The reliable operation of nuclear power plants necessitates continuous access to precise meteorological data, as adverse weather conditions can significantly impact their efficiency and safety. However, missing or incomplete meteorological records present a critical challenge in such high-stakes applications. This study explores machine learning (ML)-based approaches for imputing missing meteorological data, specifically focusing on estimating Snowpack Depth (cm)—a key parameter in environmental and operational planning for nuclear facilities. Using a robust dataset comprising predictors such as air temperature, wind speed, humidity, pressure, solar radiation, and others, we compare the performance of Linear Regression (LR), Random Forest (RF), and XGBoost Regressor models. Furthermore, we integrate Explainable AI (XAI) techniques to interpret model decisions and identify the most influential predictors for accurate estimation. The results demonstrate that ensemble-based methods, particularly XGBoost, outperform traditional models in both accuracy and generalizability. This research not only enhances the reliability of environmental assessments in nuclear energy applications but also proposes an optimum feature set for future predictive modeling efforts in similar domains. KEYWORDS: Nuclear Energy, Meteorological Data Imputation, Machine Learning, Snowpack Prediction, XGBoost Regressor, Random Forest, Linear Regression, Explainable AI, Environmental Monitoring, Energy Infrastructure.