The field of environmental prediction is witnessing a significant shift towards the adoption of physics-informed modeling techniques. These methods, which integrate physical laws and constraints into machine learning frameworks, have shown remarkable promise in improving the accuracy and reliability of predictions. Recent research has focused on developing novel architectures, such as hybrid LSTM-PINN models, that can effectively capture complex dynamics and nonlinear relationships in environmental systems. Notable applications include demographic forecasting, precipitation nowcasting, and carbon flux prediction. The use of physics-informed neural networks (PINNs) has also enabled the development of more accurate and efficient models for tasks such as climate emulation and weather forecasting. Furthermore, researchers have explored the potential of generative models, such as Spatiotemporal Pyramid Flows, to transform the way we approach climate modeling and prediction. Overall, the trend towards physics-informed modeling is expected to continue, with potential applications in a wide range of environmental domains. Some noteworthy papers in this area include: Forecasting India's Demographic Transition Under Fertility Policy Scenarios Using hybrid LSTM-PINN Model, which demonstrates the effectiveness of hybrid models in demographic forecasting. PIANO: Physics-informed Dual Neural Operator for Precipitation Nowcasting, which shows the potential of PINNs in precipitation nowcasting. Spatiotemporal Pyramid Flow Matching for Climate Emulation, which introduces a novel approach to climate emulation using generative models.