Climate change poses a significant global threat, impacting ecosystems, economies, and human societies. Accurate and timely prediction models are essential for mitigating its effects and supporting sustainable development. Traditional climate models often struggle with the complexity and scale of climate systems. In contrast, Artificial Intelligence (AI), especially Machine Learning (ML) and Deep Learning (DL) techniques, offers powerful tools to enhance climate change prediction by analyzing vast, complex, and nonlinear datasets. This research explores the role of AI in climate modeling, leveraging data from historical climate records, satellite imagery, and environmental sensors. Advanced AI architectures such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), including Long ShortTerm Memory (LSTM) models, are utilized for recognizing spatial and temporal patterns, significantly improving forecasting accuracy. AI also enables the simulation of future climate scenarios under varying emission pathways, aiding policy decisions and risk assessment. Despite its promise, challenges such as data quality, model interpretability, computational demands, and algorithmic bias remain. Addressing these issues requires interdisciplinary collaboration across climate science and AI domains. This paper reviews recent advances, current challenges, and future directions in applying AI to climate change prediction, emphasizing the potential of AI to create more adaptive, precise, and actionable climate models.
Datasets, Sensors, Convolutional Neural Networks (CNNs), Recurrent Neural Networks(RNNs), data biases, robustness, Long Short-Term Memory (LSTM).