Hargreaves Samani Research Articles

Daily reference evapotranspiration prediction using empirical and data-driven approaches: A case study of Adana plain

Precise determination of the reference evapotranspiration (ET0) is vital to studying the hydrological cycle. In addition, It plays a significant role in properly managing and allocating water resources in agriculture. The objective of this research was to examine the effectiveness of five different data-driven techniques, including artificial neural networks "multilayer perceptron" (ANN), gene expression programming (GEP), random forest (RF), support vector machine "radial basis function" (SVM), and multiple linear regression (MLR) to model the daily ET0. These methods were also compared with Hargreaves-Samani (HS), Oudin, Ritchie, Makkink (MAK), and Jensen Haise (JH) empirical models and their calibrated versions. The empirical models JH and MAK performed better than the models HS and Oudin after being calibrated by linear regression. All data-driven methods with four inputs were superior to the original and calibrated empirical models. The RF and ANN methods generally demonstrated better estimation accuracy than other data-driven methods. The performance of the RF and ANN models that utilized Tmax, Tmin, and Rs inputs, as well as those that incorporated Tmax, Tmin, Rs, and U2 inputs, proved to be superior to their corresponding MLR-based and GEP-based models for predicting ET0 in the Adana plain, which is characterized by a Mediterranean climate. Nevertheless, the GEP and MLR methods have the advantage of utilizing explicit algebraic equations, making them more convenient to apply, especially in the context of agricultural irrigation practices.

Future Reference Evapotranspiration Trends in Shandong Province, China: Based on SAO-CNN-BiGRU-Attention and CMIP6

One of the primary factors in the hydrological cycle is reference evapotranspiration (ET0). The prediction of ET0 is crucial to manage irrigation water in agriculture under climate change; however, little research has been conducted on the trends of ET0 changes in Shandong Province. In this study, to estimate ET0 in the entire Shandong Province, 245 sites were chosen, and the monthly ET0 values during 1901–2020 were computed using the Hargreaves–Samani formula. A deep learning model, termed SAO-CNN-BiGRU-Attention, was utilized to forecast the monthly ET0 during 2021–2100, and the predictions were compared to two CMIP6 climate scenarios, SSP2-4.5 and SSP5-8.5. The hierarchical clustering results revealed that Shandong Province encompassed three homogeneous regions. The ET0 values of Clusters H1 and H2, which were situated in inland regions and major agricultural areas, were the highest. The SAO-CNN-BiGRU-Attention and SSP5-8.5 forecasting results generally displayed a monotonically growing trend during the forecast period in the three regions; however, the SAO-CNN-BiGRU-Attention model displayed a declining tendency at a few points. According to the SAO-CNN-BiGRU-Attention and SSP5-8.5 results, during 2091–2100, H1, H2, and H3 will reach their peaks; the SSP2-4.5 results show that H1, H2, and H3 will peak in 2031–2040. At the end of the forecast period, for Clusters H1, H2, and H3, the prediction rate of SAO-CNN-BiGRU-Attention increased by 1.31, 1.56%, and 1.80%, respectively, whereas SSP2-4.5’s prediction rate increased by 0.31%, 0.95%, and 1.57%, respectively, and SSP5-8.5’s prediction rate increased by 10.88%, 10.76%, and 10.69%, respectively. The prediction results of SAO-CNN-BiGRU-Attention were similar to those of SSP2-4.5 (R2 > 0.96). The SAO-CNN-BiGRU-Attention deep learning model can be used to forecast future ET0.

Assessing the Accuracy of 50 Temperature-Based Models for Estimating Potential Evapotranspiration (PET) in a Mediterranean Mountainous Forest Environment

Potential evapotranspiration (PET) is a crucial parameter for forest development, having an important role in ecological, biometeorological, and hydrological assessments. Accurate estimations of PET using the FAO–56 Penman–Monteith (FAO–56 PM) benchmark method require a wide range of data parameters, which are not typically available at meteorological stations installed in forest environments. The aim of this study is to investigate the accuracy of various methods with low data requirements for assessing PET in a Mediterranean forest environment and propose appropriate alternatives for accurate PET estimation. Specifically, 50 temperature-based methods were evaluated against the FAO–56 PM method in a sub-humid forest in northern Greece, using high-quality daily meteorological data. The outcomes indicate that temperature-based methods offer a viable alternative for PET estimation when data availability is limited, with a considerable number of methods (22) presenting low deviations (up to 10%) compared to the benchmark method. Temperature-based models outperformed those incorporating water-related parameters (as relative humidity or precipitation) in Mediterranean forest environments. The top performing methods in the study site, based on several statistical indices, were the equations of Ravazzani et al., proposed in 2012, followed by Hargreaves–Samani in 1985 and Heydari and Heydari in 2014.

Short-Term Forecasting of Daily Reference Crop Evapotranspiration Based on Calibrated Hargreaves–Samani Equation at Regional Scale

Short-Term Forecasting of Daily Reference Crop Evapotranspiration Based on Calibrated Hargreaves–Samani Equation at Regional Scale

On the Sensitivity of the Potential Evapotranspiration of Egypt to Different Initial Conditions of the Soil Moisture Using a High-resolution Regional Climate Model

The present study investigated the sensitivity of the Potential Evapotranspiration (PET) of Egypt to different initial conditions of soil moisture using a high-resolution Regional Climate Model (RegCM4). To address this issue, two 32-year simulations were conducted from 1979 to 2010. The first two years were considered as spin-up. In the first simulation, the RegCM4 was initialized with the satellite product ESACCI (ESA) and the RegCM4 was initialized with version 3 of the Century Reanalysis Product (CEN) in the second simulation. The PET was computed using a calibrated version of the Hargreaves–Samani (HS) equation. In the two simulations, the RegCM4 was configured with the Community Land Model version 4.5 (CLM45). Era-Interim Reanalysis of 1.5 degrees (EIN15) was used to downscale the RegCM4 with 25 km grid spacing. The ERA5-based product (ERA5) was used as the observational dataset. The results showed that switching between ESA and CEN didn’t induce a notable influence on the daily maximum air temperature. Regarding the daily minimum air temperature, the CEN outperforms the ESA concerning the Climate Research Unit (CRU). In comparison with the ERA5 product, the CEN was able to reduce the PET negative bias (0.8 to 1.8 mm day-1) better than the one noted in the ESA (0.8 to 3 mm day-1). On a point scale, the performance of the ESA/CEN varies with the location and month. Such a point was confirmed using different statistical metrics such as mean bias, Pearson correlation coefficient and standard deviation ratio. In conclusion, the CEN can be recommended to initialize the RegCM4 to ensure a reliable estimation of the PET of Egypt either in the present climate or under different future scenarios.

Global future potential evapotranspiration signal using Penman-Monteith and Hargreaves-Samani method by latitudes based on CMIP6

This study estimated global potential evapotranspiration (PET) using six climate variables of 14 CMIP6 GCMs at different latitudes of both hemispheres. Hargreaves-Samani (HS) and Penman-Monteith (PM) were used to estimate historical and future PETs. To evaluate the historical reproducibility of the six climate variables of the CMIP6 GCMs, five evaluation indicators were used and compared with the NCEP/NCAR reanalysis data. Based on the evaluation metrics, the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) was used to calculate the weights for Multi-Model Ensemble (MME) for each band of latitude. After that, the projected annual and seasonal PETs were estimated for the near (2031–2065) and far (2066–2100) futures. The change rate of PET for five timeframes (Annual, Spring, Summer, Fall, and Winter) was calculated for four socioeconomic shared pathways (SSPs). As a result, the PM's PET and HS's PET for all scenarios showed the highest increased signals at 75.5° ∼ 90° in the northern hemisphere (NH) compared to the other latitudes. Especially, the change in the HS of SSP5–8.5 was increased by 26.4% in NH latitude 45.5° to 60° (NL4) compared to the base period. Furthermore, the changes in PET of PM and HS for December, January, and February (DJF) and September, October, and November (SON), and June, July, and August (JJA) lead to remarkably high increases in both hemispheres, irrespective of the level of the emission scenario, completely ignoring the seasonal water cycle in the historical period (1950–2014).

An evolutionary parsimonious approach to estimate daily reference evapotranspiration

The reference evapotranspiration (ETo) is an essential component in hydrological and ecological processes. The objective of this research is to develop an explicit model to estimate ETo only using commonly measurable meteorological parameters such as relative humidity, air temperature, and wind speed, where the measurements corresponding to solar radiation are omitted. The model was generated using Genetic Programming (GP), evaluated, and validated with reference data ETo using FAO56-PM. This reference data was obtained from different climates (warm-temperate and arid-warm) and latitudes, acquired from CIMIS stations in the state of California, United States, and the El Porvenir station in the state of Coahuila, located in north-central Mexico. After applying the proposed methodology, a total of 3754 results were generated, demonstrating a significant improvement in the estimation of ETo compared to the Hargreaves–Samani model. A particularly noteworthy result revealed that our approach outperformed the Hargreaves–Samani model in the training phase by 27%, and in the testing phase by 16%, on average. In order to achieve a generalized model, a dataset encompassing meteorological stations in two different climates (warm-temperate and arid-warm) and various latitudes was utilized. The obtained outcome unveiled a highly effective model for estimating ETo in diverse climatic contexts, eliminating the need for local adjustments. This model significantly surpassed the Hargreaves–Samani model, exhibiting superior performance by 17% during the training phase and 18% during the testing phase. These results conclusively underscore the capability of our approach to provide more accurate and reliable ETo estimates. These results conclusively underscore the capability of our approach to provide more accurate and reliable ETo estimates. Finally, to validate the model, four different datasets with climates similar to those used for model creation (warm-temperate, warm-arid) and different latitudes were employed. The validation stage results clearly indicate the superiority of our reference evapotranspiration ETo11 model over the Hargreaves–Samani model by 51% in warm-temperate climates. For the dataset with arid-warm climate, our model continued to show satisfactory results, surpassing the Hargreaves–Samani model by 8%. GP emerges as an innovative and effective alternative for simplified model development. This approach introduces a novel paradigm that facilitates the efficient development of models, standing out for its simplicity and effectiveness in generating solutions.

Trend Projections of Potential Evapotranspiration in Yangtze River Delta and the Uncertainty

Global warming may increase potential evapotranspiration (ETp), reducing the water resources in Yangzi River Delta. Therefore, it is important to investigate the trend of ETp there under the background of climate change. To this purpose, the systematic biases in temperature outputs of 24 global climate models (GCMs) under 3 shared socioeconomic pathways—representative concentration pathways (SSPs) emission scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5)—are first corrected by using 8 bias correction methods. Then, the trend of ETp in Yangtze River Delta is projected by using 4 ETp calculation formulas (Blaney–Criddle, Hargreaves–Samani, Makkink, and Priestley–Taylor). The uncertainty of the projections is estimated and decomposed by using multi-way analysis of variance frameworks. The influence of uncertainty on the projected change signal is quantified by using the signal-to-noise ratio. The results show that all emission scenarios indicate robust increments of ETp. Specifically, relative to 1971~2000, ETp will increase by 0.14~0.17 mm d−1 (5.7~6.8%) during 2021~2050 and by 0.21~0.41 mm d−1 (8.5~16.7%) during 2061~2090, respectively. During 2021~2050, the uncertainty of ETp projections is dominantly contributed by the main effects of GCM (63%) and the ETp calculation formula (24%). During 2061~2090, it is mainly contributed by the main effect of GCM (36%), followed by the main effects of the emission scenario (34%) and the ETp calculation formula (18%). The ETp projections are generally reliable and robust during the two projection periods.

SolarET: A generalizable machine learning approach to estimate reference evapotranspiration from solar radiation

Irrigation is the most significant consumer of freshwater worldwide. Deciding on the right amount of irrigation is crucial for sustainable water management and food production. The Penman-Monteith (P-M) reference crop evapotranspiration (ETO) is the gold standard in irrigation management and scheduling; however, its calculation requires measurements from multiple sensors over an extended reference grass surface. The cost of land, sensors, maintenance, and water to keep the grass surface green impedes having a dense network of ETO stations. To solve this challenge, this research aims to develop an input-limited ETO estimation approach based on historical weather data and machine learning (ML) algorithms to relax the need for a reference grass surface. This approach, called "SolarET," takes solar radiation (RS) data as its sole input. RS is the only meteorological driving factor of ETO that does not rely on the measuring surface. To test the generalizability of SolarET, we test its performance over unseen arbitrary locations across California. California is chosen as the case study since it is one of the world's most hydrologically altered and agriculturally productive regions. In total, 19,088,736 hourly data samples from 131 automated weather stations have been used in this study. The ML models have been trained over 114 stations and tested over 17 unseen stations, each representing a California climatic zone. Our findings point to the superiority of decision tree-based algorithms versus neural networks. SolarET works best in irrigation-oriented regions of California (e.g., Central Valley) and is less accurate in coastal and desert zones. Our results demonstrate the higher accuracy of SolarET using hourly (RMSE = 0.93 mm/day) and daily (RMSE = 0.97 mm/day) RS data in comparison to well-known empirical alternatives like Priestley-Taylor (PT) (RMSE = 1.35 mm/day) and Hargreaves-Samani (HS) (RMSE = 2.69 mm/day).

Novel Approaches for the Empirical Assessment of Evapotranspiration over the Mediterranean Region

The hydrological cycle should be scrutinized and investigated under recent climate change scenarios to ensure global water management and to increase its utilization. Although the FAO proposed the use of the Penman–Monteith (PM) equation worldwide to predict evapotranspiration (ET), which is one of the most crucial components of the hydrological cycle, its complexity and time-consuming nature, have led researchers to examine alternative methods. In this study, the performances of numerous temperature-driven ET methods were examined relative to the PM using daily climatic parameters from central stations in 11 districts of the Kahramanmaras province. Owing to its geographical location and other influencing factors, the city has a degraded Mediterranean climate with varying elevation gradients, while its meteorological patterns (i.e., temperature and precipitation) deviate from those of the main Mediterranean climate. A separate evaluation was performed via ten different statistical metrics, and spatiotemporal ET variability was reported for the districts. This study revealed that factors such as altitude, terrain features, slope, aspect geography, solar radiation, and climatic conditions significantly impact capturing reference values, in addition to temperature. Moreover, an assessment was conducted in the region to evaluate the effect of modified ET formulae on simulations. It can be drawn as a general conclusion that the Hargreaves–Samani and modified Blaney–Criddle techniques can be utilized as alternatives to PM in estimating ET, while the Schendel method exhibited the lowest performance throughout Kahramanmaras.

Comparative Analysis of Climate Change Impacts on Climatic Variables and Reference Evapotranspiration in Tunisian Semi-Arid Region

Systematic biases in general circulation models (GCM) and regional climate models (RCM) impede their direct use in climate change impact research. Hence, the bias correction of GCM-RCMs outputs is a primary step in such studies. This study compares the potential of two bias correction methods (the method from the third phase of the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP3) and Detrended Quantile Matching (DQM)) applied to the raw outputs of daily data of minimum and maximum air temperatures and precipitation, in the Cap-Bon region, from eight GCM-RCM combinations. The outputs of GCM/RCM combinations were acquired from the European branch of the coordinated regional climate downscaling experiment (EURO-CORDEX) dataset for historical periods and under two representative concentration pathway (RCP4.5 and RCP8.5) scenarios. Furthermore, the best combination of bias correction/GCM-RCM was used to assess the impact of climate change on reference evapotranspiration (ET0). Numerous statistical indicators were considered to evaluate the performance of the bias correction/historical GCM-RCMs compared to the observed data. Trends of the Hargreaves–Samani_ET0 model during the historical and projected periods were determined using the TFPMK method. A comparison of the bias correction methods revealed that, for all the studied model combinations, ISIMIP3 performs better in reducing biases in monthly precipitation. However, for Tmax and Tmin, the biases are greatly removed when the DQM bias correction method is applied. In general, better results were obtained when the HadCCLM model was used. Before applying bias correction, the set of used GCM-RCMs projected reductions in precipitation for most of the months compared to the reference period (1982–2006). However, Tmin and Tmax are expected to increase in all months and for the three studied periods. Hargreaves–Samani ET0 values obtained from the best combination (DQM/ HadCCLM) show that RCP8.5 (2075–2098) will exhibit the highest annual ET0 increase compared to the RCP4.5 scenario and the other periods, with a change rate equal to 11.85% compared to the historical period. Regarding spring and summer seasons, the change rates of ET0 are expected to reach 10.44 and 18.07%, respectively, under RCP8.5 (2075–2098). This study shows that the model can be used to determine long-term trends in ET0 patterns for diverse purposes, such as water resources planning, agricultural crop management and irrigation scheduling in the Cap-Bon region.

Integrating machine learning and empirical evapotranspiration modeling with DSSAT: Implications for agricultural water management

The availability of accurate reference evapotranspiration (ETo) data is crucial for developing decision support systems for optimal water resource management. This study aimed to evaluate the accuracy of three empirical models (Hargreaves-Samani (HS), Priestly-Taylor (PT), and Turc (TU)) and three machine learning models (Multiple linear regression (LR), Random Forest (RF), and Artificial Neural Network (NN)) in estimating daily ETo compared to the Penman-Monteith FAO-56 (PM) model. Long-term data from 42 weather stations in Florida were used. Moreover, the effect of ETo model selection on sweet corn irrigation water use was investigated by integrating simulated ETo data from empirical and ML models using the Decision Support System for Agrotechnology Transfer (DSSAT) model at two locations (Citra and Homestead) in Florida. Furthermore, a linear bias correction calibration technique was employed to improve the performance of empirical models. Results were consistent in that the NN and RF models outperformed the empirical models. The empirical models tended to underestimate and overestimate small and high daily ETo values, respectively, with the HS model exhibiting the least accuracy. However, calibrated PT and TU models performed comparably to the ML models. Results also revealed that using an inappropriate ETo model could lead to over-irrigation by up to 54 mm during a single crop season. Overall, ML models have proven reliable alternatives to the PM model, especially in regions with access to long-term data due to their site-independent performance. In areas without long-term data for ML model training and testing, calibrating empirical models is viable, but site-specific calibration is needed. It is important to highlight that distinct plant species exhibit varying transpiration characteristics and, consequently, have different water requirements. These differences play a pivotal role in shaping the overall impact of ETo models on crop water use.

Analyzing Evapotranspiration in Greenhouses: A Lysimeter-Based Calculation and Evaluation Approach

The absence of accurate measurement or calculation techniques for crop water requirements in greenhouses frequently results in over- or under-irrigation. In order to find a better method, this study analyzed the accuracy, data consistency and practicability of the Penman–Monteith (PM), Hargreaves–Samani (HS), Pan Evaporation (PAN), and Artificial Neural Network (ANN) models. Model-calculated crop evapotranspiration (ETC) was compared with lysimeter-measured crop evapotranspiration (ETC) in the National Precision Agriculture Demonstration Station in Beijing, China. The results showed that the actual ETC over the entire experimental period was 176.67 mm. The ETC calculated with the PM, HS, PAN, and ANN model were 146.07 mm, 189.45 mm, 197.03 mm, and 174.7 mm, respectively, which were different from the actual value by −17.32%, 7.23%, 11.52%, and −1.12%, respectively. The order of the calculation accuracy for the four models is as follows: ANN model > PAN model > PM model > HS model. By comprehensively evaluating the statistical indicators of each model, the ANN model was found to have a significantly higher calculation accuracy compared to the other three models. Therefore, the ANN model is recommended for estimating ETC under greenhouse conditions. The PM and PAN models can also be used after improvement.

Estimating reference crop evapotranspiration using optimized empirical methods with a novel improved Grey Wolf Algorithm in four climatic regions of China

Accurate estimation of reference crop evapotranspiration (ETo) is crucial for agricultural water management. As the simplified alternatives of the Penman-Monteith equation, empirical methods have been widely recommended worldwide. However, its application is still limited to parameters localization varied with geographical and climatic conditions, therefore developing an excellent optimization algorithm for calibrating parameters is very necessary. Regarding the above requirement, the present study developed a novel improved Grey Wolf Algorithm (MDSL-GWA) to optimize the most recommended ones among three types of ETo methods. After the optimization performance comparison among Least Square Method (LSM), Genetic Algorithm (GA), Grey Wolf Algorithm (GWA), and MDSL-GWA in four climatic regions of China, this study found that the Priestley-Taylor (PT) method was the best radiation-based (Rn-based) method and achieved better performance in temperate continental region (TCR), mountain plateau region (MPR), and temperate monsoon region (TMR) than other types. While the temperature-based (T-based) Hargreaves-Samani (HS) method performed best in subtropical monsoon region (SMR), further attaching better performance among the same type in TMR and TCR, while the Oudin method was the best T-based method in MPR. Moreover, the Romanenko method was better humidity-based (RH-based) in TCR and MPR, whereas the Brockamp-Wenner method exhibited higher in SMR and TMR. Furthermore, despite intelligence optimization algorithms significantly enhancing original ETo methods, the MDSL-GWA achieved best performance and outperformed other algorithms by 4.5–29.6% in determination coefficient (R2), 4.7–27.3% in nash-sutcliffe efficient (NSE), 3.7–44.4% in relative root mean square error (RRMSE), and 3.1–56.2% in mean absolute error (MAE), respectively. After optimization, the MDSL-GWA-PT was the most recommended ETo method in TMR, TCR, and MPR, and the median values of R2, NSE, RRMSE, and MAE ranged 0.907–0.958, 0.887–0.925, 0.083–0.103, and 0.115–0.162 mm, respectively. In SMR, the MDSL-GWA-HS produced the best ETo estimates, with median values of R2, NSE, RRMSE, and MAE being 0.876, 0.843, 0.112, and 0.146 mm, respectively. In summary, this study recommended the best ETo method and algorithms using accessible data in four climatic regions of China, which is helpful for decision-making in effective management and utilization of regional agricultural water resources.

Evaluation of Empirical Equations and Machine Learning Models for Daily Reference Evapotranspiration Prediction Using Public Weather Forecasts

Although the studies on model prediction of daily ETo based on public weather forecasts have been widely used, these studies lack the comparative evaluation of different types of models and do not evaluate the seasonal variation in model prediction of daily ETo performance; this may result in the selected model not being the best model. In this study, to select the best daily ETo forecast model for the irrigation season at three stations (Yinchuan, Tongxin, and Guyuan) in different climatic regions in Ningxia, China, the daily ETos of the three sites calculated using FAO Penman–Monteith equations were used as the reference values. Three empirical equations (temperature Penman–Monteith (PMT) equation, Penman–Monteith forecast (PMF) equation, and Hargreaves–Samani (HS) equation) were calibrated and validated, and four machine learning models (multilayer perceptron (MLP), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and gradient boosting with categorical features support (CatBoost)) were trained and validated against daily observed meteorological data (1995–2015 and 2016–2019). Based on public weather forecasts and daily observed meteorological data (2020–2021), the three empirical equations (PMT, PMF, and HS) and four machine learning models (MLP, XGBoost, LightGBM, and CatBoost) were compared in terms of their daily ETo prediction performance. The results showed that the daily ETo performance of the seven models in the irrigation season with a lead time of 1–7 days predicted by the three research sites decreased in the order of spring, autumn, and summer. PMT was the best model for the irrigation seasons (spring, summer, and autumn) at station YC; PMT and CatBoost with C3 (Tmax, Tmin, and Wspd) as the inputs were the best models for the spring, autumn irrigation seasons, and summer irrigation seasons at station TX, respectively. PMF, CatBoost with C4 (Tmax, Tmin) as input, and PMT are the best models for the spring irrigation season, summer irrigation season, and autumn irrigation season at the GY station, respectively. In addition, wind speed (converted from the wind level of the public weather forecast) and sunshine hours (converted from the weather type of the public weather forecast) from the public weather forecast were the main sources of error in predicting the daily ETo by the models at stations YC and TX(GY), respectively. Empirical equations and machine learning models were used for the prediction of daily ETo in different climatic zones and evaluated according to the irrigation season to obtain the best ETo prediction model for the irrigation season at the study stations. This provides a new idea and theoretical basis for realizing water-saving irrigation during crop fertility in other arid and water-scarce climatic zones in China.

Assessing forecasting performance of daily reference evapotranspiration: A comparative analysis of updated temperature penman-monteith and penman-monteith forecast models

Accurate forecast of reference evapotranspiration (ET0) is essential for effective water resource management and efficient irrigation scheduling. The Penman-Monteith model (PM) recommended by FAO56 is widely used as the standard method for ET0 forecasting, but PM model application is often limited by the absence of forecasted meteorological variables, especially solar radiation (Rs). Previous studies have proposed two types of models based on temperature and sunshine duration to estimate Rs. However, there is currently a lack of comprehensive comparative analysis research to evaluate the performance and applicability of these two types of models combined with weather forecast data for ET0 forecasting using the PM model. To address this issue, we selected China as the research area, which has complex climate zones and an uneven distribution of water resources. The forecasting results of Rs from temperature-based models (M1–M8) and a sunshine duration-based model (M9) are input independently into the PM model for ET0 forecasting. The updated PM model is named PMTM1–PMTM8 and PMF, respectively. We conducted a comprehensive and analytical assessment of the updated PMTM1–PMTM8 model, the PMF model and the Hargreaves-Samani (HS) model in different climate zones of China. The results showed that the accuracy of the M1–M8 and the M9 for forecasting Rs decreased as the forecasting period increased. The accuracy of the M9 was higher than that of M1–M8 in the overall accuracy in the subtropical monsoon climate (SMZ). Both the updated PMTM1–PMTM8 and PMF models were utilized to evaluate the ET0 forecasting performance in all five zones. The PMTM3 model exhibited the highest accuracy, with RMSE and MAE ranges of 0.671–1.572 mm d−1 and 0.532–1.365 mm d−1, respectively. In contrast, the PMF model displayed RMSE and MAE ranges of 0.690–1.590 mm d−1 and 0.641–1.437 mm d−1, respectively. Compared to the PMF model, the updated PMTM1–PMTM5 model showed higher accuracy in forecasting ET0 in the plateau mountain climate, temperate continental climate, temperate monsoon climate, and tropical monsoon climatic zones, but was slightly less accurate in the SMZ. Moreover, both models outperformed the Hargreaves-Samani (HS) model in terms of ET0 forecasting accuracy. Specifically, the updated PMTM1–PMTM5 model demonstrated improved accuracy compared to the temperature-based model HS model across various climate zones, with reductions in RMSE and MAE ranging from 0.117 mm d−1 to 0.616 mm d−1and 0.012 mm d−1 to 0.450 mm d−1, respectively. Overall, the updated PMTM3 model was better than the PMF and thus this model was recommended for daily ET0 forecasting for the near-future at all climate regions across China.

Short-term daily reference evapotranspiration forecasting using temperature-based deep learning models in different climate zones in China

The reference evapotranspiration (ETo) pertains to the evapotranspiration of cold-season grasses with an approximate height of 0.12 m or full-covered alfalfa with a height of 0.50 m. Accurate short-term ETo forecasts are indispensable for informed irrigation decisions by relevant departments and individuals. Four deep learning (DL) models, including Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional LSTM (Bi-LSTM), and Bidirectional GRU (Bi-GRU), as well as two calibrated empirical models (Hargreaves-Samani (HS) and reduced-set Penman–Monteith (RPM)), were used to evaluate the performance of the ETo forecast with a lead time of 1–7 d using temperature forecasts in different climates. The results reveal that the DL models and calibrated HS and RPM models exhibited comparable trends in the ETo forecasts for lead times of 1–7 d. Nonetheless, the DL models consistently outperformed the HS and RPM models across the diverse climatic regions in China. The DL models displayed an average root mean square error (RMSE) and mean absolute error (MAE) of less than 0.887 and 0.633 mm/d, respectively. Moreover, the mean correlation coefficient (R) and accuracy (ACC) exceeded 0.807% and 89.701%, respectively. Among the DL models, the LSTM model demonstrated slightly superior performance in short-term daily ETo forecasts in diverse climates. The LSTM model exhibited RMSE and MAE ranges of 0.563–0.875 mm/d and 0.418–0.626 mm/d, respectively, along with R and ACC ranges of 0.81–0.90 and 89.94–98.11%, respectively. Furthermore, even with an increase in lead time, the DL models continued to exhibit strong predictive capabilities, consistently surpassing the performance of the HS and RPM models. Overall, the trained DL models presented an exceptional ability to forecast the short-term daily ETo in various climatic regions of China. These models require only a few input variables and readily available data, making them highly advantageous for practical applications in ETo forecasting. Such models hold promise for significantly enhancing regional agricultural water-resource planning and management.

Assessment of Empirical Methods for Estimating Reference Evapotranspiration in Different Climatic Zones of Bosnia and Herzegovina

The study evaluated nine empirical methods for estimating reference evapotranspiration (ETo) in Bosnia and Herzegovina (BiH) across different climatic zones. The methods compared were the Hargreaves–Samani method (HS), the modified Hargreaves–Samani method (HM), the calibrated Hargreaves–Samani method (HC), the Priestley–Taylor method (PT), the Copais method (COP), the Makkink method (MAK), the Penman–Monteith method based on air temperature and overall average windspeed (PMT2), the Penman–Monteith method based on air temperature and regional average windspeed (PMT1.3), and the Penman–Monteith method based on air temperature and site-specific windspeed (PMTlok). These methods were tested against the “Food Agricultural Organization-Penman Monteith approach” (FAO-PM). The evaluation was performed using data from 20 meteorological stations in BiH, considering a common irrigation season (April–October) for two periods (2000–2005 and 2018–2022). The stations represented three climatic zones: semi-arid (SA), dry sub-humid (DSH), and moist sub-humid (MSH). The performance and ranking of the ETo methods were analyzed using the TOPSIS method. The trend of ETo during the common irrigation season for the period from 2018 to 2022 was determined using the Mann–Kendall test. The results of the study indicated that the HC method showed the best performance across all three climatic zones. The average root mean square error (RMSE) was 0.67 mm day−1, 0.49 mm day−1, and 0.50 mm day−1 for the SA, DSH, and MSH zones, respectively. As an alternative to the HC method, the PT method is recommended for its favorable results in both periods and in all zones. On the other hand, the HS method exhibited the highest average overestimation, particularly in the MSH zone, where ETo values were 18% higher compared with those of the FAO-PM method. The COP method also showed high overestimation and was not recommended for use. Regarding the MAK method, it resulted in underestimation during the period from 2000 to 2005, ranging from 17% in the DSH zone to 11% in the MSH zone. However, its performance improved during the period from 2018 to 2022, for which it ranked second place in the MSH zone. Among the PMT methods, the PMTlok, which utilized local average windspeed, yielded the best results. Despite performing well in the neighboring country of Serbia, the HM method showed poor overall performance in BiH. The findings of this study can serve as a foundation for further research in BiH to enhance irrigation practices in response to climate changes.

Development and evaluation of temperature-based deep learning models to estimate reference evapotranspiration

Efficient irrigation management of urban landscapes is critical in arid/semi-arid environments and depends on the reliable estimation of reference evapotranspiration (ETo). However, the available measured climatic data in urban areas are typically insufficient to use the standard Penman-Monteith for ETo estimation. Therefore, smart landscape irrigation controllers often use temperature-based ETo models for autonomous irrigation scheduling. This study focuses on developing deep learning temperature-based ETo models and comparing their performance with widely used empirical temperature-based models, including FAO Blaney & Criddle (BC), and Hargreaves & Samani (HS). We also developed a simple free and easy-to-access tool called DeepET for ETo estimation using the best-performing deep learning models developed in this study. Four artificial neural network (ANN) models were developed using raw weather data as inputs and the reconstructed signal obtained from the wavelet transform as inputs. In addition, long short-term memory (LSTM) recurrent neural network (NN) and one-dimensional convolution neural network (CNN) models were developed. A total of 101 active California Irrigation Management Information System (CIMIS) weather stations were selected for this study, with >725,000 data points expanding from 1985 to 2019. The performance of the models was evaluated against the standard CIMIS ETo. When evaluated at the independent sites, the temperature-based DL (Deep Learning) models showed 15–20% lower mean absolute error values than the calibrated HS model. No improvement in the performance of the ANN models was observed using reconstructed signals obtained from the wavelet transform. Our study suggests that DL models offer a promising alternative for more accurate estimations of ETo in urban areas using only temperature as input. The DeepET can be accessed from the Haghverdi Water Management Group website: http://www. ucrwater.com/software-and-tools.html.

Estimation of Reference Evapotranspiration in a Semi-Arid Region of Mexico.

Reference evapotranspiration (ET0) is the first step in calculating crop irrigation demand, and numerous methods have been proposed to estimate this parameter. FAO-56 Penman-Monteith (PM) is the only standard method for defining and calculating ET0. However, it requires radiation, air temperature, atmospheric humidity, and wind speed data, limiting its application in regions where these data are unavailable; therefore, new alternatives are required. This study compared the accuracy of ET0 calculated with the Blaney-Criddle (BC) and Hargreaves-Samani (HS) methods versus PM using information from an automated weather station (AWS) and the NASA-POWER platform (NP) for different periods. The information collected corresponds to Module XII of the Lagunera Region Irrigation District 017, a semi-arid region in the North of Mexico. The HS method underestimated the reference evapotranspiration (ET0) by 5.5% compared to the PM method considering the total ET0 of the study period (26 February to 9 August 2021) and yielded the best fit in the different evaluation periods (daily, 5-day mean, and 5-day cumulative); the latter showed the best values of inferential parameters. The information about maximum and minimum temperatures from the NP platform was suitable for estimating ET0 using the HS equation. This data source is a suitable alternative, particularly in semi-arid regions with limited climatological data from weather stations.