Multi-scenario risk assessment: an eighteen-hydrograph framework for dam spillway safety using copulas and integrating climate change

Abstract

ABSTRACT This study proposes a procedure for dam spillway safety verification based on a set of design hydrographs and using advancements in statistical hydrology published in the last years. The approach uses the following steps: (1) using copulas to model the dependence between peak flow and flood volume; (2) selecting representative hydrographs and automatically defining their start, end, and base flow; (3) selecting three main events of the flow rate daily recorded series (4) selecting three pairs of peak flow and hydrograph volume with joint return period of 10,000 years (5) fitting hydrograph shapes to these variables using Probability Density Functions (PDFs) (6) creating nine design hydrographs based on observed series; (7) integrating Regional Climate Model (RCM) data to assess climate change impacts (8) repeating steps 1 to 6 and creating nine design hydrographs based on future flow rate series. Unlike conventional methods that focus solely on peak flow and one form of hydrograph, the method uses flood volume and peak flow via copula-based joint probability and pdf shape adjustment to the most critical events. Implemented at the Jurumirim Hydroelectric Plant, the method presented showed to be more conservative than the traditional one, even when considering only observed flow rate series.