Search

[Display omitted] • We built hybrid Long Short-Term Memory model for Pan Evaporation (E p). • We adopt Neighbourhood Component Analysis for feature selection. • We predict E p in drought-prone Queensland with satellite predictor variables. • Our NCA-LSTM model is compared with competitive benchmark models. • NCA-LSTM model is useful for estimation of evaporative water losses. Evaporation, as a core process within the global hydrological cycle, requires reliable methods to monitor its variation, for decision-making in agriculture, irrigation systems and dam operations, also in other areas of hydrology and water resource management. Accurate monitoring of pan evaporation (E p) is one the most popular approaches to understand the evaporative process. This work aims to construct a hybrid Long Short-Term Memory (LSTM) predictive model that is coupled with Neighbourhood Component Analysis for feature selection to predict E p in drought-prone regions in Queensland, Australia (Amberley, Gatton, Oakey, & Townsville). Utilizing the daily-scale dataset [31 August 2002 to 22 September 2020], the performance of the proposed deep learning (DL) hybrid model, denoted as NCA-LSTM, is compared with competitive benchmark models, i.e., standalone LSTM, other types of DL, single hidden layer neuronal architecture and decision tree-based method. The testing results reveal the lowest Relative Root Mean Square Error ≤ 20 % , Absolute Percentage Bias ≤ 14.5 % and the highest Kling-Gupta Efficiency ≥ 87 % attained by the NCA-LSTM hybrid model (relative to benchmark models) tested for Amberley, Gatton, and Oakey sites. In respect to the predictive efficiency, the proposed NCA-LSTM hybrid model, improved with feature selection, outperforms all benchmark models, indicating its future utility in the prediction of daily E p. In practical sense, the predictive model developed for E p estimation provides an accurate estimation of evaporative water loss in hydrological cycle and therefore, can be implemented in areas of irrigation management, planning of irrigation-based agriculture, and mitigation of financial losses to agricultural and related sectors where, regular monitoring and forecasting of water resources are a vital part of sustainable livelihood and business. [ABSTRACT FROM AUTHOR]

Modelling workflow for BRF-LSTM and BRF-GRU predictive models. [Display omitted] • Deep learning hybrid predictive model (BRF-LSTM) is built for streamflow forecasting. • Climate indices, rainfall and periodicity are factored for accurate predictions. • Adopts feature algorithm (i.e. Boruta) with long short-term memory and gated recurrent unit. • Six gauging sites tested; BRF-LSTM yields > 98% errors in ± 0.015 m, ~1.30% relative error. • The proposed model is useful for hydrological and strategic water resources planning. Long-term forecasting of any hydrologic phenomena is essential for strategic environmental planning, hydrologic and other forms of structural design, agriculture, and water resources management. Climate mode indices, utilising machine learning methods, are frequently considered as predictor variables in order to forecast several different hydrological variables. In this study, a feature selection algorithm based on two different deep learning models, i.e., long short-term memory and a gated recurrent unit, is applied to improve the forecasting capability of streamflow water levels at six gauging stations in the Murray Darling Basin of Australia. This paper therefore aggregates the significant antecedent lag memory of climate mode indices, rainfall, and the monthly factor based on the periodicity as the predictor variables to attain significantly accurate stream water level forecasts. This novel method identifies an improved relationship between the stream water level and climate mode indices through the aggregation of the significant lagged datasets capturing the historical features to predict the future streamflow water level. The boruta feature selection algorithm (BRF) was then applied in a two phase process before and after attaining the significant lagged inputs to screen the optimum predictor variables. The merits of the forecast models were evaluated through different performance evaluation criteria. The results show that the accumulated significant lagged inputs based on climate mode indices, along with the rainfall and periodicity factors are seen to provide improved forecasting of the SWL over the non-BRF deep learning approaches where no prior feature selection was applied. The hybrid LSTM method (i.e., BRF-LSTM model) achieved a unique advantage in terms of SWL forecasting, particularly attaining over 98% of the predictive errors lying within a band of +/-0.015 m with relatively low relative errors (RRMSE ≈1.30% and RMAE ≈ 0.882%), outperforming all of the benchmark models. It is also found that the periodicity factor has a potential influence on the accuracy of the forecast models for the four monitored study stations. This study concludes that the newly developed hybrid deep learning approaches, coupled with the BRF feature selection, provide improved forecasting performance. The hybrid approach developed in this paper can therefore be used to provide a strong provide predictive response algorithm for the hydrological variables that were influenced by the low-frequency variability of the climate model indices in respect to streamflow water level. [ABSTRACT FROM AUTHOR]