Flood and Floodplain
Total Page:16
File Type:pdf, Size:1020Kb
15-10-29 Flood analysis and flood projections under climate change in New Brunswick (2010-2100) N. El-Jabi & N. Turkkan D. Caissie Université de Moncton Dept. of Ficheries & Oceans Moncton, NB, Canada Moncton, NB, Canada Guidelines for Your Environmental Trust Fund at Work October 26, 2015 Flood and Floodplain 1 1 15-10-29 Flood and Floodplain RETURN PERIOD HYDROMETRIC STATION 2 Decision space for Floodplain Flood Damage Flood Damage Year 3 2 15-10-29 Decision space for Floodplain Flood Damage Flood Damage Year 3 Outline The purpose of this presentation is to: § Estimate floods, based on historical data for: Ø NB Hydrometric stations, and followed by a Ø Regional regression analysis; § Identify and quantify potential impacts on floods flows based on projected climate 2100; § Develop new design criteria (based on historical data and future climate) for floods flows in New Brunswick. 64 3 15-10-29 Hydrometric stations analysis 1BF1 1BD2 1BJ7 1BJ4 1BJ3 1BE1 1BC1 1BL2 1BJ1 1BL1 1AD4 1BK4 1BL3 1AF3 1AD2 1AH5 1AD3 1BQ1 1AF2 1BP1 N 1BO3 N 1AG2 1BO2 1AG3 1BR1 W E 1BO1 S 1AJ1 1AJ4 1AJ11 1BS1 1AJ10 1AL3 1AN2 1AN1 1AL4 1AJ3 1AL2 1AP2 1AK7 1AK5 1AK2 1BU3 1AK4 1BU2 1AK8 1AK1 1BU4 1AP4 1DL1 1BV7 1AM1 1DL1 1AP6 1BV6 1BV5 1AQ2 1AR6 1AR8 1AQ1 Figure 581. Location stations; of selected hydrometric Data: stations HYDAT in New Brunswick (58 stations). 5 0 150 300 kilometers Flood Flow Statistical Analysis Generalized Extreme Value (GEV) Distribution Example: Little Southwest Miramichi River at Lyttelton 10 6 4 15-10-29 Flood Frequency by station Table 4. Results of single station flood frequency analyses using the Generalized Extreme Value (GEV) distribution Daily discharge Station QD2 QD10 QD20 QD50 QD100 (m³/s) (m³/s) (m³/s) (m³/s) (m³/s) Saint John River (Fort Kent) 2313 3254 3542 3867 4080 St Francis River 196 315 361 422 467 Saint John River (Grand Falls) 3196 4666 5115 5623 5954 Green River 219 349 394 448 486 Limestone River 33.5 48.9 55.4 64.4 71.6 Aroostook River 945 1328 1447 1582 1671 Mamozekel River 40.0 65.2 75.4 89.3 100 Saint John River (East Florenceville) 4745 7269 7982 8747 9223 Meduxnekeag River 236 378 433 505 559 Big Presque Isle Stream 92.1 156 187 232 272 …… …... …… …… …… …… 7 Regional Flood Frequency b1 bb12 QaDAT = () or QT = a()() DA MAP • DA is drainage area; • MAP is the mean annual precipitation; • a, b1 and b2 are regression constants and • QT denote the T-year flood. Table 7. Regional regression coefficient estimates and R² (GEV distribution) a b1 b2 R²* MAFL** 0.463476 0.884 * 0.984 4.2645E-06 0.926 1.617 0.990 QD2 (m3/s) 0.394690 0.897 * 0.985 1.1131E-05 0.935 1.460 0.990 QD10 (m3/s) 0.753188 0.871 * 0.981 1.3152E-06 0.919 1.848 0.988 QD20 (m3/s) 0.950031 0.857 * 0.977 5.5022E-07 0.910 2.002 0.987 QD50 (m3/s) 1.273837 0.839 * 0.971 1.7180E-07 0.896 2.205 0.983 QD100 (m3/s) 1.580312 0.824 * 0.964 7.0216E-08 0.886 2.360 0.978 * The R² was obtained from the log-transformed regression equations. ** Represents the Mean Annual Flood (MAFL), to be used in conjuction with the index-flood method. Range of application of regression equations: Drainage area = 3.89 km² to 39900 km² Mean Annual Precipitation = 925 mm to 1410 mm 8 5 15-10-29 Instantaneous Flood Flows 4.0 3.5 (QP/QD) 3.0 2.5 2.0 1.5 1.0 0.5 0.0 200 km2 800 km2 Ratio of instantaneous peak flow to daily flow daily to flow peak of instantaneous Ratio 1 10 100 1000 10000 100000 9 Drainage area (km2) Floods under climate change scenarios Location of used meteorological and METEO. Drainage STATIONS hydrometric stations. HYDROMETRIC STATIONS Area Aroostook (km2) Charlo « Charlo Saint John R. at Fort Kent 14,700 Chatham !1BC1 !1AD2 Doaktown Nashwaak R. at Durham 1,450 Bridge Fredericton !1BQ1 Canaan R. at East Canaan 668 « Aroostook Moncton «Chatham 1BO1! Saint John Kennebecasis R. at 1,100 « Doaktown Apohaqui !1AL2 !1AP2 Restigouche R. below 3,160 « Fredericton « Moncton Kedgwick SW Miramichi R. at 5,050 !1AP4 Blackville « St John NW Miramichi R. at Trout 948 ! Hydrometric station Bk. « Metrological station 1210 6 15-10-29 Atmosphere-Ocean General Circulation Model (AOGCM) grid GRID SIZE: ~ 200 x 300 km 3 4 « Charlo !1BC1 !1AD2 !1BQ1 « Aroostook «Chatham 1BO1! « Doaktown !1AL2 !1AP2 « Fredericton « Moncton !1AP4 « St John 1 2 A2 ! Hydrometric station « Metrological station B1 § Canadian Coupled Global Climate Model (CGCM3.1); § A2 and B1 defined by IPCC, (Intergovernmental Panel on Climate Change) 1311 Climate data downscaling Discharge modeling for future climate The climate change fields (temperatures and precipitation) were downscaled Future climate data from CGCM3.1 Neural Network Regression models model d) SW Miramichi R. (Doaktown) M 8 Tmin 6 Qave Future Ln(Qmax) 4 y = 1.1206x + 0.2453 è Tmax R2 = 0.8874 Flood flow 2 2 4 6 8 P Ln(Qave) Simulated future mean discharge 12 7 15-10-29 An example of simulated Qave & Qmax South West Miramichi River 600 400 /s) 3 200 Qave (m Qave 0 1985 1986 1987 1988 1989 1990 Year 1600 1200 Observed Simulated /s) 3 800 400 Qmax (m Qmax 0 1985 1986 1987 1988 1989 1990 Year 1513 High flow return periods – 2050s & 2080s SW Miramichi River (Doaktown) Example: b) SW Miramichi R. (Doaktown) - Qmax (2040-69) B1 A2 OBS 2500 2000 1918 m3/s 1854 m3/s 1500 m3/s 1000 500 1 10 100 Return Period (y ear) c) SW Miramichi R. (Doaktow n) - Qmax (2070-99) B1 A2 OBS 2500 2089 m3/s 2000 1877 m3/s 1500 m3/s 1000 500 1 10 100 Return Period (y ear) 14 8 15-10-29 Regional Climate Index (RCI) curves for floods Future floods may be computed using a: Regional Climate Index (RCI) For Flood corresponding to a recurrence period (T), x,, ts sc x ,2010 QTT= Q× RCI F in site (x), for a time slice (ts) and scenario (sc): a) Scenario B1 2010-39 2040-69 2070-99 1.6 1.5 1.4 2040-69 1.3 1.3 RCI 1.24 2070-99 1.21 1.2 1.1 2010-39 1 1 10 100 Return Period (y ear) 15 Regional Climate Index (RCI) curves for floods b) Scenario A2 2010-39 2040-69 2070-99 1.6 1.5 1.42 2040-69 1.4 2070-99 1.31 1.3 RCI 1.22 1.20 1.2 1.1 2010-39 1 1 10 100 Return Period (y ear) 16 9 15-10-29 Conclusions § Climate models used suggest for New Brunswick, an increase in precipitation and temperature in the future; § Downscaled models in connection with hydrological models can be used to predict future flows; § Flow return periods predict a significant evolution under climate change; § Regional Climate Index for floods (RCI) may be used for design purposes. 1917 Aknowledgements This study was funded by the New Brunswick Environmental Trust Fund The authors remain thankful to Mr. Darryl Pupek and his team at the Department of Environment and Local Government for support and helpful comments. 18 10 15-10-29 WWW.UMONCTON.CA/HYDRO 19 11 .