For many basins, identifying changes to water quality over time and understanding current hydrologic processes are hindered by fragmented and discontinuous water-quality and hydrology data. In the coal mined region of the New River basin and Indian Fork sub-basin, muted and pronounced changes, respectively, to concentration–discharge (C–Q) relationships were identified using linear regression on log-transformed historical (1970s–1980s) and recent (2000s) water-quality and streamflow data. Changes to C–Q relationships were related to coal mining histories and shifts in land use. Hysteresis plots of individual storms from 2007 (New River) and the fall of 2009 (Indian Fork) were used to understand current hydrologic processes in the basins. In the New River, storm magnitude was found to be closely related to the reversal of loop rotation in hysteresis plots; a peak-flow threshold of 25 cubic meters per second (m3/s) segregates hysteresis patterns into clockwise and counterclockwise rotational groups. Small storms with peak flow less than 25 m3/s often resulted in dilution of constituent concentrations in headwater tributaries like Indian Fork and concentration of constituents downstream in the mainstem of the New River. Conceptual two or three component mixing models for the basins were used to infer the influence of water derived from spoil material on water quality. Copyright © 2012 John Wiley & Sons, Ltd.