Hydrology Design Report

Transcription

1 MONTANA DEPARTMENT OF NATURAL RESOURCES AND CONSERVATION Upstream of Cold Creek Rd. Bridge, 9/18/14 Hydrology Design Report Swan River Detailed Floodplain Study Missoula County, MT By the Montana Department of Natural Resources and Conservation May 2015

2 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 HYDROLOGY DESIGN REPORT SWAN RIVER Missoula County, MT Table of Contents 1.0 INTRODUCTION LiDAR Collection Watershed Description Effective FIS (Flood Insurance Study) Hydrology Historic Data HYDROLOGIC ANALYSIS Stream Gage Analyses Systematic Estimation Two Station Comparison & MOVE Regional Regression Equations Weighted Estimation Results Comparison Transfer to Ungaged Sites Streamflow Change Locations Log Interpolation between 2 Gages HYDROLOGIC ANALYSIS SUMMARY & SELECTED DISCHARGES List of Tables Table 1: Flathead & Lake County Effective Floodplain Studies... 2 Table 2: Peak Flows used in Effective Studies... 3 Table 3: Highest Peaks Recorded at Swan River Stream Gages... 5 Table 4: Stream Gage Summary Table... 6 Table 5: Stream Gage Analysis Results Systematic... 7 Table 6: Record Extension Methods Discharge Estimates Stream Gage No Table 7: Regression Equation Estimates Stream Gage No Table 8: Weighted Estimates Stream Gage No Table 9: Flood Discharge Estimates at USGS Gage Table 10: Gage Trending Analysis Comparison Table 11: Hydraulic Model Stream Reaches i

3 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Table 12: Log Interpolation between 2 Gages Table 13: Selected Discharge Estimates List of Figures Figure 1: Study Reach Site Map... 4 Figure 2: Flood Discharge Estimates at USGS Gage Figure 3: Drainage Basin Area Appendices Appendix A: USGS Stream Gage Data Appendix B: Hydrologic Calculations ii

4 Hydrology Design Report, Swan River Detailed Floodplain Study May INTRODUCTION As part of the Missoula County Memorandum of Understanding (MOU) (Reference 9) Agreement activities, DNRC is assisting Missoula County to complete a new detailed riverine floodplain study, with floodway, for a section of the Swan River beginning at the Lake/Missoula County boundary and extending approximately 19.1 miles upstream near the Beaver Creek confluence. Per the MOU, DNRC has completed a new hydrologic analysis to be utilized for the new study. The study reach is currently mapped by FEMA using approximate methods. This study will be completed satisfying current state and FEMA standards for Detailed/Enhanced floodplain studies. A Flood Insurance Study (FIS) has been completed for Missoula County and Incorporated Areas dated August 16, 1988; however, an update to the FIS has been completed and is slated to become effective on July 6, A summary of the Digital Flood Insurance Rate Map (DFIRM) panels covering this study reach is presented below (Reference 2): Community Community No. FIRM Panels Effective Date Missoula County, MT C0200D 30063C0075D 30063C0050D 30063C0200E 08/16/ C0075E 30063C0050E 07/06/2015 This report summarizes the hydrologic analysis and results for the new detailed study stream reach described. The new study includes hydrologic analysis to estimate the 10-, 4-, 2-, 1-, and 0.2 percent-annual-chance flood discharges for the Swan River. 1.1 LiDAR Collection In 2012, Missoula County contracted with Watershed Sciences, Inc. to collect Light Detection And Ranging (LiDAR) data for four areas of interest for the purpose of supporting floodplain mapping projects: 1. Rock Creek floodplain, 2. Bitterroot River floodplain, 3. Swan River floodplain, and 4. Clearwater River floodplain. Accuracy of the topographic data meets FEMA standards for detailed level floodplain mapping. The LiDAR data covers the project reach for this study. 1.2 Watershed Description The Swan River originates from Gray Wolf Lake in the Mission Mountain Range and flows generally in a northwesterly direction before emptying into Flathead Lake at the town of Bigfork. 1

5 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 At its mouth, the river drains a watershed area of approximately 723 square miles. The Swan River valley bottom was formed by glacial melt waters and reworked by fluvial processes. The valley is bounded on the east by the Swan Mountain Range and on the west by the Mission Range. The vegetation in the valley is dominated by subalpine firs with slopes ranging from 0 to 20 percent. 1.3 Effective FIS (Flood Insurance Study) Hydrology The effective floodplain mapping for the Swan River in Missoula County stretches from the Lake County boundary upstream to the Beaver Creek confluence, which matches the extents of the study at hand. The level of detail of the effective mapping is approximate, as it has been carried forward from the historic flood hazard boundary maps (FHBMs) developed in the 1970s. In Flathead and Lake Counties, the Swan River floodplain has both approximate level and detailed level mapping; information about the effective detailed studies is shown in Table 1. Table 1: Flathead & Lake County Effective Floodplain Studies Author Study Title County Effective Date Approx. Length (mi) Level of Detail Description Simons Li & Associates, Inc. Lower Swan River Study Lake County March Detailed Beginning at the Swan River Rd. Bridge upstream of Bigfork Dam and extending upstream to the outlet of Swan Lake. Simons Li & Associates, Inc. Upper Swan River Study Lake County March Detailed Beginning approx. 6.5 miles upstream of the Swan Lake inlet and extending upstream to the Missoula Co. boundary. PBS&J Swan Lake Lake County September Limited Detail Analysis of the Swan Lake floodplain including 23 miles of shoreline. DNRC Swan River at Bigfork Flathead County December Detailed Beginning at the river mouth and Flathead Lake confluence and extending upstream 0.5 miles. Table 1 is a summary of the effective detailed level studies on the Swan River. There are also several reaches of the channel with approximate level mapping which are remnants of the FHBMs from the 1970s. Each of the effective detailed studies included a hydrologic analysis including estimation of peak flows. The following is a description of the hydrologic analysis method selected for estimating peak flows for each study: Lower Swan River Study A Log-Pearson Type III analysis was performed on the annual maximum discharges to determine peak return flows (Reference 4). Upper Swan River Study A regional frequency method was used to develop peak flows for the Swan River above Swan Lake. The USGS developed this methodology for the entire state, and discharge estimates are a function of 2

6 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 drainage areas, average annual precipitation, and location within the watershed (Reference 4). Swan Lake Procedures recommended in Bulletin 17B of the Interagency Committee on Water Data and the USGS Water Resources Investigation Report were used to develop peak flows (Reference 4). Swan River at Bigfork - A Log-Pearson Type III analysis was performed on the annual maximum discharges to determine peak return flows (Reference 3). As part of the hydrologic analyses for each of these studies, peak flows were calculated at various locations throughout the study reaches. Table 2 shows peak flow values at specified locations within the effective studies. Table 2: Peak Flows used in Effective Studies Location Drainage Area (mi 2 ) 10% Annual Chance Discharges (cfs) 2% Annual Chance 1% Annual Chance 0.2% Annual Chance Applicable FIS / Effective Year At mouth ,420 9,080 9,760 11,300 At USGS Gage Near Bigfork ,200 8,500 9,000 10,000 At Swan Lake ,860 8,460 9,120 10,700 At Wildlife Refuge Boundary At confluence with Goat Creek At confluence with Piper Creek At Missoula County Line ,950 5,150 5,700 6, ,540 4,630 5,130 6, ,070 4,040 4,490 5, ,440 3,230 3,610 4,440 Flathead County / 2015 Lake County / 2013 Lake County / 2013 Lake County / 2013 Lake County / 2013 Lake County / 2013 Lake County / 2013 The downstream end of the study at hand will tie into the upstream end of the Upper Swan River Study; the peak flows used at the upstream end of the Upper Swan River Study are shown in the row titled At Missoula County Line of Table Historic Data The most severe flooding on the Swan River typically occurs in the spring and early summer as a result of snowmelt and/or rainfall runoff. Localized flooding can occasionally result from a long sustained rainfall or ice or log jams (Reference 4). There are two USGS stream flow gages present on the Swan River (see Figure 1): 1) Gage is an inactive gage located at the upstream end of the river near the town of Condon, MT with 20 years of record ( ); 2) Gage is an active gage near the town of Bigfork with 92 years of record ( ). The five highest annual peak discharges recorded at each gage are listed in Table 3. 3

7 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Figure 1: Study Reach Site Map 4

8 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Table 3: Highest Peaks Recorded at Swan River Stream Gages Ranking USGS Gage No near Condon (69.1 mi 2 ) Date Peak Discharge (cfs) USGS Gage No near Bigfork (671.0 mi 2 ) Date Peak Discharge (cfs) 1 June 18, ,540 June 20, ,890 2 June 16, ,020 May 18, ,520 3 May 26, ,090 May 24, ,400 4 June 17, ,060 June 18, ,280 5 May 31, ,090 June 10, ,100 Comparing the dates of the peak flows at the two stations, only one of the top five peaks at the downstream gage ( ) occur within the operation period of the upstream gage ( ), which is the 1974 event and is the highest recorded flow at both stations. Statistically, the fact that the top ranked event at the downstream gage is captured within the period of record at the upstream gage is significant; the magnitude of the event will aid in increasing the confidence of the upper end of the Log Pearson Type III curve of the Systematic analysis (discussed in Section 2.1.1). For the purpose of documenting historic flood records, county officials were contacted; the only records of historic flood events through the project reach are aerial photographs of the 1997 flood event. These photos should be referenced for comparison purposes during the hydraulic analysis and floodplain mapping phases of this project; the Missoula County Floodplain Administrator has these files on record. 2.0 HYDROLOGIC ANALYSIS The new detailed study project covers approximately 19.1 miles of the Swan River, beginning at the Missoula/Lake County boundary and extending upstream to the Beaver Creek confluence (see Figure 1). This section of the report describes the various hydrologic analysis methods evaluated and the results of the flood discharges estimated for the study reach. The selected method(s) and estimated discharges are discussed in Section Stream Gage Analyses As previously mentioned, there are two USGS stream flow gages present on the Swan River (see Figure 1): Gage is located at the upstream end of the study reach and gage is located near the mouth of the river. A summary of the Swan River gages is provided in Table 4. Based on USGS WRIR (Reference 11), recurrence interval discharge estimates at gaging stations can be transferred to ungaged sites, such as at selected 5

9 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 flow change locations throughout the study reach. The transfer methodology will be discussed in Section 2.2 of this report. Table 4: Stream Gage Summary Table Gage Name Swan River near Condon, MT Swan River near Bigfork, MT Gage Number Basin Area (mi 2 ) Period of Record Number of Annual Peaks Highest Peak of Record / Year Recorded ,540 / ,890 / 1997 Prior to transferring discharges to the study reach, recurrence interval flows at the stream gages must be estimated. There are multiple ways to calculate peak flow estimates at the stream gages, and the following methods for estimating recurrence interval peaks at USGS gages and were analyzed: Gage Swan River near Bigfork, MT Systematic estimation Gage Swan River near Condon, MT Systematic estimation, 2 Station Comparison (record extension based on ), MOVE.1 analysis (record extension based on ), Regional Regression, and Weighted estimation. Estimation methods differ between the two gages due to the difference in reliability of the periods of record. Gage has a substantial period with 92 years while gage is less reliable with 20 years. To aid in the evaluation of the reliability of the 20 years of record at gage , the five peak flow estimation methods listed above were compared. The following sections provide a description of each estimation method Systematic Estimation The Systematic analysis involves development of a flood flow frequency curve using a Log- Pearson Type III (LP3) distribution based on recorded annual stream flow peaks following Bulletin 17B guidelines (Reference 7). The flood frequency analysis was performed using the USGS s PeakFQ v 7.1 program. The annual peak flow record was downloaded from the USGS website for the stream gage in the WATSTORE text format for PeakFQ input and annual peak flow frequency analysis. See Appendix B for PeakFQ output files. Prior to performing the new analysis, a baseline analysis was run for each gage in an effort to reproduce previously published flows. For gage , the most recent analysis was reported in the Swan River at Bigfork Hydrology Design Report (Reference 8) based on data 6

10 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 through 2011; and for gage , the most recent analysis known is that of WRIR through water year The results were successfully duplicated using a Generalized Skew Standard Error value of 0.64, the weighted skew method, and no defined outliers. The new LP3 flood frequency analysis for gage incorporated the current 92 year ( ) period of record for the gage and used the same options and parameters as the baseline analysis: a Generalized Skew Standard Error of 0.64, the weighted skew method, and no defined outliers. USGS personnel from the state office in Helena are currently recommending the value of 0.64 for Generalized Skew Standard Error. An updated analysis of gage is unnecessary since the gage was discontinued after 1973 and the entire period of record was captured in the previous analysis. In accordance with FEMA guidelines (Reference 1), no expected probability adjustments were made to the Bulletin 17B frequency curve. Results of the flood frequency analysis are presented in Table 5. Table 5: Stream Gage Analysis Results - Systematic Gage Description Period of Record Percent Annual Chance Peak Discharge 10% 4% 2% 1% 0.2% Baseline Analysis Updated / Current Analysis (90yrs) (92yrs) 6,690 7,870 8,540 9,190 10,700 6,960 7,860 8,500 9,130 10, Current Analysis (20yrs) 1,180 1,360 1,480 1,600 1, Two Station Comparison & MOVE.1 Comparing the dates of the five highest peak flows at each stream gage (Table 3), it is likely that some of the largest flood events on the Swan River were not captured in the 20 years of record at gage ; four of the highest top five peaks at gage are outside of the gage period of record. Therefore, the peak discharges estimated using the Systematic analysis method may not be reflective of a longer period of record; which provides greater reliability. To strengthen the period of record of gage there are several methods which can be used to extend the record by utilizing gages with long, overlapping periods of record on the same stream. The two record extension methods compared in this analysis were the Two Station Comparison (Reference 7), and a Maintenance Of Variance Extension Type 1 (MOVE.1) analysis (Reference 6). The method of the Two Station Comparison is to perform a regression on the concurrent peak flows measured at both gages and use the correlation to improve estimates of the mean and standard deviation of the short-term gage record. Similar to the Two Station Comparison, the MOVE.1 analysis also results in modifications to the sample mean and variance. However, rather than directly changing these parameters based on correlation of the data between the two gages, the procedure of the MOVE.1 analysis is to synthesize the 7

11 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 individual peaks of the target station based on a correlation analysis of the concurrent data between the two gages; and the distributional properties of the final dataset reflect adjustment of the mean and variance. Discharge estimates based on these two record extension methods at gage are shown in Table 6. Table 6: Record Extension Methods Discharge Estimates - Stream Gage No Recurrence Interval Discharge Estimates (cfs) Two Station Comparison MOVE.1 10yr 1,374 1,210 25yr 1,648 1,373 50yr 1,856 1, yr 2,068 1, yr 2,584 1,862 The MOVE.1 analysis was completed by local USGS personnel, and all calculations were performed in the USGS FORTRAN program. Calculations of the Two Station Comparison and output files for the MOVE.1 analysis are provided in Appendix B Regional Regression Equations Flood frequency discharges were estimated using USGS published regional regression equations derived from basin characteristics (Reference 11). The basin characteristics equations have been developed using characteristics of drainage basins within designated regions throughout the state with similar hydrologic properties. The study reach for this project is located in the West region, and the regression equations for the region are based on drainage basin characteristics variables of 1) area, 2) average precipitation, and 3) percent of the basin that is forested. All of the data needed to perform the regression analysis at USGS gage has been provided by USGS in WRIR (Reference 11). Results of the basin characteristics regression analysis are provided in Table 7. Table 7: Regression Equation Estimates - Stream Gage No Recurrence Interval Discharge Estimates (cfs) Basin Active Characteristics Channel Bankfull Width 10yr 1,671 3,160 3,030 25yr 1,972 3,780 3,640 50yr 2,260 4,290 4, yr 2,580 4,820 4, yr 3,437 6,140 5,950 8

12 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Similar to the regression equations derived from basin characteristics, USGS WRIR provides regression equations based on the stream active channel width and bankfull width for each region. Channel measurements representative of the stream gage locations are published in WRIR , and the discharge estimates based on the channel characteristics regression equations are shown in Table 7. The Standard Error of Prediction (SEP) of the basin characteristics equations is significantly lower than the channel characteristics equations, and therefore, the channel characteristics estimates were not included in the final comparison of discharge estimates at gaging station Weighted Estimation The final methods employed for estimating flood discharges at gage are two procedures used to combine various estimation methods. The first procedure combines the Systematic estimates with the Basin Characteristics regression estimates. According to WRIR (Reference 11), the Systematic discharge estimates may have large errors due to timesampling, especially at gages with shorter periods of record. To improve the Systematic estimates, Bulletin 17B (Reference 7) states that the Systematic estimates can be weighted with the regression estimates. Using equation 22 in WRIR , the discharges in Table 8 were estimated. Table 8: Weighted Estimates - Stream Gage No Recurrence Interval Systematic + Basin Characteristics Regression Discharge Estimates (cfs) Basin Characteristics Regression + Channel Characteristics Regression 10yr 1,216 2,160 25yr 1,421 2,550 50yr 1,567 2, yr 1,715 3, yr 2,075 4,010 The second procedure combines the estimates of all of the regression equations (basin characteristics and channel characteristics). This procedure was performed through the use of an online calculator developed by the USGS (Reference 16). These results are also shown in Table Results Comparison Due to the reliability of the 92 years of record at gage , the Systematic method was the only method selected for estimating discharges at gage Provided in Table 9 and Figure 2 are the results of the five methods used to estimate discharges at gage

13 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Table 9: Flood Discharge Estimates at USGS Gage Discharge Estimates (cfs) Recurrence Interval Systematic Two Station Comparison MOVE.1 Weighted 1* Weighted 2** Basin Characteristics Regression 10yr 1,180 1,374 1,210 1,216 2,160 1,671 25yr 1,360 1,648 1,373 1,421 2,550 1,972 50yr 1,480 1,856 1,489 1,567 2,850 2, yr 1,600 2,068 1,603 1,715 3,170 2, yr 1,890 2,584 1,862 2,075 4,010 3,437 * Weighted 1 Systematic weighted with Basin Characteristics regression ** Weighted 2 Basin Characteristics regression weighted with Channel Characteristics regressions (active channel & bankfull width) Figure 2: Flood Discharge Estimates at USGS Gage Estimates from the Systematic computations and the MOVE.1 analysis are nearly identical, while the Two Station Comparison, Basin Characteristics Regression, and Weighted regressions estimates are significantly higher. As one would intuitively suspect, the Systematic 10

14 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 weighted with the Basin Characteristics regression values are between the Systematic and Basin Characteristics Regression estimates. The initial step in both of the record extension methods (Two Station Comparison and MOVE.1) is calculation of a correlation coefficient for the purpose of determining whether the analysis is a reliable method for improving the dataset. A satisfying coefficient was calculated for both methods. Therefore, the record extension methods were assumed to be more reliable than the Systematic method. Furthermore, given the reliability of the downstream gaging station ( ) with 92 years of record, and the ability to improve the record of the upstream gage ( ) using gage as a base station, the record extension methods were also determined to be more reliable than the regression equation and weighted estimates for estimating peak flows at gage The MOVE.1 and Two Station Comparison analyses are both industry-accepted record extension methods, and neither method is preferred over the other. Comparing the 100-year event estimated using each method, the Two Station Comparison estimates are approximately 25 percent greater. In order to determine which method is more appropriate for this study, the following trending analysis was performed on gage near Bigfork, as recommended by USGS personnel: The 92-year period of record was truncated to match the 20-year record at gage near Condon. The Systematic analysis was then rerun using the truncated period of record. Comparison of the truncated Systematic results to the unmodified period of record Systematic results may provide aid in selection of the appropriate record extension method. The results of the truncated and unmodified periods of record Systematic analysis are shown in Table 10. Table 10: Gage Trending Analysis Comparison Description Complete Period of Record Truncated Period of Record Period of Record (92yrs) (20yrs) Percent Annual Chance Peak Discharge 10% 4% 2% 1% 0.2% 6,960 7,860 8,500 9,130 10,600 6,713 7,615 8,276 8,930 10,450 The results of the truncated period of record Systematic analysis are slightly (~2.5 percent) less than the results of the unmodified period of record Systematic analysis. In other words, the additional 72 years of record only slightly increases the flow estimates at this station. Assuming that this trend would also be present at the upstream gaging station if the additional 72 years of record were included in the Systematic analysis, there would be a minimal increase in the discharge estimates. Due to the much smaller flow magnitudes at the upstream gage compared to the downstream gage, the flow increases at the upstream gage would be even smaller than the increases at the downstream gage. 11

15 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Since the discharge estimates of the MOVE.1 analysis are more similar to the results of the Systematic analysis at gage and based on the results of this trending analysis, the discharge estimates of the MOVE.1 analysis have been selected for use at gaging station Also noteworthy, the USGS is currently under contract with the Montana Department of Transportation (MDT) and the Montana DNRC to perform an update to the recommended methods for estimating flood frequency in Montana based on data through water year One deliverable of the project is an updated analysis of all the gaging stations throughout the state; this deliverable includes flood frequency estimates at all gages based on Systematic analysis, record extension methods, and Systematic estimates weighted with regression estimates, as deemed necessary. In the draft report, the flood frequency flows recommended at station are much closer to the MOVE.1 results than the Two Station Comparison results included in this report. 2.2 Transfer to Ungaged Sites As described, the new study reach is located shortly downstream of USGS gage and upstream of USGS gage As documented in WRIR , the recommended procedure for estimating discharges in between two gaging stations located on the same stream is a log interpolation between the two gages (see Section 2.2.2). Identification of the appropriate transfer locations throughout the study reach is discussed in the following section Streamflow Change Locations There are several significant tributaries to the Swan River throughout the study reach. For the purpose of developing a one-dimensional steady flow hydraulic model, the stream study reach is divided into separate sections based on the flow estimated to be in the floodplain. The study reach was divided into reaches which span from immediately upstream of a tributary confluence to immediately upstream of the adjacent tributary confluence. This methodology allows the additional flow resulting from the tributary hydrology to be applied in the hydraulic model at the appropriate location. The model may also be divided at other locations to reflect changes in stream flow. The major tributaries throughout the study reach include: Holland Creek, Buck Creek, Glacier Creek, and Cold Creek. Drainage basins for each of the tributaries are shown in Figure 3 and a summary of the different model reaches is provided in Table 11. The following section describes the method used to transfer the peak flows estimated at the gaging stations to the individual reaches throughout the study reach. 12

16 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Figure 3: Drainage Basin Area 13

17 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Reach River Station 1 (ft) Table 11: Hydraulic Model Stream Reaches Reach Length (ft) ,050 11,050 Subbasin Name Cold Creek (downstream study limit) Subbasin Area (mi 2 ) Cumulative Basin Area (mi 2 ) ,050 37,800 26,750 Glacier Creek ,800 56,140 18,340 Buck Creek ,140 86,670 30,530 Holland Creek ,670 99,370 12,700 Upstream Study Extents Distance above Missoula/Lake County boundary Log Interpolation between 2 Gages As previously mentioned, WRIR documents an interpolation estimation method for ungaged sites located between two gaged sites on the same stream, which is the case for this study. The WRIR equation is as follows (Reference 11, equation 21): As mentioned in Section 2.1.5, the most reliable flood frequency estimates at gage are the Systematic estimates, and the most reliable estimates at gage are the MOVE.1 estimates. Therefore, based on these estimates at each gaging station, peak flow estimates in Table 12 were calculated using the interpolation equation. 14

18 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 Table 12: Log Interpolation between 2 Gages Recurrence Interval Cold Creek Subbasin Glacier Creek Subbasin Discharge Estimates (cfs) Buck Creek Subbasin Holland Creek Subbasin Upstream Study Extents Subbasin 10yr 3,465 3,062 2,056 1,727 1,230 25yr 3,920 3,465 2,329 1,958 1,396 50yr 4,245 3,754 2,524 2,122 1, yr 4,564 4,036 2,716 2,283 1, yr 5,287 4,677 3,150 2,650 1, HYDROLOGIC ANALYSIS SUMMARY & SELECTED DISCHARGES Stream flow gage records are typically the most reliable hydrologic information when the periods of record are of sufficient length. There are two gages on the Swan River; one near the upstream end of the study reach and one significantly downstream of the study reach. Transfer of the recurrence interval peak flow estimates from the gaging stations to the study reach is considered to be the most reliable estimation method, and interpolation of the flows between the two gages on the stream is the selected transfer methodology. Due to the 92 years of record, the Systematic estimation method was selected for the purpose of estimating flows at gage Swan River near Bigfork. With only 20 years of record, several methods were analyzed for estimating flows at gage Swan River near Condon; ultimately, the most appropriate method for estimating flows at gage was the MOVE.1 record extension method using gage as a base gage for the extension. The selected flows for this study reach have been rounded to the nearest 10 cfs and are shown in Table 13. Upstream Limit of Applicability At Missoula / Lake County Boundary Immediately upstream of Cold Creek confluence Immediately upstream of Glacier Creek confluence Immediately upstream of Buck Creek confluence Immediately upstream of Holland Creek confluence Table 13: Selected Discharge Estimates River Station Applicability Limits 1 (ft) 1 Distance above Missoula/Lake County boundary. Drainage Area (mi 2 ) Peak Discharge Estimates (cfs) 10yr 25yr 50yr 100yr 500yr 0 11, ,470 3,920 4,250 4,560 5,290 11,050 37, ,060 3,470 3,750 4,040 4,680 37,800 56, ,060 2,330 2,520 2,720 3,150 56,140 86, ,730 1,960 2,120 2,280 2,650 86,670 99, ,230 1,400 1,510 1,630 1,890 The downstream end of this study reach ties into an effective detailed study in Lake County titled the Upper Swan River Study which became effective in March of The 100-year 15

19 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 flow in the Upper Swan River Study (Table 2) is 3,610 cfs, significantly lower than the flow recommended for use in the Cold Creek Subbasin reach of this study, which is 4,560 cfs. This study will result in an increase in the predicted flow of the 100-year event through this reach of 26 percent. Differences in these flow estimates can likely be attributed to the difference in hydrologic analysis methods and additional years of record at the analyzed gages. Peak flows in the Upper Swan River Study were estimated using regional frequency methods, and the gages used in that analysis may have had up to 27 years of record less than the downstream gage ( ) used in this analysis. 16

20 Hydrology Design Report, Swan River Detailed Floodplain Study May 2015 REFERENCES 1. Federal Emergency Management Agency (FEMA), 2009, Guidelines and Specifications for Flood Hazard Mapping Partners, Appendix C: Guidance for Riverine Flooding Analyses and Mapping, 81 p. 2. Federal Emergency Management Agency (FEMA), Digital Flood Insurance Rate Maps: Community: Community No.: FIRM Panels: Effective Date: Missoula County, MT C0200D 30063C0075D 08/16/ C0050D 30063C0200E 30063C0075E 30063C0050E 3. Flood Insurance Study Flathead County, Montana and Incorporated Areas. (Preliminary). (2014). Federal Emergency Management Agency. 07/06/ Flood Insurance Study Lake County, Montana and Incorporated Areas. (2013). Federal Emergency Management Agency. 5. Flood Insurance Study Missoula County, Montana and Incorporated Areas. (1988). Federal Emergency Management Agency. 6. Hirsch, R. (1982). A Comparison of Four Streamflow Record Extension Techniques. Water Resources Research, 18, Interagency Advisory Committee on Water Data. (1982). Guidelines for Determining Flood Flow Frequency: Bulletin #17B of the Hydrology Committee (Rev. Sept ; ed.). Reston, Va.: U.S. Dept. of the Interior, Geological Survey, Office of Water Data Coordination. 8. Montana DNRC. (2012). Swan River at Bigfork Hydrology Design Report. 9. Montana DNRC, & Missoula County. (2014). Memorandum of Understanding. 10. PBS&J. (2010). Swan Lake Hydrologic Analysis Technical Support Data Notebook Parrett, C., & Johnson, D. (2004). Methods for estimating flood frequency in Montana based on data through water year Helena, Mont.: U.S. Dept. of the Interior, U.S. Geological Survey. 12. Simons, LI and Associates, Inc., Lower Swan River Flood Study, (1986, March). 13. Simons, LI and Associates, Inc., Upper Swan River Flood Study, (1986, March). 17

21 Hydrology Design Report, Swan River Detailed Floodplain Study May USGS Peak Streamflow for Montana. (n.d.). Retrieved October 22, 2014, from _form=introduction 15. Watershed Sciences Inc., 2013, 2-foot contours derived from ground-classified LiDAR point data (collected Oct. 2012), Missoula County, Montana. 16. (n.d.). Retrieved April 24, 2015, from 18

22 Appendix A USGS Stream Gage Data USGS Gage (#105) Swan River near Bigfork, MT o Systematic Peak Streamflow Table o Systematic Peak Streamflow Graph o LP3 Analysis through WY 1998 USGS Gage (#102) Swan River near Condon, MT o Systematic Peak Streamflow Table o Systematic Peak Streamflow Graph o LP3 Analysis through WY 1998

23 USGS Surface Water for Montana: Peak Streamflow of /26/2014 2:02 PM USGS Home Contact USGS Search USGS National Water Information System: Web Interface USGS Water Resources (Cooperator Access) Data Category: Geographic Area: Click to hidenews Bulletins Try our new Mobile-friendly water data site from your mobile device! Full News Peak Streamflow for Montana Available data for this site Lake County, Montana Hydrologic Unit Code Latitude 48 01'27.23", Longitude '43.75" NAD83 Drainage area 671 square miles Gage datum 3,062.6 feet above NGVD29 Output formats Table Graph Tab-separated file peakfq (watstore) format Reselect output format Water Year Date Gage Height (feet) Streamflow (cfs) 1922 Jun. 08, , Jun. 14, , May 19, , May 23, , Apr. 21, ,860

24 USGS Surface Water for Montana: Peak Streamflow of /26/2014 2:02 PM Water Year Date Gage Height (feet) Streamflow (cfs) 1927 Jun. 13, , May 28, , May 26, , Jun. 01, , May 18, , May 24, , Jun. 18, , Dec. 25, , May 25, , May 17, , May 29, , May 30, , May 06, , May 27, , Jun. 03, , May 28, , Jun. 21, , May 20, , Jun. 04, , May 30, , May 11, , May 24, , May 17, , Jun. 23, , May 14, , Apr. 30, , Jun. 05, , May 22, , Jun. 16, , Jun. 04, , May 08, ,960

25 USGS Surface Water for Montana: Peak Streamflow of /26/2014 2:02 PM Water Year Date Gage Height (feet) Streamflow (cfs) 1958 May 27, , Jun. 22, , Jun. 06, , May 29, , May 31, , Jun. 02, , Jun. 10, , Jun. 20, , Jun. 02, , May 25, , Jun. 25, , Jun. 01, , Jun. 08, , May 31, , Jun. 12, , May 21, , Jun. 20, , Jun. 17, , May 16, , May 04, , Jun. 10, , May 28, , May 27, , May 27, , Jun. 19, , Jun. 01, , Jun. 26, , Jun. 10, , Jun. 01, , May 03, , Jun. 09, ,170

26 USGS Surface Water for Montana: Peak Streamflow of /26/2014 2:02 PM Water Year Date Gage Height (feet) Streamflow (cfs) 1989 May 12, , Jun. 28, , May 21, , May 10, , May 18, , May 14, , Jun. 07, , Jun. 11, , May 18, , May 29, , May 29, , Jun. 17, , May 30, , Jun. 03, , Jun. 02, , Jun. 08, , Jun. 05, , Jun. 18, , Jun. 07, , May 22, , Jun. 02, , Jun. 06, , Jul. 02, , Apr. 29, , May 15, ,210 Questions about sites/data? Feedback on this web site Automated retrievals Help Data Tips Explanation of terms

27 USGS Surface Water for Montana: Peak Streamflow of /26/2014 2:02 PM Subscribe for system changes News Accessibility Plug-Ins FOIA Privacy Policies and Notices U.S. Department of the Interior U.S. Geological Survey Title: Surface Water for Montana: Peak Streamflow URL: Page Contact Information: Montana Water Data Maintainer Page Last Modified: :01:01 EST nadww02

28 USGS Surface Water for Montana: Peak Streamflow of /26/2014 2:05 PM USGS Home Contact USGS Search USGS National Water Information System: Web Interface USGS Water Resources (Cooperator Access) Data Category: Geographic Area: Click to hidenews Bulletins Try our new Mobile-friendly water data site from your mobile device! Full News Peak Streamflow for Montana Available data for this site Lake County, Montana Hydrologic Unit Code Latitude 48 01'27.23", Longitude '43.75" NAD83 Drainage area 671 square miles Gage datum 3,062.6 feet above NGVD29 Output formats Table Graph Tab-separated file peakfq (watstore) format Reselect output format

29 USGS Surface Water for Montana: Peak Streamflow of /26/2014 2:05 PM Download a presentation-quality graph Questions about sites/data? Feedback on this web site Automated retrievals Help Data Tips Explanation of terms Subscribe for system changes News Accessibility Plug-Ins FOIA Privacy Policies and Notices U.S. Department of the Interior U.S. Geological Survey Title: Surface Water for Montana: Peak Streamflow URL: Page Contact Information: Montana Water Data Maintainer Page Last Modified: :03:22 EST nadww02

30 Montana Flood-Frequency and Basin-Characteristic Data 1 of 3 11/26/2014 2:41 PM Flood-frequency data are based on recorded annual peak discharges through Peak discharges for specified frequencies (exceedance probabilities) were determined by fitting a log-pearson Type 3 probability distribution to base 10 logarithms of recorded annual peak discharges as described by the Interagency Advisory Committee on Water Data (1982, Guidelines for Determining Flood Flow Frequency--Bulletin 17-B of the Hydrology Subcommittee: U.S. Geological Survey, Office of Water Data Coordination). Note: Data are provisional and user is responsible for assessment and interpretation of flood-frequency data. Most of the basin characteristic data were measured in the 1970s from the best-scale topographic maps available at the time. Some data, such as mean annual precipitation, soil index data, and mean January minimum temperatures, were compiled from maps prepared by other agencies. Channel widths were measured in the field by USGS personnel. The flood-frequency and basin characteristics data were used in a new flood-frequency report just published by the USGS, entitled "Methods for estimating Flood Frequency in Montana Based on Data through Water Year 1998" (Water-Resources Investigations Report ). Information about the equations described in that report can be found at the following link. For more detailed information contact Wayne Berkas: Phone: or by . Annual peak discharge, in cubic feet per second (top line), for indicated exceedance probability, in percent (bottom line):

31 Montana Flood-Frequency and Basin-Characteristic Data 2 of 3 11/26/2014 2:41 PM NOTE: Systematic peaks are those that are recorded within the period of gaged record. The computed systematic flood-frequency curve is based only on the systematic peaks. The computed Bulletin 17-B flood-frequency curve often is different from the systematic flood-frequency curve because of differences between station skew and regional skew, low- or high-outlier adjustments, or the presence of one or more historical peaks outside the systematic record. Historical peaks also result in historical adjusted plotting positions (exceedance probabilities) for all peaks. Basin Characteristics: Value Abbrev Explanation 36.3 SLOPE Main channel slope, in ft per mile 84.5 LENGTH Total stream length, miles ELEV Mean basin elevation, ft above msl 26.0 EL6000 Percent of basin above 6,000 ft, msl 1.85 STORAGE Percent of basin in lakes, ponds, and swamps 89.8 FOREST Percent of basin in forest 8.9 SOIL_INF Soil index, in inches LAT_GAGE Latitude of gage, in decimal degrees LNG_GAGE Longitude of gage, in decimal degrees

32 Montana Flood-Frequency and Basin-Characteristic Data 3 of 3 11/26/2014 2:41 PM 23.0 PRECIP Mean annual precipitation, in inches 1.7 I24_2 Precipitation intensity for a 24-hour storm having a 2-year recurrence interval, in inches per hour 14.0 JANMIN Mean minimum January temperature, in degrees F WAC Width of active channel, in feet 1.0 W2 Mean depth for active channel, in feet WBF Width of bankfull channel, in feet 4.0 W4 Mean depth of bankfull channel, in feet Montana Flood-Frequency and Basin-Characteristic Data Retrieved on: :41:04 Department of the Interior, U.S. Geological Survey Privacy Statement Disclaimer Accessibility FOIA

33 USGS Surface Water for Montana: Peak Streamflow of /26/2014 3:13 PM USGS Home Contact USGS Search USGS National Water Information System: Web Interface USGS Water Resources (Cooperator Access) Data Category: Geographic Area: Click to hidenews Bulletins Try our new Mobile-friendly water data site from your mobile device! Full News Peak Streamflow for Montana Available data for this site Missoula County, Montana Hydrologic Unit Code Latitude 47 25'20.58", Longitude '15.50" NAD83 Drainage area 69.1 square miles Gage datum 4,015 feet above NGVD29 Output formats Table Graph Tab-separated file peakfq (watstore) format Reselect output format Water Year Date Gage Height (feet) Streamflow (cfs) 1973 May 20, Jun. 18, , Jun. 16, , May 15, May 03,

34 USGS Surface Water for Montana: Peak Streamflow of /26/2014 3:13 PM Water Year Date Gage Height (feet) Streamflow (cfs) 1978 Jun. 09, May 27, May 26, , Jun. 07, Jun. 17, , May 30, Jun. 22, Jun. 08, May 31, , May 01, Jun. 08, May 11, Jun. 26, May 19, May 09, Questions about sites/data? Feedback on this web site Automated retrievals Help Data Tips Explanation of terms Subscribe for system changes News Accessibility Plug-Ins FOIA Privacy Policies and Notices U.S. Department of the Interior U.S. Geological Survey Title: Surface Water for Montana: Peak Streamflow URL: Page Contact Information: Montana Water Data Maintainer Page Last Modified: :12:58 EST nadww02

35 USGS Surface Water for Montana: Peak Streamflow of /26/2014 3:13 PM USGS Home Contact USGS Search USGS National Water Information System: Web Interface USGS Water Resources (Cooperator Access) Data Category: Geographic Area: Click to hidenews Bulletins Try our new Mobile-friendly water data site from your mobile device! Full News Peak Streamflow for Montana Available data for this site Missoula County, Montana Hydrologic Unit Code Latitude 47 25'20.58", Longitude '15.50" NAD83 Drainage area 69.1 square miles Gage datum 4,015 feet above NGVD29 Output formats Table Graph Tab-separated file peakfq (watstore) format Reselect output format

36 USGS Surface Water for Montana: Peak Streamflow of /26/2014 3:13 PM Download a presentation-quality graph Questions about sites/data? Feedback on this web site Automated retrievals Help Data Tips Explanation of terms Subscribe for system changes News Accessibility Plug-Ins FOIA Privacy Policies and Notices U.S. Department of the Interior U.S. Geological Survey Title: Surface Water for Montana: Peak Streamflow URL: Page Contact Information: Montana Water Data Maintainer Page Last Modified: :13:36 EST nadww02

37 Montana Flood-Frequency and Basin-Characteristic Data 1 of 3 11/26/2014 2:43 PM Flood-frequency data are based on recorded annual peak discharges through Peak discharges for specified frequencies (exceedance probabilities) were determined by fitting a log-pearson Type 3 probability distribution to base 10 logarithms of recorded annual peak discharges as described by the Interagency Advisory Committee on Water Data (1982, Guidelines for Determining Flood Flow Frequency--Bulletin 17-B of the Hydrology Subcommittee: U.S. Geological Survey, Office of Water Data Coordination). Note: Data are provisional and user is responsible for assessment and interpretation of flood-frequency data. Most of the basin characteristic data were measured in the 1970s from the best-scale topographic maps available at the time. Some data, such as mean annual precipitation, soil index data, and mean January minimum temperatures, were compiled from maps prepared by other agencies. Channel widths were measured in the field by USGS personnel. The flood-frequency and basin characteristics data were used in a new flood-frequency report just published by the USGS, entitled "Methods for estimating Flood Frequency in Montana Based on Data through Water Year 1998" (Water-Resources Investigations Report ). Information about the equations described in that report can be found at the following link. For more detailed information contact Wayne Berkas: Phone: or by . Annual peak discharge, in cubic feet per second (top line), for indicated exceedance probability, in percent (bottom line):