vegetated areas, those proportions increase further around 90.6 % and 21.3 %, respectively. Furthermore, the influence of the neglect of the downscaling of the near-infrared reflectance on the ET estimation dominates that of BBR with a proportion of 42 %. The input downscaling procedure is suitable for generating the high-precision map of ET, especially in vegetated areas and during northern hemisphere summer months. Our study is helpful to obtain ET maps in fine spatial resolutions and contribute to the comprehensive understanding of the importance of the downscaling of each kind of input for reliable ET estimations.

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.

A Multi-sensor Analysis of Selected Reflectance-Based Crop Coefficient Models for Daily Maize Evapotranspiration Estimation

This study evaluated three reflectance-based crop coefficient models (RBCC) for daily maize actual evapotranspiration (ETa) estimates, using multispectral data from spaceborne, airborne, and proximal platforms. The goal was to identify the optimal multispectral sensor that gives more accurate daily ETa estimates. The remote sensing (RS) multispectral platforms included Landsat-8, Sentinel-2, Planet CubeSat, handheld multispectral radiometer (MSR), and unmanned aerial system or UAS, spatial resolution from 30 m to 0.03 m. Three RBCC models that use different vegetation indices as input variables were evaluated in the study. One RBCC uses the normalized difference vegetation index (NDVI). The second model uses the soil-adjusted vegetation index (SAVI), and the third model uses canopy cover (fc). The data for this study were from two maize research sites in Greeley and Fort Collins, Colorado, USA, collected in 2020 and 2021. The Greeley site had a subsurface drip system, while the Fort Collins site had surface irrigation (furrow). Daily maize ETa predictions were compared with observed daily maize ETa data from an Eddy Covariance system installed at each research site. Results indicated that, depending on the RS of ETa algorithm and platform, the optimal input RS data was different. The MSR sensor (1 m) provided the best remote sensing data (input) for the SAVI-based RBCC ETa model, with a maize ETa error (MBE±RMSE) of -0.13 (-3%)±0.67 (16%) mm/d. Sentinel-2 was the best sensor for the remaining two RBCC daily maize ETa algorithms, since the errors for the NDVI-based and fc-based RBCC models for maize ETa were 0.21 (5%)±0.78 (18%) mm/d and 0.59 (14%)±1.07 (25%) mm/d, respectively. These results indicate the need for methods to improve the spectral quality of the remote sensing data to improve spatial ETa estimates and advance sustainable irrigation water management.

Increasing actual evapotranspiration on the Loess Plateau of China: An insight from anthropologic activities and climate change

Actual evapotranspiration (ETa) is an important element of water circulation, and susceptible to environmental change. Precise estimation of ETa and identifying its key influence factors are of great importance. Loess Plateau has been largely affected by human activities and climate change, and the dynamics of ETa and its key influencing factors remain unknown in this region. The spatio-temporal variations of ETa over Loess Plateau from 2000 to 2020 were depicted on the basis of the ETWatch model. Then the individual contributions of climatic factors and human activities to ETa variation were quantified by applying generalized additive model. Results indicated that the annual average ETa during the period of 2000–2020 displayed a fluctuating increasing pattern from 386.67 mm in 2000 to 544.73 mm in 2020, with a growth rate of 6.25 mm⋅yr−1. Spatially, the ETa displays a downward pattern from southeast to northwest, which is highly similar to vegetation coverage with the coefficients of 0.766–0.832. The ETa of forest, grassland, cropland, shrub, barren land and impervious land exhibited positive or negative significantly correlations with their corresponding land cover area at the 0.05 or 0.01 level. Sunshine duration, temperature and precipitation are the top three factors showing high correlations with ETa. The results of contribution analysis suggested that the variations in ETa were largely controlled by LUCC in the Loess Plateau, where the LUCC accounted for 52.89% of the total ETa. The impacts of LUCC on ETa were greater than that of climate change on ETa, although climate change was dominant in a few years. These results could contribute to optimizing water resource management and developing various water-saving tactics in the Loess Plateau.

Bias-blind and bias-aware assimilation of leaf area index into the Noah-MP land surface model over Europe

Abstract. Data assimilation (DA) of remotely sensed leaf area index (LAI) can help to improve land surface model estimates of energy, water, and carbon variables. So far, most studies have used bias-blind LAI DA approaches, i.e. without correcting for biases between model forecasts and observations. This might hamper the performance of the DA algorithms in the case of large biases in observations or simulations or both. We perform bias-blind and bias-aware DA of Copernicus Global Land Service LAI into the Noah-MP land surface model forced by the ERA5 reanalysis over Europe in the 2002–2019 period, and we evaluate how the choice of bias correction affects estimates of gross primary productivity (GPP), evapotranspiration (ET), runoff, and soil moisture. In areas with a large LAI bias, the bias-blind LAI DA leads to a reduced bias between observed and modelled LAI, an improved agreement of GPP, ET, and runoff estimates with independent products, but a worse agreement of soil moisture estimates with the European Space Agency Climate Change Initiative (ESA CCI) soil moisture product. While comparisons to in situ soil moisture in areas with weak bias indicate an improvement of the representation of soil moisture climatology, bias-blind LAI DA can lead to unrealistic shifts in soil moisture climatology in areas with strong bias. For example, when the assimilated LAI data in irrigated areas are much higher than those simulated without any irrigation activated, LAI will be increased and soil moisture will be depleted. Furthermore, the bias-blind LAI DA produces a pronounced sawtooth pattern due to model drift between DA updates, because each update pushes the Noah-MP leaf model to an unstable state. This model drift also propagates to short-term estimates of GPP and ET and to internal DA diagnostics that indicate a suboptimal DA system performance. The bias-aware approaches based on a priori rescaling of LAI observations to the model climatology avoid the negative effects of the bias-blind assimilation. They retain the improvements in GPP anomalies from the bias-blind DA but forego improvements in the root mean square deviations (RMSDs) of GPP, ET, and runoff. As an alternative to rescaling, we discuss the implications of our results for model calibration or joint parameter and state update DA, which has the potential to combine bias reduction with optimal DA system performance.

Using feature engineering and machine learning in FAO reference evapotranspiration estimation

Abstract The authors of this study investigated the use of machine learning (ML) and feature engineering (FE) techniques to accurately determine FAO reference evapotranspiration (ETo) with a minimal number of climate variables being measured. The recommended techniques for areas with insufficient measurements are based solely on daily temperature readings. Various ML methods were tested to evaluate how sophisticated an ML algorithm is for this task necessary. The main emphasis was on feature engineering, which involves converting raw variables into inputs better suited for ML algorithms, resulting in improved results. FE methods for estimating evapotranspiration include approximations of clear-sky solar radiation based on altitude and Julian day, approximate relative humidity and wind velocity, a categorical month variable, and variables interactions. The authors confirmed that the ability of ML in such tasks is not solely dependent on choosing the suitable algorithm but also on this frequently ignored step. The results of computational experiments are presented, accompanied by a comparison of the proposed method against standard ETo empiric equations. Machine learning methods, mainly due to the transformation of raw variables using FE, provided better results than traditional empirical methods and sophisticated ML algorithms without FE. In addition, the authors tested the applicability of the developed models in the broader area to evaluate the possibility of their generalizability. The potential of this approach to deliver improved predictions, reduced input requirements, and increased efficiency holds interesting promise for optimizing water management strategies, irrigation planning, and decision-making within the agricultural sector.