Abstract. Floods are one of the most common and widespread disasters in India, with an estimated 40Mha of land prone to this natural disaster (National Flood Commission, India). Significant loss of property, infrastructure, livestock, public utilities resulting in large economic losses due to floods are recurrent every year in many parts of India. Flood forecasting and early warning is widely recognized and adopted as non-structural measure to lower the damages caused by the flood events. Estimating the rainfall excess that results into excessive river flow is preliminary effort in riverine flood estimation. Flood forecasting models are in general, are event based and do not fully account for successive and persistent excessive surface runoff conditions. Successive high rainfall events result in saturated soil moisture conditions, favourable for high surface runoff conditions. The present study is to explore the usefulness of hydrological model derived surface runoff, running on continuous times-step, to relate to the occurrence of flood inundation due to persistent and successive high surface runoff conditions. Variable Infiltration Capacity (VIC), a macro-scale hydrological model, was used to simulate daily runoff at systematic grid level incorporating daily meteorological data and land cover data. VIC is a physically based, semi-distributed macroscale hydrological model that represents surface and subsurface hydrologic process on spatially distributed grid cell. It explicitly represents sub-grid heterogeneity in land cover classes, taking their phenological changes into account. In this study, the model was setup for entire India using geo-spatial data available from multiple sources (NRSC, NBSS&LUP, NOAA, and IMD) and was calibrated with river discharge data from CWC at selected river basins. Using the grid-wise surface runoff estimates from the model, an algorithm was developed through a set of thresholds of successive high runoff values in order to identify grids/locations with probable flooding conditions. These thresholds were refined through iterative process by comparing with satellite data derived flood maps of 2013 and 2014 monsoon season over India. India encountered many cyclonic flood events during Oct–Dec 2013, among which Phailin, Lehar, and Madi were rated to be very severe cyclonic storm. The path and intensity of these cyclonic events was very well captured by the model and areas were marked with persistent coverage of high runoff risk/flooded area. These thresholds were used to monitor floods in Jammu Kashmir during 4–5 Sep and Odisha during 8–9 Aug, 2014. The analysis indicated the need to vary the thresholds across space considering the terrain and geographical conditions. With respect to this a sub-basin wise study was made based on terrain characteristics (slope, elevation) using Aster DEM. It was found that basins with higher elevation represent higher thresholds as compared to basins with lesser elevation. The results show very promising correlation with the satellite derived flood maps. Further refinement and optimization of thresholds, varying them spatially accounting for topographic/terrain conditions, would lead to estimation of high runoff/flood risk areas for both riverine and drainage congested areas. Use of weather forecast data (NCMWRF, (GEFS/R)), etc. would enhance the scope to develop early warning systems.