Reference crop evapotranspiration in distinct agricultural regions of Southern Brazil: a comparison of improved empirical models

The FAO56 Penman-Monteith model is globally accepted for the accurate determination of reference evapotranspiration (ETo). However, a lack of appropriate data encouraged the improved model’s approach to estimate ETo. This study compared the performance of 10 empirical models of ETo estimation (Penman, Priestley & Taylor, Tanner & Pelton, Makkink, Jensen & Haise, Hargreaves & Samani, Camargo, Benevides & Lopes, Turc, and Linacre) contrasted with the FAO56 model in two regions in Southern Brazil. Data were collected from automatic stations of the Brazilian National Institute of Meteorology (INMET) from December 21, 2019, to February 28, 2021. The determination coefficient (R²), mean square error (nRMSE), mean bias error (MBE), Willmott index (d), and Pearson’s correlation coefficient (r), clustering, and Principal Component Analysis (PCA) were performed. For the different regions, the radiation-based model proposed by Penman was the best alternative for estimating ETo. The model showed the most appropriated values for R2 (0.9015) and r (0.9494). The clustering and PCA analyses indicated the interrelations of the meteorological data and the combination of the models according to the parameters used for the determination of ETo.

Selecting Essential Factors for Predicting Reference Crop Evapotranspiration Through Tree-based Machine Learning and Bayesian Optimisation

Reference crop evapotranspiration (ETO) is a basic component of the hydrological cycle, and its estimation is critical for agricultural water resource management and scheduling. In this study, three tree-based machine learning algorithms (random forest [RF], gradient boosting decision tree [GBDT], and extreme gradient boosting [XGBoost]) were adopted to determine the essential factors for ETO prediction. The tree-based models were optimised using the Bayesian optimisation (BO) algorithm, and they were compared with three standalone models in terms of daily ETO and monthly mean ETO estimation in North China, with different input combinations of essential variables. The results indicated that solar radiation (Rs) and air temperature (Ts), including the maximum, minimum, and average temperature, in daily ETO were the key parameters affecting model prediction accuracy. Rs was the most influential factor in the monthly average ETO model, followed by Ts. Both relative humidity (RH) and wind speed at 2 m (U2) had little impact on ETO prediction at different scales, although their importance differed. Compared with the GBDT and RF models, the XGBoost model exhibited the highest performance for daily ETO and monthly mean ETO estimation. The hybrid tree-based models with the BO algorithm outperformed the standalone tree-based models. Overall, compared with other inputs, the model with three inputs (Rs, Ts, and RH/U2) had the highest accuracy. The BO-XGBoost model exhibited superior performance in terms of the global performance index (GPI) for daily ETO and monthly mean ETO prediction and it is recommended as a more accurate model predicting daily ETO and monthly mean ETO in North China or areas with a similar climate.

Splitting and Length of Years for Improving Tree-Based Models to Predict Reference Crop Evapotranspiration in the Humid Regions of China

To improve the accuracy of estimating reference crop evapotranspiration for the efficient management of water resources and the optimal design of irrigation scheduling, the drawback of the traditional FAO-56 Penman–Monteith method requiring complete meteorological input variables needs to be overcome. This study evaluates the effects of using five data splitting strategies and three different time lengths of input datasets on predicting ET0. The random forest (RF) and extreme gradient boosting (XGB) models coupled with a K-fold cross-validation approach were applied to accomplish this objective. The results showed that the accuracy of the RF (R2 = 0.862, RMSE = 0.528, MAE = 0.383, NSE = 0.854) was overall better than that of XGB (R2 = 0.867, RMSE = 0.517, MAE = 0.377, NSE = 0.860) in different input parameters. Both the RF and XGB models with the combination of Tmax, Tmin, and Rs as inputs provided better accuracy on daily ET0 estimation than the corresponding models with other input combinations. Among all the data splitting strategies, S5 (with a 9:1 proportion) showed the optimal performance. Compared with the length of 30 years, the estimation accuracy of the 50-year length with limited data was reduced, while the length of meteorological data of 10 years improved the accuracy in southern China. Nevertheless, the performance of the 10-year data was the worst among the three time spans when considering the independent test. Therefore, to improve the daily ET0 predicting performance of the tree-based models in humid regions of China, the random forest model with datasets of 30 years and the 9:1 data splitting strategy is recommended.

The Study of Drought in Future Climate Scenarios in the Huang-Huai-Hai Region

In the context of global warming, agricultural production and social and economic development are significantly affected by drought. The future change of climate conditions is uncertain; thus, it is of great importance to clarify the aspects of drought in order to define local and regional drought adaptation strategies. In this study, the meteorological data from 1976 to 2005 was used as a historical reference, and nine Global Climate Models (GCMs), downscaling to meteorological stations from 2039 to 2089, were used as future climate data. Based on Penman–Monteith, the reference crop Evapotranspiration (ET0) and Standardized Precipitation Evapotranspiration Index (SPEI) of the reference crop in three emission scenarios of RCP2.6, RCP4.5, and RCP8.5, under future climate conditions, were calculated. A non-parameter Mann–Kendall trend test was performed on temperature, precipitation, ET0, and SPEI to analyze the drought spatiotemporal distribution traits under upcoming climate scenarios. The results showed that, under future climate conditions, SPEI values in most areas of the Huang-Huai-Hai region would continuously increase year by year, and drought would be alleviated to some extent at the same pace. However, with the increase of greenhouse gas concentration in the emission scenarios, SPEI values continued to decline. In the RCP8.5 scenario, the area of severe drought was large. To sum up, in the future climate scenario, the degree of drought in the Huang-Huai-Hai region will be alleviated to some extent with the increase of rainfall, but with the increase of greenhouse gas concentration, the degree of drought will be further intensified, posing a huge challenge to agricultural water use in the region. This study provides a theoretical foundation for alleviating drought in the Huang-Huai-Hai region in future climate scenarios.