2009 – Loss of Life Estimation for Dam Safety Risk Analysis

Simon Lang

“The move to a risk-based approach to the management of dam safety requires robust estimates of the consequences of failure, and particularly the potential loss of life.” (Hill et al. 2007) In Australia to date, the empirical method developed by Graham (1999) is the most widely applied approach for estimating loss of life from dambreak flooding. However, as the move to risk-based approaches of dam safety management has gathered momentum internationally, increasingly sophisticated techniques for estimating loss of life have emerged. For example, Utah State University has developed the LIFESim model (Aboelata et al. 2002, 2003, 2004) and BC Hydro the Life Safety Model (Johnstone et al. 2003, 2005), while the United States Army Corps of Engineers have incorporated a simplified version of LIFESim into a software package they use to simulate the impacts of dambreak flooding (HEC-FIA). One advantage of the LIFESim, LSM and HEC-FIA models is that they can be used to estimate loss of life attributable to both natural and dambreak flooding. These models, along with empirical methods developed by Graham (2004, 2006), HR Wallingford (Pennning-Roswell et al. 2005, Priest et al. 2007) and Jonkman (2007) for estimating loss of life from flooding are reviewed in this paper, with an eye to their applicability in Australian contexts. This research was conducted with support from the 2009 ANCOLD travel bursary for young professionals.

Keywords: loss of life, dam safety risk analysis.