|
William R Blanton, Jr., Chief,
Engineering Management Branch,
Mitigation Directorate, Federal Management Agency,
500 C Street, S.W.
Washington, DC 20472
From: Clay Township Supervisor Jay M. DeBoyer and Clay Township Board of Trustees
RE: FEMA DOCKET # FEMA-B-1016 PROTEST LETTER
May 4, 2009
This letter is intended to file Clay Township’s official protest/appeal upon the Base Flood Elevation increase sought by Federal Emergency Management Agency referenced by the above docket number. It also is a letter that has been referenced by a large number of effected residents as to their scientific and technical data.
The study in question is based on datasets, some of which contain incomplete and unverifiable information, from areas that are remote from the principal regions of impact in Anchor Bay and utilizes predictive models, the accuracy of which have since been questioned. FEMA also fails to adequately inform the public of the impact of the conversion from NGVD29 to NAVD88. FEMA applies the conclusions of it’s study to areas entirely outside the scope of the study area without any scientific or technical basis. The Study arbitrarily uses a “sustained 40 mph wind” velocity as a critical basis for it’s modeling conclusions but ignores the actual historic wind data that is readily available for such modeling.
As required per your protest/appeals process we must base our argument on scientific and/or technical data. That being said, first let’s examine the scientific results we must refute. Our references below relate to the only document provided by FEMA in support of this initiative, titled; FLOOD LEVEL RESTUDY of LAKE ST. CLAIR AND ANCHOR BAY
IN MACOMB AND ST. CLAIR COUNTIES, MICHIGAN.
……..page 9 table 2-Revised Flood Levels for Lake St. Clair (Reach AA)
100year 176.30m 578.6feet
This increase amounts to .53 feet.
……...page 10 table 3- Difference Between Revised 2005 and 1988 Open Coast Report
100 year-0.4 feet
This is a net decrease for 1988 to 2005 reports
3.2.1 Still Water Flood Levels
This section addresses how probabilistic still water flood levels were determined. Empirical data in support of conclusions reached is totally absent.
……..page 11, -3.2.1 “Details on the actual method used were not documented in the report”
……..page 12, “the data from the North Channel gage could not be reliably reconstructed.”
These statements disqualify the data collected in that they lack the basic requirements of scientific method, which is, controlled analysis capable of reproducibility and repeatability and complete documentation in support of conclusions reached.
The Flood Level Restudy relies primarily on the projections of the HYDSTAT program for the probability predictions. This program has been determined by the scientific community to be flawed.
Below is a quote from the NOAA website under heading GLERL Publication Abstract: addressing the statistical program specifically.
“As part of the International Joint Commission (IJC) Levels Reference Study, seasonal flood risk levels were developed for selected U.S. and Canadian sites on the shores of the Great Lakes. Flood risk levels were computed through the convolution of frequency distributions which best fit the maximum seasonal monthly static water levels (lake level) for a given regulation scenario and the maximum seasonal historic storm rises (or storm surges) recorded at the site. The Ontario Ministry of Natural Resources computer program HYDSTAT, was used initially for the combined frequency analysis. It was found in HYDSTAT that the selection criteria for the "best-fit" distributions were biased and the refitting procedures were unjustified.”
This clearly disqualifies data that was compiled using the HYDSTAT statistical program on the basis of bias and unjustifiable procedures within the program.
Table 4, pg 12, clearly shows a .71 foot decrease in comparison to the 1989 study and below the table it states “the data from the North Channel gage could not be reliably reconstructed.”
Again this statement disqualifies the dataset collect by lacking the basic requirement of scientific compilation, which is, controlled methodology to be able to repeat and prove results.
…….page 13, second paragraph it states “were based on waves developed by a 40 mile per hour (mph) sustained wind.” “Sustained” is not defined, is it 1 hour, 6 hours, 24 hours etc. ? Also this modeling has not been compared to actual wind data that is available as a means of comparison. As recent as December 28, 2008 a potential “sustained” wind from a southwesterly direction of a speed in the vicinity of 40 mph over Lake St Clair occurred and the effect of this wind did not produce the results predicted by the STWAVE model. Back testing of model predictions is perhaps the best way to prove a model.
Tables 5-9 address the application of still water+wave setup. Transects 24-27 are the only ones in Clay Township. These transects show a net decrease in flood elevations and yet FEMA deems it necessary to apply a 6 inch BFE increase across the entire Township .in areas that are some 9 miles from the transects used and in the opposite direction from the wind effect area. Only a south, west or some combination there of could cause a wave setup that would effect Clay Township and if that occurred it would raise water in NW Clay Township only. Any other direction of wind would reduce water levels in Clay Township entirely. Imposing requirements on the entire township based on the results in a single very specific region is extremely irresponsible. This does not seem to be consistent with good, sound scientific practice. Again, there are no sampling stations in the region to provide necessary empirical information.
Wave run-up and overtopping are key components in FEMA’s maximum water height determination, however, throughout our research it is clear that these two components are much more important and effective when applied to sloping shoreline and much less effective when applied to blunt shore edges such as concrete walls or steel seawalls. Less than 1% of the shoreline this program was applied to consists of sloping shoreline. Further, the STWAVE program does not have a component to account for extensive backwash created by the breaking of the waves against the seawall created by a theoretical “sustained” 40 mph wind.
It seems that after the initial model results of the analysis were determined to be a net decrease in the BFE, that it was recommended “that wave run-up and overtopping be included in delineation of coastal flooding in Anchor Bay”. The resulting in BFE increases was applied in areas that would not and could not be effected by such wind circumstances. Painting with a broad brush only puts the findings, which are unarguably only a hypothesis at best, closer to an “ends justifies the means” study.
Some of the most critical statements about the study are from its own text. Page 19 under Summary and Recommendations shows the new data clearly should recommend a reduction in the BFE in the range of .2 to .6 foot and it is not until the new component of run-up and overtopping is added that an increase is justifiable. The study authors themselves express doubts concerning the data.
“More realistic run-up and overtopping elevations could be developed if actual wind data was used to develop a 1-D wave hind-cast.”
“Without wave data available to validate model results it is hard to say at this time if this is considerable. There are other models available……….that may be better suited for these types of conditions within the bay.”
The economic impact on our citizens resulting from this initiative is far to great for us to accept a cavalier scientific justification based on trumped-up data.
Appendix A
“the NOAA database seemed to be in error, based on paper gage records and plotting data. For this analysis the errors were corrected.”
“Seemed” means; to give the impression of being, i.e. may. How do we know the corrections are accurate?
“There is a break in the record from 1983 to 1993…..” again no supporting data.
……..page B 20, “it is recommended that…(LIDAR) surveys or similar data be collected and used to improve shoreline elevation assumptions”
Viewed in total, we can not help but conclude that this study was designed with a clear pre- determined outcome. The BFE in the region must be increased in order to expand FEMA's regulatory scope. The prior Anchor Bay Study did not use wave run up and overtopping in its determination. The revised study, if viewed from the perspective of the earlier study approach, would conclude that the BFE should be lowered. Addition of these two variables, which are un-verified or tested in the Lake St. Clair/Anchor Bay environment, became necessary in order to fulfill FEMA's pre-determined conclusion. It should further be noted that the restudy was ordered by a congressional hearing that determined that additional data from 19 years was relevant and that appropriate datum be reflected in the results. There was no consideration given to the addition of two new variables to the study.
At best, the FEMA study on Lake St. Clair is hypothetical. It is unfortunate that the results expressed are only reproducible on a computer and can not be tested in the real world because of a current lack of hard data and the lack of data gathering capability to support future analyses. FEMA's rush to judgment sets a dangerous precedent.
So, we go for the next best thing, which is proving theories wrong. That's easy. You just find some evidence that contradicts what the theory says. The theory is then falsified and stays that way. With that being said, we can prove that the highest recorded water level for Lake St. Clair of all time was 576.69’ falling way short of the new proposed BFE and that all the data stated in the Flood Restudy is nothing more than theory that cannot be supported by fact in any manner.
When we become comfortable with imposing mandates based on assumptions not provable by fact we have started down a road that is very hard to reverse.
Perhaps one of the most difficult issues to understand is the need to raise the flood plain for areas that cannot possibly be affected by wave run up and over topping due to the size and width of the area affected, such as the North, Middle and South Channels of the St Clair River. The report clearly indicates the data from the North Channel gage is “relatively sparse” and there is absolutely no data collection to reflect the Middle or South Channels of the St Clair River. The technical data used in the report to support the BFE increased in areas adjacent to these channels is only and specifically from Lake St Clair and Anchor Bay, therefore how can we make a scientific statement to support the increase when we do not have scientific, testable data to support such a finding.
The tools and technology certainly exist to do the type of water level study necessary to determine to a greater degree of certainty the BFE at 1% risk of flooding in Lake St. Clair and Clay Township. However, the critical tools necessary to gather the data to support a meaningful study simply do NOT EXIST along the shore of Lake St. Clair and Anchor Bay. Until these tools are in place, any study of this kind will be purely 'guesstimation'. Our citizens should not be forced to pay insurance premiums for a hypothetical flood determined by 'guesstimation'.
Attached you will find The Summary Report done by The International Joint Commission titled Impacts on Upper Great Lakes Water Levels: St Clair River. The IJC has determined that the outflow through the St Clair River was increased largely due to the ice jam of 1984 in which the outflow characteristics of the river bottom were changed dramatically. This would allow as much as an additional 6 billion gallons of water a day to flow down the St Clair River during high water periods.
“Although the river is enlarged, it would only carry an extra 6 billion gallons a day out of the river when the upper lakes are at high levels”, said Eugene Stakhiv, cochairman of the study. “If levels in those lakes get too high, the fact that more water would escape through the river should be viewed as a good thing," Stakhiv said.
With this new data coming to light it is clear that water level data used prior to 1984 would not and should not be weighted the same as data collected after 1984.
In review of the study we were unable to confirm that the most basic verification process of peer review had taken place. This process is perhaps the only way to confirm scientific credibility and accountability when dealing with such a complex issue in a scale of inches in relationship to the larger Lake St. Clair as a whole.
In light of the fact of all of the above inconsistencies as well as key factors such as regulated controls of water levels ( St Mary’s River, Niagara River, Chicago River, Wellen Canal), product exportation (packaged goods, human consumption), climate change (4-6 inches of reduction) as well as glacial isostatic adjustment (rebounding of the earth’s crust, ½-1 inch reduction), which have been agreed upon by some 100 scientist and engineers have not been considered in any manner to account for potential reduction in long term water levels for Lake St Clair it is obvious that scientifically the argument has been made that the Flood Level Restudy (March 2008) is scientifically flawed.
The Citizens of Clay Township believe that the Flood Level Restudy of Lake St Clair and Anchor Bay in Macomb and St Clair Counties, Michigan is fundamentally assumptive and can not be demonstrated by hard scientific and technical data, and that determination to increase the BFE for the addressed communities is arbitrary and capricious and officially files this protest in light of the comments presented in this letter. Enclosed please find a package of 600 letters received from our constituents relating to this issue. Among those, you will find a letter dated March 30, 2009 from the Harsen’s Island St. Clair Flats Association which addresses in some detail the technical concerns. Clay Township is in agreement with the findings and concerns and recommendations presented in that letter.
On behalf and including the some 600 protests included we do officially file our protest.
Jay M DeBoyer
Supervisor
Clay Township, St. Clair County, MI
CC: Governor Jennifer Granholm
United States Senator Carl Levin
United States Senator Debbie Stabenow
United States Congresswomen Candice Miller
State of Michigan Senator Jud Gilbert
State of Michigan Representative Phil Pavlov
St Clair County Board of Commissioners |
