Table 2.

AI-driven climate models application and impacts on decision-making

S/N	Climate actions	Models used	Performance	Impacts on decision-making	Ref
1	Monthly rainfall	ANN and ANFIS	ANFIS showed higher accuracy	It aided the meteorological stations in weather prediction and planning	(Abebe and Endalie 2023)
2	Pollutant and particle levels	Support Vector Regression (SVR)	Pollutants like CO, O3, and SO2 were predicted at 94.1% accuracy	The Environmental Protection Agency was able to plan with the data	(Castelli et al. 2020)
3	Earthquake	ANN and Random Forest	ANN model produced better earthquake depth, acceleration, and velocity accuracy	It is used in planning for the natural disaster and its influence on the environment	(Essam et al. 2021)
4	Origin of trace pollutants	Long Short-Term Memory Network (LSTM)	It provides high prediction accuracy and traces the significant sources of the pollutant	The result was used to control pollutant discharge and enhance the water quality	(Wang et al. 2019)
5	Forest fire	Classification and Regression Tree (CART)	The slope showed a substantial influence on forest fire occurrence	It assists in preparation for the forest fire occurrence	(Piao et al. 2022)
6	Projection of flue gas from waste-to-energy	ANN	The mean square error of the model validation constraints is between 0.003 and 0.19	Findings from the study were used to make decisions in renewable energy planning	(Ma et al. 2023)
7	Quality of water for drinking	SVM and Logistic Regression	The models predicted the water quality at 98 and 92%, respectively	The data from this study assisted in expanding the scope of the task	(Panigrahi et al. 2023)
8	Changes in water level	Multi-layer Perception Neural Network (MLP-NN)	The models accurately predicted changes in water level with running time	It was assisted in planning for flooding	(Nur Adli Zakaria et al. 2021)