Water Resources Engineering connects hydrology and hydraulics with global, economic, environmental, and societal issues. Our Student Adam Scicchitano makes this connection here regarding the inherent uncertainty in making predictions and the real world consequences of ignoring uncertainty.
The article entitled “Who Decides? Forecasts and Responsibilities in the 1997 Red River Flood” was reported in the Applied Behavioral Science Review in the 1999 (issue 7-2). This paper relates to the Water Resources Engineering domain of hydrology, specifically predicting the occurrence and magnitude of floods. The article, while not directly a current event, demonstrates what is at stake as predictions are made using hydrologic data, which is not only a current concern as spring approaches, but also something all in the WRE field should bear in mind in the future. In summary, the article describes the Red River flood of 1997 in Grand Forks, ND. Flood evacuation decisions were made by policy makers based on a National Weather Service prediction of a 49’ river flood stage and a levee height of 51’. What was not taken into account with this prediction was the margin of error, which would have been around 10%. The river crested at 54’ leading to widespread devastation in Grand Fork ND on the order of $1-2 billion. It was concluded in the aftermath of the flood that the NWS needed to better understand the uncertainty inherent in its forecasts; this information has value to decision makers. In this case misuse of a prediction lead to more damage than if there were no prediction at all. Based on my engineering education, I believe that WRE facts presented in this article are sound. I realize that whenever I take a measurement or produce or work with data, that that data has limitations and, especially with complex models, a slight difference in initial conditions can yield very different results. The article did a good job at looking at what went wrong in this situation and how it might be fixed, but it failed to mention how these lessons could be translated to other predictions made using WRE data in similar situations.
Water is singularly important to everyone on earth, therefore water resources engineering by definition effects individuals, societies, economies, and environments around the planet. As I write this I’m looking out my window at three-four feet of snow piled everywhere; this water storage will, hopefully, melt in a couple months and the stored water will then become storm water, ground water, and overland flow. This article demonstrates how miscommunications with respect to hydrologic data can severely impact people’s lives. Flooding takes more human lives than any other natural disaster Takeuchi (2002), therefore it is important that we as WRE professionals do the very best we can to help protect people.
Figure. 1 Downtown building destroyed by fire during the flood. Photo credit: Grand Fork Herald.
Figure. 2 “Sorlie bridge 1997”. Licensed under Public Domain via Wikimedia Commons
Wurbs, R.A., James, W.P. Water Resources Engineering. Prentice Hall, Upper Saddle River, NJ, 2002.
Who Decides? Forecasts and Responsibilities in the 1997 Red River Flood. Roger A. Pielke, Jr. Applied Behavioral Science Review, 1999: 7(2), 83-101
Takeuchi, K. Floods and Society: a Never-Ending Evolutional Relation, in Flood Defence 2002, edited by B. Wu, Z. Wang, G. Wang, G. Huang, H. Fang and J. Huang, pp. 15-22, Science Press, New York Ltd, New York, 2002.