Developing water level maps in the joint probability zone influenced by extreme tide and rainfall events

Abstract

Flood risk in estuarine regions can be difficult to estimate due to the relationship between extreme ocean levels and catchment generated streamflow. One method for estimating flood risk from two extremes is the design variable method, but this method is typically applied at an individual location along the river reach, so that, it is unclear how feasibly the method can be implemented with output from two dimensional hydraulic models. Adapting the design variable method to a spatial setting presents interesting challenges in terms of computational demand, how to incorporate regions outside the main channel and whether or not the method can be applied independently across grid cell locations. To test these issues, a case study is presented for the Nambucca catchment in the Mid North Coast region, NSW. At this location 144 water level maps were used, based on TUFLOW model output, corresponding to combinations of 12 storm tide return-levels and 12 rainfall event return-levels, from 63%AEP to 0.05%AEP. The design variable method was modified to accommodate partially wet grid cells and applied repeatedly and independently to each grid cell location. The two-dimensional design variable method is computationally intensive but yielded good results for this case study, showing considerable potential for wider application.