The impacts of climate and land use change are increasing the frequency and magnitude of rainfall-triggered landslides, debris flows, and hillslope erosion hazards in regions of the world that have already experienced increasing levels of impact and disaster risk. Understanding these hazards and their interactions requires recognizing the interrelationships of the catchment's physical characteristics such as topography, hydrology, soil properties, and land with climate variables. Physically-based distributed multi-hazard models integrate these relationships; however, these models require a large number of input parameters that are challenging to obtain in catchments where no data are available to represent all relevant catchment physical characteristics. For regions with the highest levels of hazards and disaster risks, this type of model is associated with significant challenges related to data scarcity, uncertainty, model complexity, and possible over-parameterisation. Uncertainties arise due to the quality of the available data and the accessibility of different spatial resolution data to accurately represent these hazards within different catchment scales. This thesis addresses such uncertainties by developing a new modelling workflow that enables physically-based distributed multi-hazard models to be applied in data-scarce regions. Using this workflow, model parameterisation and uncertainty management were addressed to explore climate and land use scenarios in the two case study sites proposed in this thesis (Soufriere catchment, Saint Lucia, and The Maipo sub-catchment, Chile) to demonstrate the utility of this approach for informing resilient land use planning and policy. Applying the workflow to the two selected study sites identified the parameter-set values for land use and soil types that best approximated the spatial representation of rainfall-triggered landslides, debris flows, and hillslope erosion hazards for registered rainfall events, allowing the exploration of climate and land use scenarios at both study sites. This thesis has contributed by introducing a systematic modelling workflow that addresses the uncertainties in multi-hazard modelling, thereby improving the representation of hillslope hydrological hazard interactions for catchments with data scarcity.