Appendix E Hydrology, Erosion and Sediment Transport Studies

Transcription

1 Appendix E Hydrology, Erosion and Sediment Transport Studies Hatch 2012/10

2 Appendix E1 EA Hydrology Memorandum February 2011 (Hatch. 2011a) Hatch 2012/10

3 Project Memo February 23, 2011 TO: Larry King FROM: Mark Orton Jim Law cc: File H Horizon Hydro Inc. Trout Lake River Hydro Project EA Hydrology Memorandum 1. Introduction This memorandum presents the information requested by the Ministry of the Environment (MOE), under Hydrological Analysis, items a) to g) of Table 1, for the proposed Trout Lake River Hydro Project on the Trout Lake River at Big Falls. Big Falls is located in Kenora District of the Northwest Region of Ontario. The Trout Lake River flows south out of Trout Lake and drains 2370 km 2 at Big Falls. Figure 1 shows the project location. A full listing of the hydrology requirements and the location in this document where they can be found area given in Appendix A. The software used to extract the information presented herein is listed in Appendix B. Table 1 lists the information required by the MOE under Hydrological Analysis (as per project correspondence). These analyses of existing data have been undertaken in support of the ongoing Environmental Assessment process and should not be used in the detailed engineering design without further review. Seasons are defined as: Winter January 1 to March 31 Julian Day 1 to 90 Spring April 1 to June 30 Julian Day 91 to 181 Summer July 1 to September 30 Julian Day 182 to 273 Fall October 1 to December 31 Julian Day 274 to 365 If you disagree with any information contained herein, please advise immediately. H , Rev. 0, Page 1

4 Table 1 Information Required under Hydrological Analysis a) Descriptive flow statistics using all available daily flows for all years: mean, median, minimum, maximum, flow exceeded 20% time, flow exceeded 80% time. b) Extreme low flow statistics: 7Q2 (2 year return period 7-day-average-low flow), 7Q10 (10 year return period 7-day-average-low flow) and 7Q20 (20 year return period 7-dayaverage-low flow). c) Flow duration curves and tables using total daily average flow data for the entire period, for all four seasons and for all twelve months. d) Flow duration curves and tables using daily baseflow data for the entire period, for all four seasons and for all twelve months. e) Flow duration curves derived using both the percentile method and the median of percentiles method. Both methods are incorporated into the flow analysis tool, developed by Schmidt and Metcalfe (2009), which can be downloaded for free from f) Flood frequency analysis using instantaneous maximum flow of each year for the entire period of records. g) Low flow frequency analysis using 7-day-average-low flow for the entire period of records. h) Altered flow of the bypass reach and the reach below tailrace, if applicable. i) Compensation flow for the bypass reach and the reach below tailrace, if applicable. 2. Descriptive Flow Statistics Descriptive flow statistics for the Trout Lake River at Big Falls have been prepared using the Streamflow Analysis and Assessment Software (SAAS) from the Ontario Ministry of Natural Resources (MNR). The long term daily flow series for was extracted from the Water Survey of Canada archived flow data website for streamflow station 05QC003, Trout Lake River above Big Falls. The station information for 05QC003 indicates that significant flow estimation was required in 2009 due to gauge problems, so the flow record was screened for stationarity using Environment Canada s Consolidated Frequency Analysis software, CFA 3.1. This screening indicated possible nonstationarity at the 5% significance level in the homogeneity test, but not at the 1% significance level. The tests for serial dependence, randomness and trend did not reveal any problems is the highest year on record at Big Falls in terms of flood peak and flow volume so a further check on the flow series was undertaken, comparing Trout Lake River flows to flows in the Chukuni River near Ear Falls [03QC001] and the Cedar River below Wabaskang Lake [05QE008]. Double mass analyses between the flow records for revealed no change of slope or other anomalies. Furthermore 1974 and 2009 were the highest flow years and 1988 was the lowest flow year in both the Trout Lake River and the Chukuni River. As a result of this preliminary screening analysis it was concluded that the full 40-year flow record for the Trout Lake River above Big Falls [05QC003] can be used in the Hydrological Analysis. Figure 2 shows the superimposed daily flow hydrographs from these 40 years as well as for the median of each day of the year. Table 2 shows the descriptive flow statistics using all available daily flows for all years. H , Rev. 0, Page 2

5 Table 2 Descriptive Mean Daily Flow Statistics for Big Falls Parameter Value Units Date of Occurrence Mean Flow 17.5 m 3 /s - Median Flow 14.3 m 3 /s - Minimum Flow 1.92 m 3 /s 12/2/1987 Maximum Flow 103 m 3 /s 8/5/2009 Flow Exceeded 20% time 24.2 m 3 /s - Flow Exceeded 80% time 9.65 m 3 /s - Number of Zero Flow Days Extreme Low Flow Statistics Low flow frequency analysis of 7-day average low flows has been undertaken using the Engsoft Low Flow Frequency Analysis software ESLOFFAN. This software reads the full daily flow series for Big Falls and creates a 7-day (or 30-day) average flow series. The annual maxima of this average flow series are then extracted for low flow frequency analysis using the 2-parameter Lognormal, Log Pearson III and Gumbel distributions. Table 3 shows the annual 7-day average flow minima for the Trout Lake River at Big Falls. Table 4 show 7QT-year extreme low flow statistics from the frequency analyses. The three frequency distributions shown in Table 4 give similar annual 7-day average flow minima for the 7Q2, 7Q10 and 7Q20 extreme low flow statistics. The Log Pearson III distribution appears to give the best fit to the data in Table 3. The extreme low flow estimates from the Log Pearson III distribution are: 7Q m 3 /s 7Q m 3 /s 7Q m 3 /s H , Rev. 0, Page 3

6 Table 3 Annual 7-day Average Flow Minima Trout Lake River at Big Falls Year Month 7-day Average Flow (m 3 /s) H , Rev. 0, Page 4

7 Table 4 7-day Average Extreme Low Flow Statistics Trout Lake River at Big Falls Return Period Minimum 7-day Average Low Flow (m 3 /s) (years) 2 Parameter LogNormal Log Pearson Type III Gumbel Total Daily Average Flow Duration Curves SAAS v2.1.1 has been used to generate flow duration curves using total daily flow for the entire period of record, for all four seasons and for all twelve months. These flow duration curves are shown in Figures 3 to 5, respectively. The flow duration analysis results used prepare Figures 3 to 5 are given in Tables 5 and 6. Table 5 Total Daily Average Flow Duration Analysis for All Four Seasons and the Period of Record Percent of Total Daily Average Flow (m 3 /s) Time Period of Exceeded Winter Spring Summer Fall Record H , Rev. 0, Page 5

8 Percent of Total Daily Average Flow (m 3 /s) Time Period of Exceeded Winter Spring Summer Fall Record H , Rev. 0, Page 6

9 Percent of Total Daily Average Flow (m 3 /s) Time Period of Exceeded Winter Spring Summer Fall Record Table 6 Total Daily Average Flow Duration Analysis for All Twelve Months of the Year Percent of Time Exceeded Total Daily Average Flow (m 3 /s) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec H , Rev. 0, Page 7

10 Percent Total Daily Average Flow (m 3 /s) of Time Exceeded Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec H , Rev. 0, Page 8

11 Percent Total Daily Average Flow (m 3 /s) of Time Exceeded Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Daily Average Baseflow Flow Duration Curves SAAS v2.1.1 separates baseflow using a recursive digital filtering procedure. The resulting daily average baseflow series is tabulated by SAAS and has been used to generate baseflow flow duration curves for the entire period of record, for all four seasons and for all twelve months. These baseflow flow duration curves are shown in Figures 6 to 8, respectively. The flow duration analysis results used prepare Figures 6 to 8 are given in Tables 7 and 8. H , Rev. 0, Page 9

12 Table 7 Daily Average Baseflow Flow Duration Analysis for All Four Seasons and the Period of Record Percent of Daily Average Baseflow (m 3 /s) Time Winter Spring Summer Fall Period of Exceeded Record H , Rev. 0, Page 10

13 Percent of Daily Average Baseflow (m 3 /s) Time Winter Spring Summer Fall Period of Exceeded Record H , Rev. 0, Page 11

14 Percent of Daily Average Baseflow (m 3 /s) Time Winter Spring Summer Fall Period of Exceeded Record Table 8 Daily Average Baseflow Flow Duration Analysis for All Twelve Months of the Year Percent of Time Exceeded Daily Average Baseflow (m 3 /s) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec H , Rev. 0, Page 12

15 Percent Daily Average Baseflow (m 3 /s) of Time Exceeded Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec H , Rev. 0, Page 13

16 Percent Daily Average Baseflow (m 3 /s) of Time Exceeded Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Median Annual Flow Duration Curves SAAS v2.1.1 includes the option to generate median annual flow duration curves for the entire period of record, for all four seasons and for all twelve months. The median annual method calculates a flow duration curve (FDC) for each year (or month/season) of the period of record, records the value for each percent exceedance (0 to 100) on each year s FDC, calculates the median for each, and draws a FDC using the median values. The median annual flow duration curves are shown in Figures 9 to 11, respectively. The flow duration analysis results used prepare Figures 9 to 11 are given in Tables 9 and 10. Table 9 Median Annual Flow Duration Analysis for All Four Seasons and the Period of Record Percent of Median Annual Flow (m 3 /s) Time Winter Spring Summer Fall Period of Exceeded Record H , Rev. 0, Page 14

17 Percent of Median Annual Flow (m 3 /s) Time Winter Spring Summer Fall Period of Exceeded Record H , Rev. 0, Page 15

18 Percent of Median Annual Flow (m 3 /s) Time Winter Spring Summer Fall Period of Exceeded Record Table 10 Median Annual Flow Duration Analysis for All Twelve Months of the Year Percent of Time Exceeded Median Annual Flow (m 3 /s) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec H , Rev. 0, Page 16

19 Percent Median Annual Flow (m 3 /s) of Time Exceeded Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec H , Rev. 0, Page 17

20 Percent Median Annual Flow (m 3 /s) of Time Exceeded Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Flood Frequency Analysis Instantaneous maximum flow data are available for the Trout Lake River above Big Falls for the full period of record, from 1970 to The instantaneous flood maxima at Big Falls are shown in Table 11. H , Rev. 0, Page 18

21 Table 11 Estimated Instantaneous Flood Maxima for the Trout Lake River at Big Falls Year Peak (m 3 /s) Year Peak (m 3 /s) This 40-year instantaneous flood maxima series was entered into the SAAS model and also into Environment Canada s Consolidated Frequency Analysis software CFA3.1. In CFA3.1 the data set was tested for independence, trend, homogeneity and randomness and was found to be free of any statistical inconsistencies except at the 5% significance level for homogeneity (see Section 2). The data set was then subjected to parametric flood frequency analysis using the following frequency distributions: General Extreme Value (GEV) 3 parameter Lognormal Log Pearson Type III Wakeby. Table 12 shows the flood frequency analysis results at Big Falls. Figure 12 shows the GEV distribution, which was found to give the best fit to the data. H , Rev. 0, Page 19

22 Table 12 Return Period Years Flood Frequency Analysis for the Trout Lake River at Big Falls Instantaneous Flood Peak in m 3 /s GEV 3 para Lognormal Log Pearson III Wakeby (1) (1) (1) Note: (1) Flood estimates with a return period of greater than 100 year should be used with caution. The Flood flows are intended for use in the EA and shall not be used for engineering design without further review. H , Rev. 0, Page 20

23 N Project Site Source: Red Lake Touring Region Adventure Guide H/327203_rm Figure 1.1 Horizon Hydro Inc. Trout Lake Hydroelectric Project Project Location

24 Figure 2 (SAAS) Superimposed Daily Hydrographs at Big Falls H , Rev. 0, Page 22

25 Figure 3 (SAAS) Period of Record Total Daily Average Flow Duration Curve H , Rev. 0, Page 23

26 Figure 4 (SAAS) Period of Record Seasonal Composite Flow Duration Curves H , Rev. 0, Page 24

27 Figure 5 (SAAS) Period of Record Monthly Composite Flow Duration Curves H , Rev. 0, Page 25

28 100 Period-of-Record Baseflow Flow Duration Curve Discharge [m 3 sec -1 ] Exceedence Probability [%] Figure 6 (SAAS) Period of Record Baseflow Flow Duration Curve H , Rev. 0, Page 26

29 100 POR Seasonal Composite Baseflow Flow Duration Curve Winter Spring Summer Fall Discharge [m 3 sec -1 ] Exceedence Probability [%] Figure 7 (SAAS) Period of Record Seasonal Composite Baseflow Flow Duration Curves H , Rev. 0, Page 27

30 Discharge [m 3 sec -1 ] POR Monthly Composite Baseflow Flow Duration Curve Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Exceedence Probability [%] Figure 8 (SAAS) Period of Record Monthly Composite Baseflow Flow Duration Curves H , Rev. 0, Page 28

31 Figure 9 (SAAS) Median Annual Flow Duration Curve H , Rev. 0, Page 29

32 Figure 10 (SAAS) Median Seasonal Composite Flow Duration Curves H , Rev. 0, Page 30

33 Figure 11 (SAAS) Median Monthly Composite Flow Duration Curves H , Rev. 0, Page 31

34 Trout Lake River above Big Falls - Instantaneous Maxima GEV Distribution 1000 Return Periods Observation (m 3 /s) Standard Normal Variate Annual Maximum Discharge Best Fit Figure 12 (CFA3.1) Instantaneous Flood Frequency Analyses for Big Falls H , Rev. 0, Page 32

35 Appendix A MOE Hydrology Requirements H , Rev. 0, Page 33

36 What information is required under hydrological analysis? a) Descriptive flow statistics using all available daily flows for all years: mean, median, minimum, maximum, flow exceeded 20% time, flow exceeded 80% time. b) Extreme low flow statistics: 7Q 2 (2 year return period 7-dayaverage-low flow), 7Q 10 (10 year return period 7-dayaverage-low flow) and 7Q 20 (20 year return period 7-dayaverage-low flow). c) Flow duration curves and tables using total daily average flow data for the entire period, for all four seasons and for all twelve months. d) Flow duration curves and tables using daily baseflow data for the entire period, for all four seasons and for all twelve months. e) Flow duration curves derived using both the percentile method and the median of percentiles method. Both methods are incorporated into the flow analysis tool, developed by Schmidt and Metcalfe (2009), which can be downloaded for free from f) Flood frequency analysis using instantaneous maximum flow of each year for the entire period of records. g) Low flow frequency analysis using 7-day-average-low flow for the entire period of records. h) Altered flow of the bypass reach and the reach below tailrace, if applicable. i) Compensation flow for the bypass reach and the reach below tailrace, if applicable. Location Section 2 Section 3 Section 4 Section 5 Section 6 Section 7 Section 3 n/a n/a H , Rev. 0, Page 34

37 Appendix B MOE Hydrology Requirement Calculation Procedures H , Rev. 0, Page 35

38 Hydro Site Hydrology in Ontario Ontario Ministry of the Environment Requirements Input Data Source Software Comments a) Descriptive flow statistics using all available daily flows for all Long term daily flow series SAASv2.1.1 The Hydrograph Explorer (Basic Table) and Flow years: mean, median, minimum, maximum, flow exceeded 20% synthesized for hydro site. Analysis (Flow Duration Curves) applications. time, flow exceeded 80% time. Must be > 20 years. b) Extreme low flow statistics: 7Q 2 (2 year return period 7-dayaverage-low flow), 7Q 10 (10 year return period 7-day-average-low synthesized for hydro site. ESLOFFAN Long term daily flow series ENGSOFT/ flow) and 7Q 20 (20 year return period 7-day-average-low flow). Must be > 20 years. c) Flow duration curves and tables using total daily average flow Long term daily flow series SAASv2.1.1 data for the entire period, for all four seasons and for all twelve synthesized for hydro site. months. Must be > 20 years. d) Flow duration curves and tables using daily baseflow data for the Long term daily flow series entire period, for all four seasons and for all twelve months. synthesized for hydro site. Must be > 20 years. e) Flow duration curves derived using both the percentile method Long term daily flow series SAASv2.1.1 and the median of percentiles method. Both methods are synthesized for hydro site. incorporated into the flow analysis tool, developed by Schmidt and Must be > 20 years. Metcalfe (2009), which can be downloaded for free from f) Flood frequency analysis using instantaneous maximum flow of Long term instantaneous flood CFA3.1 each year for the entire period of records. series synthesized for hydro site. Must be > 20 years. g) Low flow frequency analysis using 7-day-average-low flow for the Long term daily flow series ENGSOFT/ entire period of records. synthesized for hydro site. ESLOFFAN Must be > 20 years. h) Altered flow of the bypass reach and the reach below tailrace, if Design flow series. applicable. i) Compensation flow for the bypass reach and the reach below Design flow series. tailrace, if applicable. 7-day average low flow frequency analysis application; 2LN, 3LN, Log Pearson and Gumbel distributions. Flow Analysis (Flow Duration Curves); period of record, seasonal and monthly applications. SAASv2.1.1 Flow Analysis (Baseflow analysis table); 1, 5, 25, 50, 75, 95 and 99% baseflow exceedance values for: period of record, seasonal and monthly applications. Flow Analysis (Flow Duration Curves); percentile and median of percentiles methods for; period of record, seasonal and monthly applications. See Flood Frequency Analysis sheet 7-day average low flow frequency analysis application; 2LN, 3LN, Log Pearson and Gumbel distributions. To be defined in design phase To be defined in design phase H , Rev. 0, Page 36

39 Hydro Site Hydrology in Ontario Flood Frequency Analysis Step 1 Pre-process instantaneous annual flood peak series from synthesized daily flow series at the hydro site: 1. Select instantaneous and daily annual flood peaks from regional streamflow stations 2. Compare peaking factors (instantaneous/daily peaks) with: - Drainage area - Season - Regulation (natural and man-made) - Flood magnitude (divided by drainage area) - previous and following day flows 3. Derive fuction to estimate instantaneous flood peaks from daily flood peaks 4. Use function to estimate instantaneous annual flood peak series from synthesized daily flow series at hydro site Step 2 Input the synthesized instantaneous flood peak series to the CFA3.1 software Step 3 Screen synthesized instantaneous flood peak series for statistical consistency, including: - Independence - Trend - Homogeneity - General Randomness Step 4 Test for high and low outliers and adjust accordingly Step 5 Perform parametric frequency analysis using: - The General Extreme Value (GEV) Distribution - The Three-Parameter Lognormal (3LN) Distribution - The Log Pearson Type III Distribution - The Wakeby Distribution Step 6 Select best fit frequency distribution Step 7 Plot and tabulate instantaneous flood peak estimates for hydro site H , Rev. 0, Page 37

40 Hydro Site Hydrology in Ontario Hydrological Software used in Analysis Software Title Source Year Description SAASv2.1.1 Streamflow Analysis Ontario Ministry of Natural and Assessment Resources Software 2010 ESLOFFAN EngSoft LOw Flow ENGSOFT 1995 Frequency ANalysis CFA 3.1 Consolidated Frequency Analysis Environment Canada 2000 SAAS is a MATLAB application to analyze an annual maxima flow series, daily streamflow time-series and hourly streamflow time-series. The software produces tabular and graphic reports on: - flow hydrographs - flood frequency analysis - flow duration curves - baseflow - rate of change ESLOFFAN extracts the annual minimum 7-day or 30-day average flows from a HYDAT formatted daily flow series and undertakes low flow frequency analysis using the 2LN, 3LN, Log Pearson III and Gumbel distributions. The software produces tabular and graphic output for each distribution. CFA 3.1imports an annual maxima flow series and stores it in a direct access file. The software performs the following tests on the data series: - Spearman rank order serial correlation coefficient test for independence - Spearman rank order correlation coefficient test for trend - Mann-Whitney split sample test for homogeneity - Runs above and below the median for general randomness Flood estimates are computed using the following distributions: - The General Extreme Value (GEV) Distribution - The Three-Parameter Lognormal (3LN) Distribution - The Log Pearson Type III Distribution - The Wakeby Distribution The software produces tabular and graphic output for each distribution. MO:GJL:ll H , Rev. 0, Page 38