Nitrogen oxide (NOx) emissions from heavy-duty vehicles pose significant environmental and health risks, especially in urban regions with high traffic density. This study presents a comparative multi-strategy forecasting framework to forecast NOx emissions using real-world fleet data from Brazil, enriched with interpolated meteorological variables. A preprocessing pipeline integrates onboard sensor readings, trip metadata, and weather station data through temporal and spatial interpolation, including kriging. Four regression models: Linear Regression, Random Forest, XGBoost, and LSTM, were evaluated under single-step, recursive, multi-output, and direct per-horizon forecasting strategies. Results demonstrate that performance depends on forecasting strategy and vehicle type. The multi-output strategy provided the most balanced long-horizon performance for LSTM, where it achieved the lowest relative errors and competitive variance explanation. In its best configuration, LSTM reached a MAPE of 10.28% in single-step forecasting. Adjusted average speed emerged as the most relevant predictor, while meteorological factors improved predictive performance in selected settings. Unlike previous studies, this work pioneers the integration of continuous onboard fleet data in Brazil with spatio-temporal meteorological interpolation via kriging, this study also advances the scientific frontier by proposing a methodological framework for NOx forecasting and benchmarking under multi-horizon strategies in real-world conditions.