Evaluation of several soft computing methods in monthly evapotranspiration modelling

Abstract

Evapotranspiration assessment is one of the most substantial issues in hydrology. The methods used in modelling reference evapotranspiration (ET 0 ) consist of empirical equations or complex methods based on physical processes. In arid and semi-arid climates, determining the amount of evapotranspiration has a major role in the design of irrigation systems, irrigation network management, planning and management of water resources and water management issues in the agricultural sector. This paper presents a case study of five meteorological stations located in Kurdistan province in the west of Iran. The ability of three different soft computing methods, an artificial neural network (ANN), an adaptive neuro-fuzzy inference system (ANFIS) and gene expression programming (GEP), were compared for modelling ET 0 in this study. The FAO56 Penman−Monteith model was considered as a reference model and soft computing models were compared using the Priestley−Taylor, Hargreaves, Hargreaves−Samani, Makkink and Makkink−Hansen empirical methods, with respect to the determination co-efficient, the root mean square error, the mean absolute error and the Nash-Sutcliffe model efficiency co-efficient. Soft computing models were superior to the empirical methods in modelling ET 0 . Among the soft computing methods, the ANN was found to be better than the ANFIS and GEP.

Using data‐driven methods to explore the predictability of surface soil moisture with FLUXNET site data

“…In recent years, machine-learning methods have been successfully applied to the estimation of meteorological and hydrological variables, such as air temperature (Cobaner, Citakoglu, Kisi, & Haktanir, 2014;Kisi & Shiri, 2014;Kisi & Sanikhani, 2015a;), solar radiation (Mohammadi, Shamshirband, Kamsin, Lai, & Mansor, 2016), dew point temperature (Kisi, Kim, & Shiri, 2013), soil temperature (Hadi, Deo, Mundher, Okan, & Ozgur, 2018;Kim & Singh, 2014), precipitation (Kisi & Sanikhani, 2015b), and evapotranspiration (Feng, Cui, Zhao, Hu, & Gong, 2016;Gavili, Sanikhani, Kisi, & Mahmoudi, 2017;Sanikhani, Kisi, Maroufpoor, & Yaseen, 2018). A few studies have aimed to predict SM using machine-learning methods.…”

Comprehensive assessment of 12 soft computing approaches for modelling reference evapotranspiration in humid locations

“…Wang et al (2017c) compared the FG, least square support vector regression (LSSVR), M5Tree, MARS and multi-variate linear regression (MLR) models to analyse climatic factors on the pan evaporation modelling process and introduced the LSSVR and FG models as superior models over the rest. Gavili et al (2018) applied the ANN, ANFIS and GEP to model ET o in arid and semi-arid climates. Sanikhani et al (2018) investigated six artificial intelligence models (multi-layer perception neural network (MLPNN), GRNN, radial-basis neural network (RBNN), ANFIS-GP, ANFIS-SC and GEP) to model ET o in arid areas.…”

Modelling daily soil temperature at different depths via the classical and hybrid models

“…Over the last years, the AI-based techniques have been proposed in various areas of research, such as meteorological, hydrological and soil sciences (Tabari et al, 2011;Talaee, 2014;Aitkenhead and Coull, 2016;Liu et al, 2016;Citakoglu, 2017;Mehdizadeh et al, 2017aMehdizadeh et al, , 2018aGavili et al, 2018;Massawe et al, 2018;Singh et al, 2018;Maroufpoor et al, 2019;Azad et al, 2020;Mehdizadeh, 2020). For example, the AI approaches were applied successfully by Mehdizadeh (2018a), Mehdizadeh et al (2017b) and Azad et al (2020) to estimate the dew point and air temperatures.…”

“…For example, the AI approaches were applied successfully by Mehdizadeh (2018a), Mehdizadeh et al (2017b) and Azad et al (2020) to estimate the dew point and air temperatures. In another works, the potential of AI models was verified to estimate the wind speed and evapotranspiration time series (Gavili et al, 2018;Maroufpoor et al, 2019;Mohammadi and Mehdizadeh, 2020).…”