2010 Maine Water Conference
Wednesday, March 17, 2010
Augusta Civic Center, Augusta, Maine

SESSION F.
Historical and Future Extreme Hydrologic Events in Maine
Chair: Robert M. Lent, Glenn Hodgkins, USGS, Augusta, ME

Description: Maine has a long history of extreme hydrologic events, including major floods in 1936, 1987, and 2008, and droughts in the 1960’s and early 2000’s. There is a considerable amount known about these historical events and there is some information available on future projected events. What do historical records tell us about the frequency and severity of floods and droughts in Maine? Have they changed over time? What can we say about the potential for future floods and droughts and what methods and models can we use to predict them? How much uncertainty is associated with the predictions? What have been the impacts of historical extreme events and what will be the impacts of future extreme events? This Maine Water Conference session will attempt to answer many of these questions.

ABSTRACTS

• Establishing Maine Coastal Flood Zones in a Time of Rising Sea Level
  Robert G. Gerber, P.E., C.G., VP Environmental Engineering Sebago Technics Inc.
  
  In the summer of 2009, FEMA released new provisional coastal zone flood maps for York and Cumberland Counties.  In most exposed coastal locations, the new maps will result in significant increases in the elevation of the high velocity zones (V-zones) where waves in water depths of greater than 3 feet under the 100-year storm condition have high energy and can cause damage to man-made structures.  The new flood maps implement two major improvements in coastal flood zone mapping:  1) the use of LIDAR to produce much more accurate topographic maps than have been available heretofore; and 2) the use of new FEMA modeling methodologies explained in major guidance manuals released in 2003, 2007, and 2008.  What is NOT included, however, is any allowance for future sea level rise or possible increase in storm intensity.  Given that it has been 20 years since the last maps were updated and it is likely to be 20 more before the next generation of maps are released, there is a need to make these maps more dynamic and adjustable year-to-year with minimal effort to determine which properties are eligible for low-cost federal flood insurance.  This presentation will present examples from wave modeling of bathymetric/topographic conditions that are most vulnerable to V-zone category wave action as sea level rises, and will explore data needs and means of adjusting coastal flood zone designations and elevations to adjust to rise in sea level as it occurs.
  
  
• Recent Floods in maine: A Comparison of Causes, Regions of Influence and Exceedance Probabilities
  Pam Lombard, USGS Maine Water Science Center
  

  Three floods with annual exceedance probabilities less than 1 percent have occurred in Maine in the last three years.  Reports describing the Mother’s Day storm of May 2006 and the Patriots’ Day storm of April 2007 floods in southern Maine; and the April/May 2008 flood in northern Maine were written in cooperation with the Federal Emergency Management Agency.  The documentation of peak stream elevations and peak streamflow magnitudes and exceedance probabilities provides essential information for the delineation of flood plains and for flood-mitigation decisions by local, State, and Federal emergency management officials.  The three recent floods will be compared with regards to the antecedent conditions and hydrologic events that caused them, the regions that they affected, and how they fit into historical context. 

• Historical changes in annual peak flows in Maine and implications for flood frequency analyses
  Glenn A. Hodgkins, USGS Maine Water Science Center
  
  Flood frequency analyses use statistical methods to compute peak streamflows of selected recurrence intervals (such as the 100-year peak flow). These peak flows are used for bridge design and floodplain management. It has long been assumed that peak streamflows are stationary (non-changing) over very long periods of time, for relatively undeveloped basins. This assumption has recently been questioned because of the potential influence of global warming on peak flows. Maine has many streamflow gaging stations with 50-105 years of recorded annual peak streamflows. This allowed the testing of historical flood frequency stationarity.
  Changes over time in annual peak streamflows were analyzed for 28 USGS streamflow-gaging stations in Maine that had at least 50 years of data. The median peak-flow change over time was an increase of 15.9 percent based on a linear change and 12.4 percent based on a step change around 1970. Peak flows with 100-year and 5-year recurrence intervals were computed for the 28 stations using the full annual peak-flow record and selected intervals of that record. The 1967-1996 interval had the largest 100-year and 5-year peak flows overall when compared to estimates based on the full record; the median difference for both was +8 percent. The 1977-2006 and 1971-2006 intervals also showed overall larger flows, but not as large as the 1967-1996 interval. Is it better to use the full peak-flow record or recent intervals when computing peak flows for selected recurrence intervals? One possible conservative approach is compute both and choose the higher value.
  
  
• Trends in New England Flood Magnitudes and Frequencies Revealed by Partial Duration Flood Series Analyses of Long Gauge Records
  William H. Armstrong, Boston College, Dept. of Geology & Geophysics; Mathias J. Collins, NOAA; Noah P Snyder, Boston College
  

  Recent studies document increasing precipitation and streamflow in the Northeastern United States throughout the 20th-21st centuries.  Low flow quantiles like the annual minimum and median flows have increased significantly over the period on many New England rivers with dominantly natural streamflow, as have annual peak discharges—especially for smaller, more frequent floods.  In order to better investigate the high frequency flooding trends (events occurring at least once every five years), this study analyzes the partial duration flood series for 23 of these New England rivers.  The study rivers have continuous records through 2006 and an average period of record of around 70 years, with a minimum of 59 years and a maximum of 81 years.  Eighteen rivers show positive trends in flood magnitude using the Mann-Kendall non-parametric trend test.  Six of these trends are significant at p<0.1.  We also investigate a potential hydroclimatic shift in the region around 1970.  Twenty-two rivers show increased numbers of partial duration series floods per year in the post-1970 period when comparing pre- and post-1970 records using the non-parametric Wilcoxon rank sum test.  Thirteen of these increases are significant at p<0.1.  On average, each river has approximately one more flood per year above the gauge threshold value.  Because frequent floods are important channel-forming flows, these results have significant implications for channel and floodplain geomorphology, as well as aquatic habitat.