The writers’ main purpose was to simultaneously forecast different types of drought. This was first done using different time windows of the MSPI index. Research was conducted to find answers to the following questions: (1) What types of droughts can be predicted simultaneously with relatively good accuracy; and (2) What types of droughts can be predicted simultaneously with low accuracy. In this study, artificial intelligence (AI) models were used to answer these questions. The writers believe that the details of artificial intelligence models and sophisticated tools are not so important to hydrologists as to be published in the Journal of Hydrologic Engineering. The discussers divert the reader from themain concepts of our research. Numerous AI modeling studies have not mentioned details (e.g., model parameters) in order not to deviate from the main conceptual subject—for example, Emadi et al. (2021) on evaporation modeling, Fang et al. (2018) on evapotranspiration modeling, Zhang et al. (2017) on drought prediction, and many other studies (Aghelpour and Varshavian 2021; Aghelpour et al. 2019, 2020a, b, 2021b; Ahmadi et al. 2021; Graf and Aghelpour 2021; Guan et al. 2020; Islam et al. 2021; Karbasi 2018; Mehdizadeh et al. 2020; Mohammadi et al. 2020, 2021; Pham et al. 2021). These studies used AI models to solve hydrological problems. They provided the model accuracy in the training and testing stages, but avoided information about the models used (e.g., type and number of model parameters). Zeynoddin et al. (2019) used multilayer perceptron neural networks (MLPNNs) in soil temperature forecasting. The discussers did not provide sufficient detail about the parameters of the developed models.
