Examination of Flood Characteristics at Selected Streamgages in the Meramec River Basin, Eastern Missouri, December 2015–January 2016

By Robert R. Holmes, Jr., Todd A. Koenig, Paul H. Rydlund, and David C. Heimann

Overview methods (submersible pressure transducer, non-submersible pressure transducer, or non-contact radar). These obser- Heavy rainfall resulted in major flooding in the Mera- vations are recorded autonomously at a predetermined mec River Basin in eastern Missouri during late December programmed frequency (typically either 15 or 30 minutes) 2015 through early January 2016. Cumulative rainfall from dependent on drainage-area size and concomitant flashiness December 14 to 29, 2015, ranged from 7.6 to 12.3 inches of the stream. Although stage data are important, stream- at selected precipitation stations in the basin with flood- flow data are equally or more important for streamflow ing driven by the heaviest precipitation (3.9–9.7 inches) forecasting, water-quality constituent loads computation, between December 27 and 29, 2015 (National Centers for flood-frequency analysis, and flood mitigation planning. Environmental Information, 2016). Financial losses from Streamflows are computed from recorded stage data using flooding included damage to homes and other structures, an empirically determined relation between stage and damage to roads, and debris removal. Eight of 11 coun- streamflow termed a “rating.” Development and verification ties in the basin were declared a Federal Disaster Area of the rating requires periodic onsite discrete measurements (Nelson, 2016). of streamflow throughout time and over the range of stages The U.S. Geological Survey (USGS), in coopera- to define local hydraulic conditions. tion with the U.S. Army Corps of Engineers and St. Louis The purpose of this report is to examine character- Metropolitan Sewer District, operates multiple streamgages istics of flooding that occurred in the Meramec River along the Meramec River and its primary tributaries Basin in December 2015–January 2016 including peak including the Bourbeuse River and Big River (fig. 1). stages, peak streamflows, and the flood-frequency statis- The period of record for streamflow at streamgages in the tics associated with the peak flows. A comparison between basin included in this report ranges from 24 to 102 years the December 2015–January 2016 flood and a similar (table 1). Instrumentation in a streamgage shelter auto- flood in December 1982 in the Meramec River Basin also matically makes observations of stage using a variety of is included.

Homes flooded by the Meramec River in Pacific, Missouri, December 29, 2015. Photograph by Doug Moore, St. Louis Post-Dispatch, used with permission.

U.S. Department of the Interior Open-File Report 2016–1140 U.S. Geological Survey 2 Examination of Flood Characteristics in the Meramec River Basin, Missouri, December 2015-January 2016

90°30'

Missouri Valley Park 07019130 # Eureka 07019000 91° Pacific # 38°30' # 07019300# Union 07016500

# 07017020

Arnold

#07018500

M

r

i e s

91°30' r s v e

i i R iv Dittmer 2.1 WSW s eus e R s rb i u c p o e B am p r i

e R M i v er e Ownsville 6.7 S iv r R

g i

B

Sullivan 07018100 #07014500 # #07015720 Vichy/Rolla National Airport

07013000 38° # Steelville 2 N

Steelville Rolla Potosi

St. James 10.1 SSE # #07017200 07010350 Caledonia 1.8 NNW

0 5 10 20 MILES

0 5 10 20 KILOMETERS

Base from U.S. Geological Survey 37°30' 1:24,000 Watershed Boundary Dataset, 2010

EXPLANATION Meramec River Basin 07010350 # U.S. Geological Survey streamgage and number Bourbeuse River Basin

National Climatic Data Center precipitation station Big River Basin and name Cities Main-stem Meramec River Basin

Figure 1. Meramec River Basin including major tributaries—Bourbeuse River and Big River. Characteristics of December 2015–January 2016 Flooding in the Meramec River Basin 3

Characteristics of December 2015– streamgages, no peak streamflow during the December 2015– January 2016 flood exceeded the peak streamflow for the January 2016 Flooding in the Meramec period of record, but the peaks generally were ranked in the top five at the selected stations(table 1). The streamgages at River Basin which the December 2015–January 2016 flood did not rank in the top five of streamflow peaks during the period of record Flood Summary Data by Streamgage were in the upstream portion of the basin. These streamgages included the Meramec River near Steelville, Missouri, and the Streamflow hydrographs for December 20, 2015– Big River at Irondale, Missouri, but the December 2015 peaks January 7, 2016, at selected main-stem and tributary at these streamgages still ranked in the top 10 for peak stream- streamgages in the Meramec River Basin show the upstream flow during the period of record. to downstream increase in streamflow magnitude and the temporal progression of streamflow peaks in the basin(fig. 2). Table 1. Peak stage and streamflow data at selected There were peak stages during the December 2015–January streamgages in the Meramec River Basin for the flood of 2016 flood in the basin that exceeded the period-of-record December 2015–January 2016 (available online at http://dx.doi. peak stage at selected streamgages (table 1). Of these selected org/10.3133/ofr20161140).

70,000 0 180,000 0 0.02 1.07 0.08 07019000 0.16 0.5 160,000 60,000 07016500 Meramec River 1 0.62 Bourbeuse River 1 140,000 0.92 near Eureka, 50,000 1.34 Missouri at Union, Missouri 1.5 120,000 2 (precipitation from (precipitation from 40,000 2 100,000 2.72 3 30,000 2.5 80,000 3 60,000 4 20,000 3.5 40,000 3.51 5 10,000 4.11 4 20,000 5.28 0 4.5 0 6 90°30' 30,000 0 60,000 0 07015720 0.5 0.73 25,000 0.5 1.53 91° 0.21 0.19 0.5 Bourbeuse River near 1 38°30 # 50,000 0.71 1 High Gate, Missouri 1.5 91°30' # 07018100 20,000 40,000 (precipitation from Vichy/ 2 Big River near 1.5 Rolla national Richwoods, Missouri 15,000 2.34 2.5 30,000 (precipitation from 2 3 2.5 10,000 3.5 # 20,000 # # 2.65 4 3 5,000 10,000 4.5 3.5 4.31 38° 3.41

0 5 0 4 tion, in inches 70,000 0 # #

07014500 30,000 0 ecipit a 60,000 0.21 0.5 Meramec River near 0.65 07017200 50,000 0.73 Sullivan, Missouri 1 25,000 0.5 (precipitation from Big River at 40,000 1.5 20,000 Irondale, Missouri 1 Daily p r (precipitation from 30,000 2 15,000 1.35 Caledonia 1.8 NNW, 1.5 1.54 1.5 20,000 2.5 10,000 2 2.83 10,000 3 3.29 5,000 2.5 0 3.5 0 2.82 3 20,000 0 12/20 12/23 12/26 12/29 1/1 1/4 1/7 18,000 0.49 0.5 0.53 0.49 16,000 07010350 1 Meramec River at 14,000 1.5 EXPLANATION 12,000 Cook Station, Missouri (precipitation from 2 10,000 St. James 10.1 SSE, 2.5 8,000 6,000 3 3.5 4,000 3.81 3.14 2,000 4 Daily precipitation and value 0 4.5 0.71 12/20 12/23 12/26 12/29 1/1 1/4 1/7

Figure 2. Streamflow hydrographs and precipitation distributions for selected streamgages and precipitation stations within the Meramec River Basin, December 20, 2015–January 7, 2016. 4 Examination of Flood Characteristics in the Meramec River Basin, Missouri, December 2015-January 2016

Comparison of 1982 and 2015 Floods For the upstream streamgages on the Meramec and Bourbeuse Rivers (first three streamgages listed in table 2), The heavy rainfalls that occurred in December 2015 in the December 1982 peak stages were higher than the Decem- the Meramec River Basin were similar in timing and mag- ber 2015 flood stages as would be expected with higher nitude to those in December 1982, which also resulted in streamflows and when the hydrodynamic conditions of the widespread flooding in the basin (Sauer and Fulford, 1983). channel (which would affect the relation between stage and Both events were associated with an El Niño climatic pattern streamflow) have not changed appreciably between floods. (National Oceanic and Atmospheric Administration, 2016), The December 1982 peak stages at the upstream Bourbeuse which is characterized by warm ocean temperatures in the and Meramec River streamgages were 0.72 to 1.25 feet higher equatorial Pacific Ocean and increased rainfall across the than in December 2015. For the downstream streamgages on southern tier of the United States (Earth Systems Research each river (table 2), the 1982 flood peak stages were 0.51 to Laboratory, 2016). A comparison of peak stages and stream- 4.24 feet lower than the December 2015 flood, and the Decem- flows during the December 1982 and December 2015 floods at ber 1982 streamflows generally were lower than in December selected locations within the Meramec River Basin is shown 2015. The exception was the Bourbeuse River at Union, Mis- in table 2. The December 1982 flood peak streamflows were souri, streamgage with a peak stage that was 0.51 feet higher greater than the December 2015 flood peak streamflows at in December 2015 than in December 1982 despite a lower upstream Meramec River and Bourbeuse River streamgages peak streamflow. A higher peak stage for a lower peak stream- (first four streamgages listed intable 2), whereas the Decem- flow is the result of changing hydrodynamic conditions of ber 2015 streamflows were greater than December 1982 peak the channel, either from downstream backwater effects from flows at the Meramec River near Eureka and all Big River tributaries or the Meramec River, the timing of the flood wave, streamgages. Streamflows in the Bourbeuse and upper Mera- or physical changes to the downstream channel. mec River Basins were 1,700 to 21,300 cubic feet per second (ft3/s) higher in December 1982 compared to December 2015, Table 2. Comparison of peak stages from the December 1982 whereas flows in the lower Meramec River and Big River and December 2015 floods at selected streamgages in the Basins were 8,600 to 17,000 ft3/s higher in December 2015 Meramec River Basin (available online at http://dx.doi.org/10.3133/ compared to December 1982. ofr20161140).

Flooding from the Bourbeuse River in Union, Missouri, December 29, 2015. Photographs by Huy Mach, St. Louis Post-Dispatch, used with permission. Characteristics of December 2015–January 2016 Flooding in the Meramec River Basin 5

Flood-Frequency Analyses The current (2016) standard methodology for the determi- nation of flood-frequency studies is Bulletin 17B of the Inter- Flood-frequency estimates for the Bourbeuse River agency Advisory Committee on Water Data (U.S. Interagency at Union, Missouri, and the Meramec River near Eureka, Advisory Committee on Water Data, 1982). The Bulletin 17B Missouri, are shown in table 3. In the past, flood-frequency method fits a log-Pearson Type III distribution curve to the estimates have been presented in terms of the recurrence inter- logarithms of annual peak discharges at a given station using val, which represents the average number of years between the method-of-moments to compute a mean, standard devia- occurrences of a streamflow of equal or greater magnitude. tion, and station skew of the log-transformed peak flow data. This term, however, often results in a misunderstanding that The user has the option to weigh the individual station skew a 100-year flood, once it has occurred, will not occur again estimate with a generalized/regional skew estimate, which for another 99 years, which is not necessarily true. A 100-year typically improves the accuracy because skews tend to follow flood means that, if all things remain the same in the basin regional trends. At the time of this report, modest changes to (for example, land use and climate conditions), the observed Bulletin 17B, recommended by the Advisory Committee on annual floods during thousands of years would average around Water Information (http://acwi.gov), are being drafted into 100 years between events of this magnitude. For example, Bulletin 17C (http://acwi.gov/hydrology/Frequency/b17c/bul- sometimes it could be 200 years and other times it could be letin17c_draft_for_public_review.pdf ). Modifications include 30 years between “100-year” magnitude flood events during the adoption of a generalized method-of-moments estimator, those thousands of years. The term annual exceedance prob- known as the expected moments algorithm (EMA) procedure ability (AEP) is now more commonly used to refer to flood- (Cohn and others, 1997), and a generalized version of the frequency estimates because it more clearly conveys that the Grubbs-Beck test for low outliers (Cohn and others, 2013). flood-frequency estimates are probabilistic in nature. The AEP The EMA is an updated method for fitting the log-Pearson is the probability that the streamflow of a certain magnitude Type III frequency distribution that has been shown to be a will be equaled or exceeded in any year and is the reciprocal more effective means of incorporating historical peak stream- of the recurrence interval. A 100-year flood streamflow, for flow information into a flood-frequency analysis. example, is the same as a streamflow having a 0.01 AEP, or a flood streamflow with a 1-percent chance of occurring in any year.

Flooding from the Meramec River near Eureka, Missouri, December 30, 2015. Photographs by David Carson, St. Louis Post-Dispatch, used with permission. 6 Examination of Flood Characteristics in the Meramec River Basin, Missouri, December 2015-January 2016

The USGS computer program PEAKFQ (version 7.1; 2-percent, 1-percent, 0.5-percent, and 0.2-percent AEP flood Flynn and others, 2006) was used to compute the flood-fre- quantiles for each streamgage, along with the 95-percent quency estimates for the Bourbeuse River at Union, Missouri, confidence interval range for each quantile, are also shown and the Meramec River near Eureka, Missouri. The program in table 3. The observed peak streamflow of 64,000 ft3/s falls automates many of the flood-frequency analyses procedures, within the estimated 1-percent annual exceedance flow of including identifying and adjusting for high and low outli- 55,700 ft3/s and the estimated 0.5-percent annual exceedance ers and historical periods, and fitting the log-Pearson Type flow of 64,400 ft3/s based on 103 years of peak record used in III distribution to the streamflow data. The program includes the analysis. the EMA procedure for flood-frequency analysis and multiple Grubbs-Beck outlier screening (Veilleux and others, 2013). Table 3. Summary of peak stages, streamflows, and flood- The PEAKFQ program and associated documentation are frequency estimates for the December 2015 flood at selected available at http://water.usgs.gov/software/peakfq.html. Meramec River Basin streamgages (available online at http:// The estimated AEP corresponding to the December dx.doi.org/10.3133/ofr20161140). 2015 flood at the Bourbeuse River at Union, Missouri, is 0.52 percent (recurrence interval of 191 years), with the The estimated AEP corresponding to the December 2015 uncertainty of this estimate, expressed by the 66.7-percent flood at the Meramec River near Eureka, Missouri, is 1.1 per- confidence interval for the true exceedance probability, being cent (recurrence interval of 91 years). A 66.7-percent confi- from 0.7 (recurrence interval of 143 years) to 3.1 percent dence interval for the true annual exceedance probability of (recurrence interval of 32 years) (table 3). The 66.7-percent the flood extends from 0.7 (recurrence interval of 143 years) confidence interval for the AEP is a non-parametric estimate to 3.2 percent (recurrence interval of 31 years). The observed based on a flood of a given rank within a specified period peak streamflow of 162,000 ft3/s falls within the estimated of record. These confidence intervals are not centered about 2-percent annual exceedance flow of 140,000 ft3/s and the esti- the interpolated AEP estimate. Consequently, the range of mated 1-percent annual exceedance flow of 165,000 ft3/s based the confidence interval may exclude the AEP estimate. The on 99 years of peak record used in the analysis.

Flooding from the Meramec River in Valley Park, Missouri. Upper left photograph taken December 30, 2015, by David Carson, St. Louis Post-Dispatch, used with permission. Lower right photograph taken December 31, 2015, by J.B. Forbes, St. Louis Post-Dispatch, used with permission. References Cited 7

References Cited National Weather Service, 2016, Advanced hydrologic predic- tion service: National Oceanic and Atmospheric Adminis- tration, accessed May 16, 2016, at http://water.weather.gov/ Cohn, T.A., England, J.F., Berenbrock, C.E., Mason, R.R., ahps/. Stedinger, J.R., and Lamontagne, J.R., 2013, A generalized Grubbs-Beck test statistic for detecting multiple potentially Nelson, Alisa, 2016, FEMA approves Nixon’s request to influential low outliers in flood series: Water Resources expand Missouri’s federal disaster declaration: Missou- rinet, accessed May 18, 2016, at http://www.missourinet. Research, v. 49, no. 8, p. 5047–5058. com/2016/02/10/fema-approves-nixons-request-to-expand- Cohn, T.A., Lane, W.L., and Baier, W.G., 1997, An algorithm missouris-federal-disaster-declaration/. for computing moments-based flood quantile estimates Sauer, V.B., and Fulford, J.M., 1983, Floods of December when historical flood information is available: Water 1982 and January 1983 in central and southern Mississippi Resources Research, v. 33, no. 9, p. 2089–2096. River Basin: U.S. Geological Survey Open-File Report 83–213, 41 p. Earth Systems Research Laboratory, 2016, Multivariate ENSO Index (MEI): National Oceanic and Atmospheric Admin- U.S. Army Corps of Engineers, 2016, River and reservoir istration, accessed May 18, 2016, at http://www.esrl.noaa. daily report, St. Louis District: accessed May 13, 2016, at gov/psd/enso/mei/. http://mvs-wc.mvs.usace.army.mil/dresriv.html.

Flynn, K.M., Kirby, W.H., and Hummel, P.R., 2006, User’s U.S. Geological Survey, 2016, National Water Information manual for program PeakFQ, Annual flood frequency analy- System: U.S. Geological Survey Water Data for Missouri, sis using Bulletin 17B guidelines: U.S. Geological Survey accessed May 13, 2016, at http://waterdata.usgs.gov/mo/ nwis/nwis. Techniques and Methods, book 4, chap. B4, 42 p. U.S. Interagency Advisory Committee on Water Data, 1982, National Centers for Environmental Information, 2016, Data Guidelines for determining flood flow frequency (revised): access: accessed May 16, 2016, at http://www.ncdc.noaa. U.S. Geological Survey, Office of Water Data Coordination, gov/data-access. Bulletin 17B, 28 p., 14 app., and 1 pl. National Oceanic and Atmospheric Administration, 2016, Veilleux, A.G., Cohn, T.A., Flynn, K.M., Mason, R.R., and Explaining El Niño—Impacts: El Nino Theme Page, Pacific Hummel, P.R., 2013, Estimating magnitude and frequency Marine Environmental Laboratory, accessed May 18, 2016, of floods using the PeakFQ 7.0 program: U.S. Geological at http://www.pmel.noaa.gov/elnino/impacts-of-el-nino. Survey Fact Sheet 2013–3108, 2 p.

Homes flooded by the Meramec River in Arnold, Missouri, December 31, 2015. Photograph by David Carson, St. Louis Post-Dispatch, used with permission. For more information concerning this publication, contact: Chief, USGS Office of Surface Water 415 National Center 12201 Sunrise Valley Drive Reston, VA 20192 (703) 648-5301 Or visit the Office of Surface Water Web site at: http://water.usgs.gov/osw/

For more information on the USGS—the Federal source for science about the Earth, its natural and living resources, natural hazards, and the environment—visit http://www.usgs.gov or call 1–888–ASK–USGS. For an overview of USGS information products, including maps, imagery, and publications, visit http://store.usgs.gov. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Although this information product, for the most part, is in the public domain, it also may contain copyrighted materials as noted in the text. Permission to reproduce copyrighted items must be secured from the copyright owner.

Suggested citation: Holmes, R.R., Jr., Koenig, T.A., Rydlund, P.H., and Heimann, D.C., 2016, Examination of flood characteristics at selected streamgages in the Meramec River Basin, Eastern Missouri, December 2015–January 2016: U.S. Geological Survey Open-File Report 2016–1140, 7 p., http://dx.doi.org/10.3133/ofr20161140.

ISSN 2331-1258 (online) Publishing support provided by Rolla Publishing Service Center http://dx.doi.org/10.3133/ofr20161140