The present study deals with the characterization of subsurface coal fires of East Basuria colliery in Jharia coal field, India using tilt derivative and downward continuation of magnetic data. Magnetic data processing methods such as diurnal correction, noise removal, reduction to pole, tilt derivative and downward continuation have been used to process the data and for the interpretation of results on the basis of magnetic properties of overlying materials which change with the temperature variation above or below the Curie temperature. Most of the magnetic anomalies are associated with coal fire and non-coal fire regions which are correlated with tilt-derivative anomaly and corresponding downward-continued anomaly at different depths. The subsequent surface and subsurface characteristics are explained with good agreement. Approximate source depth of principal anomaly inferred from tilt derivatives method are corroborated with multi-seam occurrences, mine working levels and surface manifestation which are also correlated well with 3D model of downward continued anomaly distribution.