uses - USGS · S. D. Warren bridge in Westbrook, Maine 20 4. Elevation and usable storage data for...
Transcript of uses - USGS · S. D. Warren bridge in Westbrook, Maine 20 4. Elevation and usable storage data for...
Flood of October 1996 in Southern Maine
by Glenn Hodgkins and Gregory J. Stewart
U.S. DEPARTMENT OF THE INTERIORU.S. GEOLOGICAL SURVEYWater-Resources Investigations Report 97-4189
Prepared in cooperation with theFEDERAL EMERGENCY MANAGEMENT AGENCY
uses Augusta, Mainescience for a changing world 1997
U.S. DEPARTMENT OF THE INTERIOR
Bruce Babbitt, Secretary
U.S. GEOLOGICAL SURVEY
Mark Schaefer, Acting Director
For additional information Copies of this report cancontact: be purchased from:
Chief, Maine District USGS Information ServicesU.S. Geological Survey Box 2528626 Ganneston Drive Denver, CO 80225Augusta, Me 04330 Telephone: (303) 202-4200 Telephone: (207) 622-8201
CONTENTS
Abstract 1 Introduction 1
Purpose and Scope 1Acknowledgments 2
Antecedent Hydrologic Conditions 2Streamflows 2Rainfall 2
Storm Characteristics 2Rainfall Amounts 4Rainfall Frequency 5
Description of Flood 6Peak Water-Surface Elevations 6Peak Flows 6Peak-Flow Frequency 20Extrapolation of Peak-Flow Frequency 21Reservoir Storage 22Flood Damages 23
Historical Perspective on Flood 23Historical Peak Flows 24Historical Peak Water-Surface Elevations 25
Summary and Conclusions 26 References Cited 27
FIGURES
1. Map showing location of study area and selected points 32. Graph showing daily rainfall totals from September 1, 1996 to October 19, 1996 at Portland
International Jetport 43. Graph showing cumulative rainfall from October 20-22, 1996 at Portland International
Jetport 54. Photograph showing aerial view of Presumpscot River at Route 302 in Westbrook, Maine on
October 22, 199675. Photograph showing Presumpscot River at Route 302 in Westbrook, Maine on October 22,
199676. Photograph showing flooding at Winona Avenue in the Ocean Park area of Old Orchard Beach,
Maine on October 22, 1996 87. Photograph showing Stroudwater River at Saco Street in Westbrook, Maine on
October 22, 199688. Photograph showing Cascade Brook (Stuart Brook) at Old Blue Point Road in Scarborough,
Maine on October 22, 1996 99. Map showing total rainfall isolines and location of sites where peak water-surface elevations were
determined for October 1996 southern Maine flood 10
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 jjj
10. Map showing total rainfall isolines and location of sites where peak flows and recurrence intervals were determined for October 1996 southern Maine flood 19
11. Graph showing relation between recurrence interval of peak flows and basin-average rainfall for October 1996 southern Maine flood 22
12. Graph showing Crippen and Bue region 1 envelope curve with data from October 1996 southern Maine flood 25
TABLES
1. Peak water-surface elevations for October 1996 southern Maine flood 112. Peak-flow data for October 1996 flood and previous floods in southern Maine 183. Recurrence intervals, exceedance probabilities, and peak flows for the Presumpscot River at the
S. D. Warren bridge in Westbrook, Maine 204. Elevation and usable storage data for Sebago Lake 235. Significant historical peak flows on the Presumpscot River in Westbrook, Maine 246. Flood flows that defined Crippen and Bue region 1 envelope curve 257. Historical peak water-surface elevations on the Presumpscot River in Maine 26
CONVERSION FACTORS
Multiply
Length
inch (in)
foot (ft)
mile (mi)
Area
square mile (mi2)
Volume
cubic foot (ft3 )
Flow
cubic foot per second (ft3/s)
By
25.40
0.3048
1.609
2.590
0.02832
0.02832
To obtain
millimeter
meter
kilometer
square kilometer
cubic meter
cubic meter per second
VERTICAL DATUM
In this report, all elevations are referenced to the National Geodetic Vertical Datum of 1929, a geodetic datum derived from a general adjustment of the first-order level nets of both the United States and Canada. It was formerly called Sea Level Datum of 1929.
jV Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Flood of October 1996 in Southern Maine
By Glenn Hodgkins and Gregory J. Stewart
ABSTRACTUp to 19.19 inches of rain were recorded in southern Maine from October 20-22, 1996. This rainfall caused severe flooding that resulted in one death, damage to more than 2,100 homes and businesses, and the destruction of bridges and dams in 8 communities. Peak flows, with estimated recurrence intervals of greater than 500 years, were recorded on 3 streams. The peak flow on the Presumpscot River in Westbrook, Maine was 68 percent larger than any other flow at that location in the last 102 years.
This report provides a detailed description of the October 1996 flood in southern Maine. It presents peak streamflows for 13 sites, peak-flow recurrence intervals for 12 of these 13 sites, and peak water-surface elevations for 74 sites; describes hydrologic conditions prior to the October 20-22 storm; provides a method for estimating the recurrence intervals of peak flows at sites where peak flows were not determined (for this flood); and reports a significant amount of historical flood data for the Presumpscot River.
INTRODUCTIONTypically, the most important factor contributing to extreme flood flows on streams in Maine is rainfall. Total rainfall amounts, rainfall intensity, and the spatial distribution of rainfall in a drainage basin are all important in determining the magnitude of flood flows. Other significant contributing factors include the antecedent hydrologic conditions (the conditions prior to a flood, such as soil moisture and streamflows) and the physical characteristics of the drainage basin in question (such as drainage area and stream slope). Factors such as snowmelt and dam breaks, in certain situations, can also be important. Extreme water- surface elevations are typically caused by extreme flood flows, but can be aggravated or caused by ice jams, debris jams, and other factors.
Documented flood data are valuable to many people. Peak flows, recurrence intervals of peak flows, and peak water-surface elevations (associated with the peak flows) are used to establish, assess, and verify the accuracy of Federal Emergency Management Agency (FEMA) flood insurance study 100-year and 500-year floodplain limits and flood profiles. These flood insurance studies are used to promote sound flood plain management by local, State, and Federal officials. Flood information is also important in the design of hydraulic structures (such as bridges, culverts, and dams), helping to ensure the safety and cost effectiveness of the structures.
Extraordinary amounts of rain fell in southern Maine from October 20-22, 1996. Record-breaking streamflows and consequent flooding caused one death, damage to more than 2,100 homes and businesses, and destruction of bridges and dams in 8 communities. The U.S. Geological Survey (USGS) in cooperation with the Federal Emergency Management Agency (FEMA), made observations and measurements to describe and document the October 1996 flood.
Purpose And ScopeThis report provides a detailed description of the October 1996 .flood in southern Maine (fig. 1). The meteorological characteristics of the October 20-22 storm are summarized and the rainfall amounts from the storm and their associated frequencies are given. Hydrologic conditions (streamflow and rainfall) in the area, prior to the storm, are described. This report gives peak water-surface elevations for 74 sites, peak streamflows for 13 sites, and peak-flow recurrence intervals for 12 of these 13 sites from the October 1996 flood. This report also describes a method for estimating the recurrence intervals of peak flows at sites where peak flows were not determined (for this flood). Reservoir storage data for Sebago Lake are given, flood damages in southern Maine are described,
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 1
and a significant amount of historical peak flow and peak water-surface elevation data, mostly for the lower Presumpscot River, are presented.
AcknowledgmentsS. D. Warren Co. provided historical streamflow and peak water-surf ace elevation data for the Presumpscot River and provided information on the Presumpscot River and Sebago Lake for the October 1996 southern Maine flood. Meteorological data were provided by the National Weather Service. Central Maine Power Co. provided streamflow data for the lower Saco River.
This report is the culmination of an intensive effort by dedicated personnel of the U.S. Geological Survey (USGS). The following USGS employees provided significant help collecting the data, analyzing the data, preparing the final report, and/or reviewing the final report: William P. Bartlett, Jr., James M. Caldwell, Andrew R. Cloutier, Gary D. Firda, Richard A. Fontaine, William H. Kirby, Robert G. Lippert, Gloria L. Morrill, Joseph P. Nielsen, Scott A. Olson, Perry N. Silverman, and Chester Zenone.
ANTECEDENT HYDROLOGIC CONDITIONSThe hydrologic conditions in a watershed prior (antecedent) to a storm can be important in determining the severity of flooding. For example, high streamflows prior to extreme rainfall would result in higher flood flows than if prior streamflows were low. Prior to the extreme rainfall on October 20-22, 1996, which caused the flooding in southern Maine, only minimal antecedent condition data are available. Information that is available on streamflows and rainfall is presented in this section.
StreamflowsThe Royal River in Yarmouth, Maine (fig. 1, site 74 in this report) was the closest active long-term streamflow gaging station to the severely flooded area of southern Maine in October 1996. The flow conditions at this site just prior to the October 1996 flood are assumed to be generally representative of flow conditions at other streams in the area shown in figure 1.
The daily mean flow on October 19, 1996 at the Royal River gaging station was 52 cubic feet per second
s)
(ft /s) and was receding from a small peak on October 10, 1996. Eighteen percent of all the daily mean flows at the Royal River gaging station from October 1,
1949, to September 30, 1996, were less than 52 ft3/so
and 50 percent were less than 120 ft /s. Thirty nine percent of all the October daily mean flows from October 1, 1949, to October 31, 1995, were less than
o o
52 ft /s and 50 percent were less than 61 ft /s. The streamflow at the Royal River gaging station prior to the October 1996 southern Maine flood was a typical October flow.
RainfallThe National Weather Service operates a recording rain gage at the Portland International Jetport (fig. 1). The daily rainfall totals at the Jetport from September 1, 1996 to October 19, 1996 are shown in figure 2. The total rainfall for September 1996 was 3.54 inches and the total rainfall from October 1 through October 19, 1996, was 1.68 inches. Using data from 1961 to 1990, daily normal rainfall amounts were computed (National Oceanic and Atmospheric Administration, 1990). The normal rainfall total for September is 3.09 inches. The normal rainfall total for October 1 to October 19 is 2.21 inches. The total rainfall in the weeks prior to the October 1996 southern Maine flood was near normal.
STORM CHARACTERISTICSVery large amounts of rain fell in southern Maine from the afternoon of October 20, 1996, to early in the morning of October 22. The maximum recorded rainfall from this storm was 19.19 inches in the Camp Ellis area of Saco, Maine (fig. 1). This record-breaking precipitation was caused by a combination of several meteorological factors. On Sunday, October 20, a strong coastal low pressure system (northeaster) was located near Washington, D. C. At the same time, Hurricane Lili was located about 900 miles southeast of Portland, Maine, and a strong high pressure system was centered over Labrador. The high pressure system acted as a block, preventing other systems from moving. The northeaster remained nearly stationary for about 36 hours while moisture from Hurricane Lili became entrained into the large circulation created by
2 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
70°45' JO'301 70° 15'
44°00'
43°45'
43 "30'
irftero Vyj v /-v rSk>A
>~lx^V / /I1-Df"'^/^I n> ( *f/ ,'V\ /I ^^'f.
STJVKLC'-.
71./ MAINE L rVr )
Base from U.S. Geological Survey Portland, and Kittery quadrangles 1:100,000,1985 0 SCALE SMILES
0 5 KILOMETERS
Figure 1 . Location of study area and selected points.
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 3
10 15
SEPTEMBER
20 25
1996
30 10
OCTOBER
15 9
Figure 2. Daily rainfall totals from September 1,1996 to October 19,1996 at Portland International Jetport (National Oceanic and Atmospheric Administration, 1996a).
the northeaster. This "conveyor belt" of moisture from Hurricane Lili, combined with the slow movement of the northeaster, resulted in the tremendous rainfall totals in Maine (Tom Hawley, National Weather Service, written commun., 1996).
Rainfall AmountsThe rainfall totals from the October 20-22, 1996 storm are shown with lines of equal rainfall (rainfall isolines) in figure 9 (page 10). These isolines were computed by the National Weather Service (NWS), based on rainfall totals from NWS observers in southern Maine. The largest recorded totals were: 19.19 inches in the Camp Ellis area of Saco, 19.00 inches in Gorham, 17.62 inches in Westbrook, 16.00 inches in South Windham, and 15.52 inches in Scarborough (Tom Hawley, written commun., 1997).
The National Weather Service rain gage at the Portland Internationa] Jetport was the only recording rain gage in southern Maine during the October 20-22, 1996 storm. Figure 3 is a graph of the cumulative rainfall amounts at the Jetport from October 20-22, 1996 and is based on hourly rainfall data from the Jetport gage (National Oceanic and Atmospheric Administration, 1996b). Maximum rainfall totals for different time periods were calculated using this same hourly rainfall data. The maximum 48-hour, 24-hour, 12-hour, 6-hour, and 2-hour rainfall totals at the Jetport were 14.65 inches, 13.32 inches, 8.86 inches, 4.81 inches, and 2.06 inches, respectively.
The data in figure 3 show that most of the rain at the Jetport from this storm fell in a 26-hour period from October 20 at 8:00 p.m. to October 21 at 10:00 p.m. The relatively uniform slope in this figure indicates
Flood of October 1996 in Southern Maine USGS WRIR 97-4189
that most of the rain at the Jetport fell at a relatively constant rate.
At the Jetport, the total rainfall for October 20-22, 1996 was 14.65 inches and the maximum 24-hour total was 13.32 inches. This 24-hour total shattered the previous 24-hour record of 7.83 inches set during Hurricane Bob in August 1991. Rainfall records for Portland have been kept since 1871 (Tom Hawley, oral commun., 1997). The total rainfall isolines in figure 9, indicate that the 13.32-inch 24-hour rainfall at the Jetport was exceeded at other locations.
Rainfall FrequencyThe Atlas of Precipitation Extremes for the Northeastern United States and Southeastern Canada (Wilks and Cember, 1993) contains maps with recurrence-interval isolines for 1 -day and 2-day rainfall
totals. The n-year (where n equals 100, 50, 25, 10, 5, and 2), 1-day and 2-day rainfall recurrence intervals for Portland were interpolated from these maps. The recurrence interval is the average period of time between rainfalls that are greater than, or equal to, a specified magnitude. As an example of recurrence interval, the 100-year rainfall is the rainfall that, on long-term average, would be equaled or exceeded once every 100 years. Conversely, this means that there is a one percent chance, every year, that a rainfall of this magnitude will be equaled or exceeded.
The 48-hour, n-year rainfall amounts for Portland were computed by applying the adjustment factor from the Atlas of Precipitation Extremes for the Northeastern United States and Southeastern Canada to the 2-day rainfall totals. The 24-hour rainfalls were computed by applying the adjustment factor to the I-day rainfalls.
16
14
UJ Xo
LL
<DC
10
O
12am 4am Sam 12pm 4pm 8pm 12am 4am Sam 12pm 4pm 8pm 12am 4am Sam 12pm 4pm 8pm 12am
October 20, 1996 October 21, 1996 October 22, 1996
Figure 3. Cumulative rainfall from October 20-22, 1996 at Portland International Jetport.
Flood of October 1996 in Southern Maine USGS WRIR 97-4189
The 48-hour and 24-hour adjustment factors in the Atlas were obtained from Hershfield (1961). The 12- hour, 6-hour, and 2-hour rainfalls were computed by applying adjustment factors to the I-day rainfall totals. These adjustment factors in the Atlas are originally from Huff and Angel (1992) and correspond closely to those in Hershfield (Wilks and Cember, 1993).
The Portland International Jetport rainfall totals from the October 20-22, 1996 storm were compared to the n- year recurrence interval rainfall totals. The 48-hour, 24- hour, and 12-hour rainfall totals were greater than 100- year recurrence interval events. The 6-hour rainfall total was a 50-year recurrence interval event and the 2- hour rainfall total was a 2- to 5-year event. The 2-hour recurrence interval shows that the October 20-22 storm did not produce the extreme rainfall intensities that can be associated with thunderstorms and hurricanes. As discussed in the Rainfall Amounts section, most of the rain from this storm fell at a relatively constant rate for about 26 hours (at the Portland International Jetport).
DESCRIPTION OF FLOODThe rain that fell in southern Maine on October 20-22, 1996, caused record-breaking streamflows and consequent flooding. On October 24, 1996, President Clinton declared an emergency in the State of Maine, and on October 28 a major disaster declaration was issued for York and Cumberland Counties (Federal Emergency Management Agency, 1996). The following sections describe the flood and its effects. The photographs included as figures 4 through 8 illustrate the severity of the flooding.
Peak Water-Surface ElevationsPeak water-surf ace elevations were determined for 74 sites in southern Maine following the October 1996 flood (table 1, page 11). The site locations and the total rainfall isolines from the October 20-22, 1996 storm are shown in figure 9 (page 10). Most of the sites are at bridges or culverts, which typically have a large influence on peak water-surface elevations. For example, bridges and culverts often cause constrictions in a stream during flood flows that result in higher peak water-surface elevations on the upstream side of the structure than on the downstream side. Because of this, both upstream and downstream peak water-surface elevations were measured and reported, when possible.
Peak water-surface elevations were based on high- water marks which were identified using the techniques of Benson and Dalrymple (1967). These marks were surveyed to the National Geodetic Vertical Datum of 1929. At each site, personnel would typically identify three or more high-water marks on the upstream side of a structure and three or more marks on the downstream side of the structure. High-water marks can include seed, conifer needle, and mud lines on trees and buildings; and debris and wash lines on the ground. The elevations listed in table 1 are the best estimates of the true peak water-surface elevations from the October 1996 flood. These elevations are based on the quality of the individual flood marks at each site. It is possible for individual flood marks to be lower or higher than the true peak flood elevation for a variety of reasons. For example, wave action can cause marks to be higher than peak water-surface elevations and settling of flood debris can cause marks to be lower than the peak water-surface elevations. Any other available information, such as observations by local residents, was also used in the process of evaluating the high water marks.
Flood peak elevation data should be used with an understanding of the methods used to obtain them and their relation to the river profile. For example, flood elevations may be affected by pileup on the upstream side of hydraulic structures (bridges, culverts, etc.) and by drawdown on the downstream side of these structures (Fontaine and Nielsen, 1994). Interpolation of peak water-surface elevations between the sites listed in table 1 should not be done without further investigation.
The time of the peak elevation or the peak flow from the October 1996 southern Maine flood is not known for most sites. The peak elevation on the Presumpscot River at S. D. Warren Co. in Westbrook was on October 22 at about 2:00 a.m. (Tom Howard, S. D. Warren Co., oral commun., 1997).
Peak FlowsThe peak streamflows determined for the October 1996 southern Maine flood and their associated recurrence intervals and exceedance probabilities are listed in table 2 (page 18). Also included in the table are the highest peak flows previously known at these sites, the periods for which peak flow data are available (periods
6 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Figure 4. Aerial view of the Presumpscot River at Route 302 in Westbrook, Maine (site 11) on October 22, 1996 (Photograph by John Patriquin / The Port land Newspapers).
Figure 5. Presumpscot River at Route 302 in Westbrook, Maine (site 11) on Octo ber 22, 1996. Note the high-water mark on Route 302 and the debris on the upstream guardrail of the bridge (Photograph by David A. Ftodgers / The Portland Newspapers).
Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Figure 6. Flooding at Winona Avenue in the Ocean Park area of Old OrchardBeach, Maine (site 54) on October 22, 1996 (Photograph by David A. Rodgers / The Portland Newspapers).
Figure 7. Stroudwater River at Saco Street in Westbrook, Maine (site 26) on Octo ber 22, 1996 (Photograph by Doug Jones / The Portland Newspapers).
8 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Figure 8. Cascade Brook (Stuart Brook) at Old Blue Point Road in Scarborough, Maine (site 34) on October 22, 1996 (Photograph by David A. Rodgers/ The Portland Newspapers).
of known peak flows), and the drainage areas of the sites. The location of the sites with peak-flow determinations and their associated recurrence intervals are shown in figure 10 (page 19). Recurrence intervals are discussed in the Peak-Flow Frequency section of this report. Total rainfall isolines for the October 20-22, 1996 storm are also included in this figure.
The flows in table 2 were determined by several different methods. The Presumpscot River peak flow in Falmouth (site 14) was computed by using a contracted opening computation technique (Matthai, 1967). The Presumpscot River in Westbrook (site 10) is at a discontinued USGS streamflow gaging station. However, an accurate peak flow determination at this site for the October 1996 flood was not possible. For this reason, the peak flow was calculated by applying a drainage area correction (Morrill, 1975) to the calculated flow at site 14. Flows released from Sebago Lake were insignificant compared to the peak flows on the lower Presumpscot River for the October 1996 flood (as shown in the Reservoir Storage section of this report). Because of this, only drainage areas for the part of the drainage basin downstream of the outlet of Sebago Lake were used for the drainage area correction
between site 14 and site 10. The peak flows at Stroudwater River in Portland (site 28) and at Capisic Brook in Portland (site 55) were computed by using a culvert flow computation technique (Bodhaine, 1967). The peak flow at Cascade Brook in Saco (site 33) was calculated by using the Water-Surface Profile Computation (WSPRO) step-backwater computer model (Shearman, 1990) using critical depth assumptions (Chow, 1959) and was verified by use of a slope-area peak-flow computation (Dalrymple and Benson, 1968). The contracted opening, culvert, critical depth, and slope-area methods all use hydraulic principles to compute peak flows from peak water- surface elevations and channel geometry data.
Crooked River in Casco/Naples (site 41), Stony Brook in Sebago (site 62), and Royal River in Yarmouth (site 74) are active (1997) USGS stream gaging stations. The peak flows for these sites were computed from the relation between water-surface elevation and flow, using standard USGS techniques (Rantz, 1982). Pease Brook in Cornish (site 42), Collyer Brook in Gray (site 49). Little Ossipee River in Limington (site 50), Mousam River in Sanford (site 59), and Branch Brook in Wells (site 68) are former USGS gaging stations.
Flood of October 1996 in Southern Maine USGS WRIR 97-4189
70 "SO1 70° 15'
u iuu r\i
EXPLANATION
Line of equal precipitation, in inches
-Base from U.S. Geological Survey Portland, and Kittery quadrangles 1:100,000,1985
Figure 9. Total rainfall isolines and location of sites where pi for October 1996 southern Maine flood. [Rainfall i
f October 1996 In Southern Maine USGS WRIR 97-4189
SCALESMILES
_______ I
0 5 KILOMETERS
I location of sites where peak water-surface elevations were determined ... _ Jtober 1996 southern Maine flood. [Rainfall isolines provided by National Weather Service.]
10 Flood of October 1996 In Southern Maine USGS WRIR 97-4189
Table 1. Peak water-surf ace elevations for October 1996 southern Maine flood [na indicates not applicable]
Site number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Location
Presumpscot River
Hurricane Rd. Bridge, Windham/Gorham
Gambo Rd. Bridge (discon tinued)0, Windham/Gorham
Route 202 Bridged,Windham/Gorham
Mallison St. Bridge6, Windham/Gorham
7 Rousseau Road,Windham
40 Lincoln Streetf,Westbrook
Bridge St. Bridge, Westbrook
20 Water St.,Westbrook
Cumberland St. Bridges8, Westbrook
S. D. Warren Bridge,Westbrook
Route 302 Bridge,Westbrook/Portland
Blackstrap Rd. Bridge, Falmouth
Route 26/100 Bridge,Falmouth
1-95 Bridges, Falmouth
Smelt Hill Dam,Falmouth
Latitude (degrees, minutes, seconds)
43 45 58
434451
43 44 02
43 43 40
43 41 55
43 40 52
43 40 39
43 40 50
43 40 59
4341 11
434211
43 43 28
43 43 42
43 43 40
43 43 06
Longitude (degrees, minutes, seconds)
70 26 55
70 26 21
70 25 35
702515
7023 11
702219
70 22 03
70 21 29
70 21 08
70 20 55
70 19 32
7018 11
70 17 45
70 17 10
701611
Upstream elevation3
(feet)
138.3 (50ft)
137.2 ' (20ft)
115.7(600 ft)
None found.
na
na
58.8 (130 ft)
na
52.5 (10ft)
47.5(5ft)
43.8(50 ft)
41.0 (50ft)
38.8(100ft)
37.2 (200 ft)h
26.5(350 ft)
Downstream elevation3
(feet)
None found.
118.0 (500 ft)
97.6(300 ft)
95.5 (75 ft) 87.5 (900 ft)
na
na
55.5 (75 ft)
na
51.6 (30 ft)
None found.
None found.
40.0 (200 ft)
None found.
36.4 (150ft)1
14.4(170ft)
Elevationb (feet)
na
na
na
na
80.9
76.8
na
54.5
na
na
na
na
na
na
na
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 11
Table 1. Peak water-surface elevations for October 1996 southern Maine flood--Continued [na indicates not applicable]
Site number
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Location
Nonesuch River
Watson Mill Rd. Culvert, Saco
Broadturn Rd. Culvert,Scarborough
Mitchell Hill Rd. Culverts,Scarborough/Gorham
Beech Ridge Rd. Culvert, Scarborough
Payne Rd. Culvert, Scarborough
Route 114 (GorhamRd.) Culvert, Scarborough
Route 1/9 Bridge, Scarborough
Route 207 (Black Point Rd.) Bridge, Scarborough
Stroudwater River
Route 114 Bridge,Gorham
Brackett St. Culverts,Gorham
Saco St. Culvert, Westbrook
Spring St. Culvert, Westbrook
Congress St. Bridge, Portland
Westbrook St. Bridgek,Portland
Latitude (degrees, minutes, seconds)
43 35 10
43 36 13
43 37 16
43 37 39
43 36 58
433651
43 36 33
43 34 47
43 38 36
43 39 23
43 40 03
43 39 38
43 39 26
43 39 22
Longitude (degrees, minutes, seconds)
70 27 59
70 26 48
70 25 41
70 24 33
702120
70 20 30
701851
70 19 36
70 25 20
70 24 04
70 22 28
702143
701851
701838
Upstream elevation3
(feet)
89.8 (50 ft)
50.0(5ft)
41.8(25 ft)
37.9 (20 ft)
25.3 (30ft)
24.6 (70ft)
17.4 (5ft)
9.2(5ft)
91.5(50ft)
None found.
74.8 (15ft)
59.3 (10ft)
33.8J (right bank) (25 ft)
34.7J (left bank) (20 ft)
26.0(100ft)
Downstream elevation3
(feet)
80.2 (100ft)
49.2(5ft)
41.5(20 ft)
37.5 (30ft)
None found.
None found.
15.0 (200 ft)
7.8 (5ft)
91.0(160ft)
80.4(15 ft)
71.2 (40 ft)
50.6 (50ft)
29.4 (60 ft)
None found.
Elevation (feet)
na
na
na
na
na
na
na
na
na
na
na
na
na
na
12 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Table 1. Peak water-surface elevations for October 1996 southern Maine Rood Continued [na indicates not applicable]
Site number
30
31
32
33
34
35
Location
Cascade Brook
Jenkins Rd. Bridge,Saco
Route 1 Culvert,Saco
Route 98 Culvert,Saco
Cascade Falls,Saco
Old Blue Point Rd. Culvert, Scarborough1
136 Pine Point Rd., Scarborough1
Latitude (degrees, minutes, seconds)
43 33 03
43 32 38
43 32 37
43 32 34
43 33 27
43 33 56
Longitude (degrees, minutes, seconds)
70 26 17
70 24 54
70 24 33
70 24 23
70 22 54
70 22 30
Upstream elevation3
(feet)
125.9(20 ft)
88.0(50ft)
85.4(60 ft)
66.0(300 ft)
20.3 (20 ft)
10.4 (10ft)m
Downstream elevation3
(feet)
121.5(80 ft)
None found.
76.0(60 ft)
None found.
15.9 (100 ft)
8.6 (25 ft)n
Elevationb (feet)
na
na
na
na
na
na
OTHER SITES BY TOWN
36
37
38
39
40
Biddeford
Moors Brook Culvert,Route 208
Thatcher Brook Culvert,South St.
West Brook Culvert,West St.
Cape Elizabeth
Spurwink River Culvert, Spurwink Ave.
Willow Brook Culverts,Scott Dyer Rd.
43 28 16
43 29 41
43 28 19
43 35 24
43 35 41
70 24 48
70 28 46
70 27 23
70 14 44
701431
16.1(25 ft)
61.9(40 ft)
124.5(130 ft)
8.8 (10ft)
19.5(5ft)
None found.
59.0(30 ft)
None found.
7.0 (30 ft)
17.0(10ft)
na
na
na
na
na
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 13
Table 1. Peak water-surface elevations for October 1996 southern Maine Rood Continued [na indicates not applicable]
Site number
41
42
43
44
45
46
47
48
49
Location
Casco/Naples
Crooked River Bridge, Route 11
Cornish
Pease Brook Culvert, Route 25/1 17
Cumberland
East Branch Piscataqua River Culvert, Route 9
Falmouth
Piscataqua River Culverts, Leighton Rd.
Gorham
Little River Bridge, Route 202
Little River Bridge, Route 237
South Branch Stroudwater River Culvert, Hodgdon Rd.
Tannery Brook Culverts, Route 114
Gray
Collyer Brook Bridges, Route 202/1 00/4
Latitude Longitude ,, ,, Upstream (degrees, (degrees, , . a
. . elevationa minutes, minutes, ._ .
A \ A \ (feCt)seconds) seconds)
43 58 45 70 33 51 None found.
434715 704557 326.4 (30 ft)
434627 701509 44.5 (55 ft)
43 44 15 70 17 40 38.2 (20 ft)
434315 702528 92.5 (60 ft)
43 42 48 70 24 52 88.6 (120ft)
43 37 42 70 26 55 None found.
434111 702651 147.1 (160ft)
43 55 05 70 19 02 None found.
Downstream elevationa
(feet)
280.8 (5ft)
323.7 (30 ft)
44.1 (105 ft)
38.1 (85 ft)
92.3 (50 ft)
87.3 (100ft)
103.9 (100 ft)
140.8 (50 ft)
182.9 (50 ft)0
Elevationb (feet)
na
na
na
na
na
na
na
na
na
14 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Table 1. Peak water-surf ace elevations for October 1996 southern Maine Hood--Continued [na indicates not applicable]
Site number
50
51
52
53
54
55
56
57
58
59
Latitude LongitudeT . (degrees, (degrees, Location . .
minutes, minutes,seconds) seconds)
Limington
Little Ossipee River Bridge, 43 41 21 70 40 15 Nason's Mills Rd.
Old Orchard Beach
42 Walnut St. 43 3 1 22 70 22 28
15 Casco Ave. 43 30 06 70 23 07
18 Roanoke Ave. 43 30 09 70 23 05
1 8 Winona Ave. 43 30 12 70 23 03
Portland
Capisic Brook Culvert, 43 39 52 70 1 8 27 Capisic St.
Fall Brook Culvert, 43 41 18 70 16 26Washington Ave.
Saco
Goosefare Brook Culvert, 43 3 1 06 70 25 1 1Ross Rd.
Sawyer Brook Culvert, 43 30 18 70 26 53 Sawyer St.
Sanford
Mousam River Bridge, 43 25 05 70 39 30Wieners Mill Rd.
Upstream elevation3
(feet)
280.9 (35 ft)
na
na
na
na
36.3 (65 ft)
50.0(120ft)
33.4(40 ft)
66.1 (50ft)
171.1(120ft)
Downstream elevation3
(feet)
None found.
na
na
na
na
34.0 (5ft)
47.6(10ft)
30.0(20 ft)
66. 1? (30ft)
None found.
Elevation13
(feet)
na
10.3
9.6
9.5
9.6
na
na
na
na
na
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 15
Table 1. Peak water-surface elevations for October 1996 southern Maine flood Continued [na indicates not applicable]
Site number
60
61
62
63
64
65
66
67
68
Location
Scarborough
521 Route 1
Stuart Brook Culvert,Route 1
Sebago
Stony Brook Culvert,Route 114
South Portland
Kimball Brook Culverts,Highland Ave.
Long Creek Culverts,Maine Mall Rd.
Long Creek Culverts,Foden Rd.
Long Creek Culvertq,Route 9
Wells
Blacksmith Brook Culvert,Route 1
Branch Brook Culvert,Route 9A
Latitude Longitude ., ,-. Upstream (degrees, (degrees, , . fl
. 5 . 5 elevation3 minutes, minutes, .,.
A \ A\ (feCt)seconds) seconds)
43 34 37 70 22 40 na
43 33 40 70 23 58 77.4(10ft)
435120 703825 282.7(5ft)
433801 701501 12.4(40 ft)
43 38 30 70 20 02 52.5(20 ft)
433810 701946 50.9(5ft)
433758 701850 26.9(20 ft)
431959 703417 47.4(75 ft)
43 22 45 70 35 02 50.6(50ft)
Downstream elevation3
(feet)
na
69.6(25 ft)
None found.
12.2(20 ft)
52.5(20 ft)
43.9(70ft)
8.4(400 ft)
42.1(20 ft)
50.0(30 ft)
Elevationb (feet)
9.7
na
na
na
na
na
na
na
na
Westbrook
69 39 Eisenhower Drive 43 39 35 70 22 03 na na 59.5
70 29 Eisenhower Drive 43 39 34 70 21 59 na na 59.6
16 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Table 1. Peak water-surface elevations for October 1996 southern Maine Rood Continued [na indicates not applicable]
Site number
71
72
73
Location
Windham
Colley Wright Brook Bridge, Montgomery Rd.
Colley Wright Brook Bridge, River Rd.
Pleasant River Bridge, Pope Rd.
Latitude (degrees, minutes, seconds)
43 44 38
43 43 35
4347 11
Longitude (degrees, minutes, seconds)
70 23 55
70 24 24
7025 18
Upstream elevation3
(feet)
None found.
96.3(70ft)
167.0 (70ft)
Downstream elevation3
(feet)
119.6 (50ft)
96.2 (100ft)
161.9 (60 ft)
Elevationb (feet)
na
na
na
Yarmouth
74 Royal River Bridge, East Main St.
43 47 56 70 10 45 15.1 (150ft)
None found. na
a The location of the upstream and downstream flood elevations in relation to the hydraulic structure are referenced from the upstream or downstream side of the structure, respectively. All distances are approximate and are shown in parenthesis under the flood mark elevations.
b This column is used for flood elevations that are not associated with hydraulic structures.c Newhall Dam is located about 150 feet downstream of discontinued bridge.
d Little Falls Dam is located about 200 feet downstream of bridge.e South Windham Dam is located about 150 feet downstream of bridge.
f Saccarappa Dam is located about 1000 feet downstream of this site.
g Cumberland Mills Dam is located about 200 feet downstream of bridges.
h Upstream of the upstream bridge.
1 Downstream of the downstream bridge.
J The difference in elevations was probably caused by superelevation of the water around a bend.
k Dam is located about 50 feet upstream of bridge.
1 Cascade Brook is also known as Stuart Brook at this location.m Upstream of the road at 136 Pine Point Rd.
n Downstream of the road at 136 Pine Point Rd.
0 Downstream of discontinued Route 202/100/4 Bridge.
p At normal flows, Sawyer Brook flows underground, starting at Sawyer Street.
q Clark Pond Dam is located about 40 feet upstream of culvert.
Peak flows were determined at these sites by using the most recent elevation-flow relation at each site with the peak water-surface elevation from the October 1996 flood.
The highest peak flows previously known for the sites in table 2 were taken from USGS records, which are published by the USGS in the annual report series titled
"Water Resources Data - Maine". A significant amount of historical information for the Presumpscot River in Westbrook near site 10 (explained further in the Historical Perspective on Flood section) was obtained from S. D. Warren Co.
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 17
00
Tab
le 2
. Pe
ak-f
low
dat
a fo
r O
ctob
er 1
996
flood
and
pre
viou
s fl
oods
in
sout
hern
Mai
neo
[das
h in
dica
tes
not a
vaila
ble,
non
e in
dica
tes
not a
cur
rent
or
form
er U
.S.
Geo
logi
cal
Surv
ey s
trea
m g
agin
g st
atio
n, c
ubic
fee
t per
sec
ond
abbr
evia
ted
as f
t /s
, sq
uare
mile
s ab
brev
iate
d as
mi
, > i
ndic
ates
gre
ater
than
, <
indi
cate
s le
ss t
han]
Flood
of
October
1996
in 0 <D 5 Q 5' (D C 0 i WRIR
97 00
Peak
flo
w d
urin
g O
ctob
er
Site
nu
mbe
r3
10 14 28 33 41 42 49 50 55 59 62 68 74
USG
Sga
ging
st
atio
n nu
mbe
r
0106
4118
Non
e
0106
4158
Non
e
0106
3100
0106
6100
0105
9800
0106
6500
Non
e
0106
9500
0106
3310
0106
9700
0106
0000
Stre
am a
nd lo
catio
n
Pres
umps
cot R
iver
at
S. D
. War
ren
Bri
dge,
Wes
tbro
ok
Pres
umps
cot R
iver
at 1
-95,
Fal
mou
th
Stro
udw
ater
Riv
er a
t Con
gres
s St
.,Po
rtla
nd
Cas
cade
Bro
ok a
t Cas
cade
Fal
ls,
Saco
Cro
oked
Riv
er a
t Rou
te 1
1 , C
asco
/ N
aple
s
Peas
e B
rook
at R
oute
25/
117,
Cor
nish
Col
lyer
Bro
ok a
t Rou
te 2
02/1
00/4
,G
ray
Litt
le O
ssip
ee R
iver
, at
Nas
on's
Mill
sR
d., L
imin
gton
Cap
isic
Bro
ok a
t C
apis
ic S
t., P
ortla
nd
Mou
sam
Riv
er a
t Wic
hers
Mill
Rd.
,Sa
nfor
d
Ston
y B
rook
at R
oute
114
, Seb
ago
Bra
nch
Bro
ok a
t Rou
te 9
A, W
ells
Roy
al R
iver
at E
ast M
ain
St.,
Yar
mou
th
Flow
(ft3
/s)
23,3
00b
31,5
00
13,4
00
1,65
0
2,81
0
352
1,15
0
5,80
0
865
3,60
0
106
1,02
0
4,41
0
1996
flo
od
Rec
urre
nce
Exc
eeda
nce
inte
rval
pr
obab
ility
250
year
s0
-
>500
yea
rs0
>500
yea
rs0
4 ye
ars
1 0 y
ears
15 y
ears
30 y
ears
>500
yea
rs0
50 y
ears
7 ye
ars
40 y
ears
3 ye
ars
0.00
4
-
<0.0
02
<0.0
02
0.25
0
0.10
0
0.06
7
0.03
3
<0.0
02
0.02
0
0.14
3
0.02
5
0.33
0
Hig
hest
pea
k fl
ow p
revi
ousl
y kn
own
Flow
Pe
riod
of k
now
n (f
t3/s
) pe
ak f
low
s
1991
13
,900
18
95-p
rese
nt
-
1977
1,
160
1975
- 197
7 (s
easo
nal)
-
1996
2,
800
1975
- 197
7 (s
easo
nal)
, 19
95 -p
rese
nt
1969
48
6 19
64-1
974
1969
1,
220
1964
-198
2
1936
8,
530
1936
, 19
40-1
982
-
1983
4,
020
1939
-198
4
1996
63
19
95-p
rese
nt
1972
72
3 19
64-1
974
1977
11
,500
19
49-p
rese
nt
Dra
inag
e ar
ea
(mi2
)
577d
640e 27
.5 3.13
150 4.
80
13.8
168 2.
63
99.0 1.50
10.3
141
aThe
site
num
ber
refe
rs to
the
site
num
bers
fro
m T
able
1.
b B
ased
on
a dr
aina
ge a
rea
corr
ectio
n fr
om s
ite 1
4.c
Rec
urre
nce
inte
rval
s gr
eate
r th
an 1
00 y
ears
hav
e a
larg
e un
cert
aint
y as
soci
ated
with
them
.T
he d
rain
age
area
is 1
36 m
i2 b
elow
the
out
let o
f Seb
ago
Lake
. e T
he d
rain
age
area
is 1
99 m
i2 b
elow
the
outle
t of
Seba
go L
ake.
70 45' 70° 15'
44'00'
43° 45'
43 "30'
Site where peak flow was determined. Top number is site number, bottom number is recurrence interval of peak flow in years (asterisk indicates not determined).
Base from U.S. Geological Survey Portland, and Kittery quadrangles 1:100,000,1985
SCALE 0 SMILES
~
5 KILOMETERS
Figure 10. Total rainfall isolines and location of sites where peak flows and recurrence intervals were determined for October 1996 southern Maine flood. [Rainfall isolines provided by National Weather Service.]
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 19
Peak-Flow FrequencyThe recurrence interval is the average period of time between peak flows that are greater than, or equal to, a specified magnitude. For example, the 50-year peak flow is the flow that would be exceeded or equaled, on long-term average, once in 50 years. This does not imply that flooding will happen at regular intervals. Two 50-year peak flows could be experienced in 2 consecutive years. Conversely, a 50-year peak flow might not be experienced for 100 years or longer. The reciprocal of the recurrence interval is called the annual exceedance probability; that is, the probability that a given peak flow will be exceeded or equaled in any given year. For example, the annual exceedance probability of the 50-year peak flow would be 0.02. In other words, there is a 2 percent chance that the 50- year peak flow will be exceeded or equaled in any given year.
The 500-, 100-, 50-, 25-, 10-, 5-, and 2-year recurrence-interval peak flows for sites in table 2 were determined as explained below. The peak flows from the October 1996 flood were then compared to these recurrence-interval peak flows to determine the recurrence intervals of the October 1996 peak flows.
All recurrence interval estimates have an uncertainty associated with them. The uncertainty generally increases as the recurrence interval increases. Recurrence intervals that are greater than 100 years have a large amount of uncertainty associated with them. For example, the calculated 500-year peak flow
for Cascade Brook in Saco (site 33) is 650 ft3/s, but the one-standard-error-of-estimate range of the 500-year
peak flow extends from about 200 ft3/s to about 1200
ft3/s. This means that the actual 500-year peak flow
will be between 200 ft3/s and 1200 ft3/s for about two- thirds of sites with the same drainage basin characteristics (drainage area, stream slope, etc.) as site 33. This corresponds to a flow uncertainty factor of about 2 to 3. The recurrence interval uncertainty is greater than the flow uncertainty. The recurrence
interval for a given flow of 650 ft3/s at site 33 is estimated to be 500 years with a recurrence interval uncertainty factor of about 9 (William H. Kirby, U.S. Geological Survey, written commun., 1997). That is, at two-thirds of sites similar to site 33, the actual recurrence interval is estimated to be in the range of 60 to 4500 years. Given these uncertainties, a reasonable
estimate of the recurrence interval of the 1650 ft3/s October 1996 peak flow at site 33 is not possible. In general, because of the extreme uncertainties in peak- flow recurrence intervals greater than 500 years, applicable peak-flow recurrence intervals from the October 1996 flood are reported as >500 years (in table 2), instead of their calculated recurrence intervals.
The 500-, 100-, 50-, 25-, 10-, 5-, and 2-year peak flows for the Presumpscot River in Westbrook (site 10) were calculated using the guidelines (Bulletin 17B) of the Interagency Advisory Committee on Water Data (1982). The calculations involve fitting the log-Pearson Type III probability distribution to the observed annual peaks at a site. Bulletin 17B provides a procedure for incorporating historical peak flow information into the recurrence interval calculations. The historical peak flow information provided by S. D. Warren Co. (explained further in the Historical Perspective on Flood section) was used in the Bulletin 17B analysis.
Table 3. Recurrence intervals, exceedance probabili ties, and peak flows for the Presumpscot River at the S.D. Warren Bridge in Westbrook, Maine (site 10) [cubic feet per second abbreviated as ft /s]
Recurrence interval
2 years
5 years
10 years
25 years
50 years
100 years
200 years3
500 years3
Exceedance probability
0.500
0.200
0.100
0.040
0.020
0.010
0.005
0.002
Peak flow (ft3/s)
5,310
7,830
9,890
13,000
15,700
18,900
22,400
28,000
a Recurrence intervals greater than 100 years have a large uncertainty associated with them.
20 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
The calculated peak flow from the October 1996 flood was also used in the calculations. The recurrence intervals, exceedance probabilities, and peak flows for site 10 are shown in table 3.
The peak flows for the 500-, 100-, 50-, 25-, 10-, 5-, and 2-year recurrence intervals for Pease Brook in Cornish (site 42), Collyer Brook in Gray (site 49), Little Ossipee River in Limington (site 50), Mousam River in Sanford (site 59), Branch Brook in Wells (site 68), and Royal River in Yarmouth (site 74) were all calculated using the guidelines in Bulletin 17B.
The peak flows for the 100-, 50-, 25-, 10-, 5-, and 2- year recurrence intervals for Stroudwater River in Portland (site 28), Cascade Brook in Saco (site 33), Crooked River in Casco/Naples (site 41), Capisic Brook in Portland (site 55), and Stony Brook in Sebago (site 62) were calculated using the procedures of Morrill (1975). These procedures involve computing drainage basin characteristics (such as drainage area and stream slope) and inputting these values into regression equations to compute the peak flows for specified recurrence intervals. The 500-year flood flows for these 5 sites were computed by the extrapolation method of Thomas and Kirby (1994). For Capisic Brook, an urban adjustment to the peak flows was applied using the nationwide 7-parameter urban regression equations (Sauer and others, 1983).
The extreme rains of October 20-22, 1996, in southern Maine did not cause extreme flows on the lower Saco River. The Saco River at Skelton Dam in Dayton
}
peaked at 17,200 cubic feet per second (ft /s) on October 22, 1996 (Francis R. (Mike) Towns, Jr., Central Maine Power Co., written commun., 1997).The 10-year peak flow at this location is 25,800 ft3/s (Federal Emergency Management Agency, 1980).
Extrapolation of Peak-Flow FrequencyIt is not feasible to determine at-site flows and their associated recurrence intervals for all streams in a flood-affected area. It is sometimes possible, however, to draw general conclusions based on the streams where peak-flow recurrence intervals have been determined.
A strong positive relation was found between recurrence interval of peak flows and basin-average rainfall for the October 1996 southern Maine flood (fig. 11). The 12 data points in figure 11 represent the peak- flow recurrence interval and the basin-average rainfall of the sites in table 2. The greater-than-500-year recurrence intervals for sites 28, 33, and 55 were plotted at the 500-year level. The basin-average rainfall is an estimate of the total rainfall that fell in a given basin during the October 20-22, 1996 storm. To compute basin-average rainfall, approximate drainage basin boundaries were drawn on an enlarged version of figure 10. The drainage basin for each site was divided into subareas based on the rainfall isolines. The areal average of the rainfall for all of the subareas was then computed to give the basin-average rainfall.
It is probable that unmeasured streamflows from the October 1996 flood fall in the gray range shown in figure 11. Given this conclusion, additional conclusions can be made, all of which apply only to the October 1996 flood in the area covered in this study. Also, the following conclusions may not apply to basins with drainage areas smaller than 1.5 square miles because no peak flow data was collected on basins of this size. Any drainage basin with an basin- average rainfall of more than 15 inches probably experienced a peak flow (at the drainage basin outlet) with a recurrence interval of greater than 100 years. Drainage basins with average rainfalls of less than 11 inches probably had peak flows with recurrence intervals of less than 100 years. Those with less than 10 inches of rain likely had peak flows with recurrence intervals of less than 50 years.
A range of likely recurrence intervals for peak streamflows at any location (for basins in the area covered by this study, with drainage areas greater than 1.5 square miles) could be calculated by computing a basin-average rainfall for the stream's drainage basin. This basin-average rainfall would then be entered into figure 11 to estimate a range of recurrence intervals for the October 1996 flood. For example, for the October 1996 flood, a drainage basin with an average total rainfall of 10 inches probably experienced a peak flow with a recurrence interval of 12 to 50 years. A basin with an average total rainfall of 13 inches likely had a peak flow with a recurrence interval of 40 to 250 years, while one with an average total rainfall of 16 inches
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 21
probably had a peak flow with a recurrence interval of 160 to greater than 500 years. It must be remembered that peak-flow recurrence intervals greater than 100 years have a large uncertainty associated with them (as discussed in the Peak-Flow Frequency section). Estimates of peak-flow recurrence intervals greater than 500 years have extreme uncertainty associated with them. This is why recurrence intervals greater than 500 years are not reported exactly.
Reservoir StorageFlow records for the Presumpscot River at the outlet of Sebago Lake and storage data for Sebago Lake (Tom Howard, written commun., 1996) indicate that most precipitation that fell on October 20-22, 1996, on the Sebago Lake drainage area, was stored in Sebago Lake. The daily mean flow at the outlet of the lake was
o
257 cubic feet per second (ft /s) from October 16 to 20,
1996, 165 ft3/s on October 21, 1996, and 75 ft3/s from October 22 to 31, 1996.
Lake elevation and usable storage data for Sebago Lake are shown in table 4. Usable storage is the volume of water normally available for release from a reservoir, between the minimum and maximum controllable elevations. The usable storage statistics were obtained from the monthly report, Current Water Resources Conditions in Maine (U.S. Geological Survey, 1996). From October 16 to October 23, 1996; 2,238 million cubic feet of water were stored in Sebago Lake.
The peak flow for the October 1996 flood on the Presumpscot River at 1-95 in Falmouth (site 14) was
o
31,500 ft /s. The outflow from Sebago Lake during the October 1996 flood was less than one percent of this peak flow. Therefore, the drainage area above the outlet
C/5
LU
DC
LLJ OzLLJ DC DC12 OLU DC
500400
300
200
100
70
5040
30
20
10
7
5
4
3
EXPLANATION Site number
41
"62
33
8 10 12 14 16 18
BASIN-AVERAGE RAINFALL, OCTOBER 1996 STORM (INCHES)
2C
Figure 11. Relation between recurrence interval of peak flows and basin-average rainfall for October 1996 southern Maine flood.
22 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Table 4. Elevation and usable storagea data for Sebago Lake. [dash indicates not available]_______________________
, . Volume _ Long-term Elevation . .,,. PercentDate ._ . (million _ average
(feet) , . _ . fullcubic feet) percent full
Sept. 30, 1996
Oct. 16, 1996
Oct. 23, 1996
Oct. 30, 1996
263.19
262.76
264.52
265.72
5,301
4,754
6,992
8,517
55
49
72
88
46
-
-
57
a Usable storage for Sebago Lake is the 9700 million cubic feet of water between the elevations of 259.0 feet and 266.6 feet.
of Sebago Lake did not contribute a significant amount of water to the flooding, at site 14, in October 1996. The drainage area of the Presumpscot River at the
0outlet of Sebago Lake is 441 square miles (mi ). The drainage area of the Presumpscot River at site 14 is
^
640 mi . Therefore, the October 1996 flooding at site/-\
14 was caused by rain falling on the 199 mi of drainage area below the outlet of Sebago Lake. The contribution of Sebago Lake outflows to October 1996 peak flows and elevations in the entire lower Presumpscot River (defined for this report as downstream from the confluence with the Little River in Gorham) was very likely insignificant. If the large amount of rain that fell in the Sebago Lake drainage basin in October 1996 had not been stored in Sebago Lake, peak flows and peak water-surface elevations on the lower Presumpscot River would have been even higher than they were.
The Small Business Administration estimated that 45 homes and 27 businesses had major damage from the flood and 2,002 homes and 110 businesses had minor damage.
Damage to large water mains left the City of Portland without water for several hours. A boil order was issued because of contaminated water, forcing most businesses in Portland to close from October 21, 1996 to October 25, 1996. Sewer treatment plants in Berwick, Old Orchard Beach, and Saco suffered significant damage. Clam flats from Kittery to Rockland were closed due to pollution.
Many state and local roads were washed away when their drainage structures failed. Bridges were destroyed in Cape Elizabeth, Falmouth, Old Orchard Beach, Scarborough, Wells, Westbrook, and Windham. The Maine Turnpike was closed for many hours at several locations.
Highland Lake Dam in Westbrook, Milliken Mills Dam in Old Orchard Beach, Rocky Gorge Lower Dam in South Berwick, and 2 dams owned by the State Inland Fisheries and Wildlife in Scarborough were destroyed in the flood. The Balch Pond Outlet in Shapleigh was damaged and the Rod and Gun Club Dam in the Lower Falls area of Gorham failed. Concern over possible loss of structural integrity has provoked an inspection and assessment of major dams in southern Maine.
HISTORICAL PERSPECTIVE ON FLOOD
Flood DamagesThe October 1996 flood in southern Maine affected the lives of many area residents by destroying homes and businesses, contaminating water, and washing out roads, bridges, and dams. The following damage information was taken from the Federal Emergency Management Agency's Interagency Hazard Mitigation Team Report (1996).
A motorist drowned after driving into floodwaters in Scarborough. There were fifteen reported injuries due to the flood.
Peak flows and peak water-surface elevations from the southern Maine flood of 1996 are compared to historical data in this section. Based on the recurrence intervals listed in table 2 (page 18), the 4 measured streams with the most extreme flooding in the October 1996 flood were the Presumpscot River (site 10, 250- year recurrence interval), the Stroudwater River (site 28, >500-year recurrence interval), Cascade Brook (site 33, >500-year recurrence interval), and Capisic Brook (site 55, >500-year recurrence interval).
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 23
Historical Peak Flows
Measured peak flows from the October 1996 flood are compared to the highest peak flows previously known, in table 2 (page 18). All previously known flows are from the USGS annual report series "Water Resources Data - Maine" with the exception of many flows on the Presumpscot River in Westbrook. The flows at the USGS gaging stations were calculated by standard USGS methods (Rantz and others, 1982).
Historical peak flows for the Presumpscot River at the Saccarappa Dam in Westbrook (which is about 1000 feet downstream from Site 6, see figure 9 for site locations) were supplied by S.D. Warren Co. (Tom Howard, written commun., 1996). The flows were calculated using weir computations. Some historical flows on the Presumpscot River are also available from the USGS at stream gaging stations in Westbrook (site 10, USGS station 01064118) and Falmouth (site 12, USGS station 01064140). The October 1996 peak flow for the Presumpscot River at 1-95 in Falmouth (site 14) was determined by using a contracted opening indirect measurement (Matthai, 1967). The difference in drainage area between Saccarappa Dam and the USGS gaging station at Westbrook (site 10) is less than 2%, making the flows at the two sites comparable. Flows from the USGS Falmouth gaging station and from the 1-95 bridge were adjusted by drainage area corrections (Morrill, 1975) to make them comparable to the flows at the USGS gaging station in Westbrook. All known significant historical peak flows on the Presumpscot River are listed in table 5. The October 1996 peak flow of 23,300 cubic feet per second was 68 percent larger than any other flow in the last 102 years.
Because no significant historical flows have been recorded for the Stroudwater River (site 28), Cascade Brook (site 33), or Capisic Brook (site 55), a site- specific historical perspective of the October 1996 flood at these sites is not possible. A general comparison, however, is possible. Crippen and Bue (1977) determined envelope curves for peak flow versus drainage area. These envelope curves were developed to provide a guide for estimating potential maximum flood flows. The curves were developed by analyzing thousands of sites with recorded flood flows, and using the sites with the most extreme flows to draw the envelope curves. The United States was divided
Table 5. Significant historical peak flows on the Presumpscot River in Westbrook, Maine
Date
April 15, 1895
March 1, 1896
February 13, 1900
March 2, 1900
May 17, 1916
June 17, 1917
March 12, 1936
September 11, 1954
March 14, 1977
May 12, 1989
August 20, 1991
October 22, 1996
Flow (cubic feet per second)
13,000a
13,800a
ll,300a
9,720a
12,400a
9,7 10a
ll,200a
12,400a
ll,200b
9,200C
13,900C
23,300d
a Flow computed by S.D. Warren Co. using weir equations at Saccarappa Dam in Westbrook. Flow computed by U.S. Geological Survey at streamflow gaging station in Falmouth and adjusted to Westbrook using a drainage area correction.
c Flow computed by U.S. Geological Survey at streamflow gaging station in Westbrook. Flow computed by U.S. Geological Survey by con tracted opening measurement at 1-95 bridge in Fal mouth and adjusted to Westbrook using a drainage area correction.
into 17 different regions in the Crippen and Bue report. Maine is in region 1 (region 1 covers Maine, Vermont, most of New Hampshire, most of Connecticut, western Massachusetts, and a small part of Rhode Island) and this envelope curve is shown in figure 12. The points that were used to define the Crippen and Bue envelope curve for region 1 are listed in table 6. A general historical flood comparison is made by plotting the Presumpscot River (site 14), Stroudwater River (site
24 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
QZo oLJJ C/)DC LJJ Q_
LJJ LJJ LJ_
O CD13 O
Q
1,000,000
500,000
100,000
50,000
10,000
5,000
1,000
500
100
EXPLANATIONMA17 Historical Flows
Site 55 1996 Flows »NH7
CT7 CT6
VT12 ASite14
ASite 28
MA17
^ Site 33
Site 55
0.1 1 10 100 1,000 10,000
DRAINAGE AREA (SQUARE MILES)
Figure 12. Crippen and Bue (1977) region 1 envelope curve with data from October 1996 southern Maine Flood.
28), Cascade Brook (site 33), and Capisic Brook (site 55) peak flows from October 1996 on the Crippen and Bue envelope curve for region 1. Site 14 is plotted using the drainage area below the Sebago Lake outlet (for reasons discussed in the Reservoir Storage section). Flows in figure 12 that plot closest to the envelope curve are more extreme flows than those that plot further away. On this basis, the most extreme measured flow from the October 1996 southern Maine flood was the Stroudwater River in Portland (site 28). The measured flows from this flood that are not plotted in Figure 12 are less extreme than the 4 sites that are plotted.
Historical Peak Water-Surface ElevationsComparisons of historical flood elevations can be made, but changes that occur over time, such as physical changes in structures (for example, dams and bridges) and changes in land uses can make such comparisons difficult and unreliable. The only known significant historical peak water-surface elevations (in
Table 6. Flood flows that defined Crippen and Bue (1977) region 1 envelope curve
Point
MA17
VT12
CT6
CT7
Site name
PowdermilkBrook nearWestfield, MA
Rock River atWilliamsville,VT
Salmon Brooknear Granby, CT
Naugatuck River at Tho-maston, CT
Peak flow (cubic feet
per second)
5,740
27,200
40,000
53,400
Drainage area
(square miles)
2.50
42.6
66.8
101
Date
8-19-1955
9-21-1938
8-19-1955
8-19-1955
CT4 Farmington 140,000 360 8-19-1955 River at Collins- ville, CT
NH7 Merrimack River 150,000 3,092 3-20-1936 below Manches ter, NH
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 25
the area severely affected by the October 1996 flood) are located on the Presumpscot River and are listed in table 7. The peak elevations for 1896, 1916, and 1954 were provided by S. D. Warren Co. (Tom Howard, written commun., 1996). The elevations for 1977, 1991, and 1996 are from USGS records. The peak elevations in table 7 were associated with the peak flows that are listed in table 5. The peak elevations from the October 1996 flood exceeded all previously recorded peak elevations on the lower Presumpscot River.
SUMMARY AND CONCLUSIONS
Flood flows are typically caused by a complexinterrelationship of rainfall, rainfall intensity,antecedent drainage basin conditions, physical
drainage basin characteristics and other factors. The cause of the southern Maine flood of October 1996, however, was not complex. Up to 19.19 inches of rain were recorded in southern Maine from October 20-22, 1996. The maximum 24-hour rainfall total of 13.32 inches at the Portland International Jetport exceeded the previous record (records have been kept since 1871) by 5.49 inches. Most of the rain from this storm fell in about 26 hours at a relatively constant rate (at the Jetport). None of the available data regarding hydrologic conditions prior to the October 20-22 storm were abnormally high or low. These facts and the strong relation (for this flood) between basin-average rainfall and the recurrence interval of peak flows indicate that the extreme rainfall totals from the October 1996 storm were the cause of the extreme peak streamflows experienced in southern Maine.
Table 7. Historical peak water-surface elevations on the Presumpscot River in Maine[feet abbreviated as ft, na indicates not applicable, dash indicates not available]
SiteYear
Location1896 1916 1954 1977 1991 1996
73.8 ft 73.7 ft
46.0 ft 45.8 ft
na Upstream of Saccarappa Dam, Westbrook
9 Upstream of Cumberland Mills Dam, Westbrook
10 Upstream of S. D. Warren Bridge, Westbrook
11 Upstream of Route 302 Bridge, 36.6ft 34.1ft 36.5ft Westbrook
11 Downstream of Route 302 Bridge, Westbrook
12 Upstream of Blackstrap Rd. Bridge, Falmouth
13 Upstream of Route 267100 Bridge, Falmouth
76.8a ft
51.6b ft
38.9 ft 47.5 ft
43.8 ft
33.9ft 35.5ft
35.8ft 32.0ft 33.5ft 32.3ft
35.1ft 31.7ft 31.8ft
15 Upstream of Smelt Hill Dam, 22.5ft 25.2ft 23.4ft Falmouth
41.0ft
38.8 ft
26.5 ft
a About 1000 feet upstream of Saccarappa Dam at Site 6.
This peak water-surface elevation on the upstream side of Cumberland Mills Dam is the downstream elevation of the Cumberland Street bridge (see table 1).
26 Flood of October 1996 in Southern Maine USGS WRIR 97-4189
Peak water-surface elevations from the October 1996 flood were determined at 74 locations. Peak flows were calculated at 13 sites and recurrence intervals were calculated at 12 of these 13 sites. Three of these flows had recurrence intervals of greater than 500 years. The peak flow on the Presumpscot River in Westbrook, Maine had a recurrence interval of 250 years. Peak- flow recurrence intervals greater than 100 years have a large amount of uncertainty associated with them. Based on historical data gathered for this report, the October 1996 peak flow on the Presumpscot River in Westbrook was 68 percent larger than any other flow at this location in the last 102 years. All known significant historical peak flows and peak elevations (in the area covered by this report) have been presented.
A strong positive relation was found between basin- average rainfalls and the recurrence intervals of peak flows from the October 1996 flood (in the area covered by this study). Based on this relation, it is probable that any drainage basin with an basin-average rainfall of more than 15 inches experienced a peak flow (at the drainage basin outlet) with a recurrence interval of greater than 100 years. Drainage basins with average rainfalls of less than 11 inches probably had peak flows with less than 100-year recurrence intervals. Those with less than 10 inches of rain likely had peak flows with less than 50-year recurrence intervals. Because of the strong relation between basin-average rainfall and peak-flow recurrence interval, peak-flow recurrence intervals for the October 1996 flood can be estimated for sites where peak flows were not determined (for basins in the area covered by this study with drainage areas greater than 1.5 square miles).
The October 1996 flood devastated parts of southern Maine. One person drowned after driving into floodwaters. More than 2,100 homes and businesses were damaged in the flood. Bridges, dams, or both were destroyed in Cape Elizabeth, Falmouth, Old Orchard Beach, Scarborough, Wells, South Berwick, Westbrook, and Windham. Clam flats from Kittery to Rockland were closed due to pollution.
Sebago Lake stored most of the rain that fell on the 441 square miles of the Presumpscot River drainage basin that is above the outlet of the lake. If this water had not been stored, peak flows and peak water-surface elevations on the lower Presumpscot River would have been even higher than they were in October 1996.
REFERENCES CITED
Benson, M. A., and Dalrymple, Tate, 1967, General field and office procedures for indirect discharge measurements: U.S. Geological Survey Techniques of Water-Resources Investigations, Book3, Chapter A 1,30 p.
Bodhaine, G. L., 1967, Measurement of peak discharge at culverts by indirect methods: U.S. Geological Survey Techniques of Water-Resources Investigations, Book 3, Chapter A3, 60 p.
Chow, V. T., 1959, Open-channel Hydraulics: New York, McGraw-Hill, 680 p.
Crippen, J. R., and Bue, C. D., 1977, Maximum floodflows in the conterminous United States: U.S. Geological Survey Water-Supply Paper1887, 52 p.
Dalrymple, Tate, and Benson, M. A., 1968,Measurement of peak discharge by the slope area method: U.S. Geological Survey Techniques of Water-Resources Investigations, Book 3, ChapterA2, 12 p.
Federal Emergency Management Agency, 1980, Flood Insurance Study, Town of Dayton, Maine, York County: December 1, 1980, 19 p.
Federal Emergency Management Agency, 1996, Interagency Hazard Mitigation Team Report: FEMA-DR-1143-ME, 37 p.
Fontaine, R. A., and Nielsen, J. P., 1994, Flood of April 1987 in Maine: U.S. Geological Survey Water-Supply Paper 2424, 50 p.
Hershfield, D. M., 1961, Rainfall Frequency Atlas of the United States: U.S. Department of Commerce, Weather Bureau Technical Paper 40, 115 p.
Huff, F. A., and Angel, J. R., 1992, Rainfall Frequency Atlas of the Midwest: Illinois State Water Survey, Champaign IL, Bulletin 71, 141 p.
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 27
Interagency Advisory Committee on Water Data, 1982, Guidelines for determining flood flow frequency: Water Resources Council Bulletin 17B, 28 p.
Matthai, H. F., 1967, Measurement of peak discharge at width contractions by indirect methods: U.S. Geological Survey Techniques of Water- Resources Investigations, Book 3, Chapter A4,44 p.
Morrill, R. A., 1975, A technique for estimating the magnitude and frequency of floods in Maine: U. S. Geological Survey Open File Report 75-292,44 P-
National Oceanic and Atmospheric Administration, 1990, Daily Normals of Temperature, Precipitation and Heating and Cooling Degree Days, 1961-1990, Maine, Portland WSMO AP: National Climatic Data Center, Climatography of the United States No. 84.
Thomas W. O., Jr., and Kirby, W. H., 1994, Estimation of extreme floods, in Nationwide summary of U.S. Geological Survey regional regression equations for estimating magnitude and frequency of floods for ungaged sites, 1993: U. S. Geological Survey Water-Resources Investigations Report 94-4002, p. 13-16.
U.S. Geological Survey, 1996, Current water resources conditions in Maine: Augusta, Maine (published monthly).
Wilks, D. S., and Cember, R. P., 1993, Atlas of Precipitation Extremes for the Northeastern United States and Southeastern Canada: Cornell University, Northeast Regional Climate Center, 40 p.
National Oceanic and Atmospheric Administration, 1996a, Climatological Data, New England, September 1996-October 1996: National Climatic Data Center, v. 108, nos. 9-10.
National Oceanic and Atmospheric Administration, 1996b, Local Climatological Data, Portland, ME, October 1996 (revised): National Climatic Data Center.
Rantz, S. E., and others, 1982, Measurement and computation of streamflow: U. S. Geological Survey Water Supply Paper 2175, 2 v., 631 p.
Sauer, V. B., Thomas, W. O., Jr., Stricker, V. A., and Wilson, K. V., 1983, Flood characteristics of urban watersheds in the United States: U. S. Geological Survey Water Supply Paper 2207, 63 p.
Shearman, J. O., 1990, Users Manual for WSPRO - A Computer Model for Water-Surface Profile Computations: Federal Highway Administration Report FHWA-IP-89-027, 177 p.
Flood of October 1996 in Southern Maine USGS WRIR 97-4189 28