Predicting Modified Fournier Index by Using Artificial Neural Network in Central Europe

Endre Harsányi; Bashar Bashir; Firas Alsilibe; Muhammad Farhan Ul Moazzam; Tamás Ratonyi; Abdullah Alsalman; Adrienn Széles; Aniko Nyeki; István Takács; Safwan Mohammed

doi:10.3390/ijerph191710653

Predicting Modified Fournier Index by Using Artificial Neural Network in Central Europe

Int J Environ Res Public Health. 2022 Aug 26;19(17):10653. doi: 10.3390/ijerph191710653.

Authors

Affiliations

¹ Institute of Land Use, Technical and Precision Technology, Faculty of Agricultural and Food Sciences and Environmental Management, University of Debrecen, 4032 Debrecen, Hungary.
² Institutes for Agricultural Research and Educational Farm, University of Debrecen, Böszörményi 138, 4032 Debrecen, Hungary.
³ Department of Civil Engineering, College of Engineering, King Saud University, P.O. Box 800, Riyadh 11421, Saudi Arabia.
⁴ Department of Transport Infrastructure and Water Resources Engineering, Széchenyi István University, Egyetem tér 1, 9026 Gyor, Hungary.
⁵ Department of Civil Engineering, College of Ocean Science, Jeju National University, 102 Jejudaehakro, Jeju 63243, Korea.
⁶ Department of Biosystems and Food Engineering, Faculty of Agricultural and Food Sciences, Széchenyi István University, Vár Square 2, 9200 Mosonmagyarovar, Hungary.
⁷ Doctoral School of Humanities, University of Debrecen, Egyetem Tér 1, 4032 Debrecen, Hungary.

Abstract

The Modified Fournier Index (MFI) is one of the indices that can assess the erosivity of rainfall. However, the implementation of the artificial neural network (ANN) for the prediction of the MFI is still rare. In this research, climate data (monthly and yearly precipitation (p_i, P_total) (mm), daily maximum precipitation (P_d-max) (mm), monthly mean temperature (T_avg) (°C), daily maximum mean temperature (T_d-max) (°C), and daily minimum mean temperature (T_d-min) (°C)) were collected from three stations in Hungary (Budapest, Debrecen, and Pécs) between 1901 and 2020. The MFI was calculated, and then, the performance of two ANNs (multilayer perceptron (MLP) and radial basis function (RBF)) in predicting the MFI was evaluated under four scenarios. The average MFI values were between 66.30 ± 15.40 (low erosivity) in Debrecen and 75.39 ± 15.39 (low erosivity) in Pecs. The prediction of the MFI by using MLP was good (NSE_{Budapest(SC3)} = 0.71, NSE_Pécs(SC2) = 0.69). Additionally, the performance of RBF was accurate (NSE_{Debrecen(SC4)} = 0.68, NSE_Pécs(SC3) = 0.73). However, the correlation coefficient between the observed MFI and the predicted one ranged between 0.83 (Budapest (SC2-MLP)) and 0.86 (Pécs (SC3-RBF)). Interestingly, the statistical analyses promoted SC2 (P_d-max + p_i + P_total) and SC4 (P_total + T_avg + T_d-max + T_d-min) as the best scenarios for predicting MFI by using the ANN-MLP and ANN-RBF, respectively. However, the sensitivity analysis highlighted that P_total, p_i, and T_d-min had the highest relative importance in the prediction process. The output of this research promoted the ANN (MLP and RBF) as an effective tool for predicting rainfall erosivity in Central Europe.

Keywords: Hungary; climate change; land degradation; machine learning.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Europe
Hungary
Neural Networks, Computer*
Temperature

Grants and funding

Project no. TKP2021-NKTA-32 was implemented with the support provided by the National Research, Development, and Innovation Fund of Hungary, financed under the TKP2021-NKTA funding scheme. This research was supported by the RESEARCHERS SUPPORTING PROJECT, grant number RSP-2021/296, King Saud University, Riyadh, Saudi Arabia.