Damaging debris flows and other large erosion events are hazards that often emerge in mountainous landscapes due to the combination of fire disturbance and intense rainfall. Quantifying the water quality risk associated with these hazards is a complex task requiring deterministic catchment response models in combination with models that represent the stochastic conditioning by fire disturbance and storms in space and time. This presentation summarises three years of Bushfire CRC research where modeling and intensive data collection were combined in order to develop operational and strategic management tools for evaluating water quality risk. Operational land managers are typically concerned with spatial variation in risk as a function of fire severity, recovery and catchment properties. In this context our research has aimed to identify the key landscape variables that can be used to quantify spatial variation in risk across large catchment where severity, soil properties and topography are often highly variable. At a strategic level, the land managers want to evaluate risk as a function of fire and rainfall regimes that change over time due to climate change and prescribed burning (i.e. risk is non-stationary). In this context our research has aimed to isolate the first-order effects of fire and rainfall regimes on the frequency of significant water quality impacts. This was achieved by modeling fire disturbance and storms as random processes (or patches) that overlap in space and time. In this type of model the risk responds directly to changes in fire and rainfall regimes, thus eliminating some of the uncertainties associated with the exact response of any particular hillslope or catchment. The presentation concludes with examples of management scenarios where operational and strategic modeling tools can be applied to evaluate the water quality risk due to fire disturbance.