Simulating several flood events using Nays 2D Flood
Download as PPTX, PDF1 like271 views
1. The document summarizes a study report on simulating past flood events in Surabaya City to derive probabilistic flood maps. It outlines flood modeling done for 3 events using hydrological and hydraulic models. 2. Key results presented include flood inundation maps, discharge and velocity outputs from the hydrological model, and evaluation of routing results comparing simulated and observed downstream hydrographs. 3. Next steps discussed are rating curve estimation, evaluating model parameters like roughness coefficients, and developing probabilistic flood maps and estimating damage.
Simulating several flood events using Nays 2D Flood
- 1. Study Report 3 M2 - Putika Ashfar Khoiri Water Engineering Laboratory Department of Civil Engineering October 17th , 2017
- 2. General Purpose: Simulate past flood event in Surabaya City to derive probabilistic flood maps in flood prone area by assessing several uncertainties in hydraulic modelling and flood mapping. Problems to be discussed : 1. How river discharge and other parameters have influence on flood simulations (How the model results change depending on the input data) 2. What is the problems while simulating floods? Objective Outline 1
- 3. Previous Task • Development of flood scenarios and model simulations Perform flood inundation modelling for the study area NAYS 2D Flood for 1 flood event Inflow discharge hourly observed discharge data from February 11st, 2015 to February 13rd, 2015 water surface at downstream free outflow Initial water surface Depth = 0 Calculation time 42 hours Time step 2 mins Calculation Condition Parameters which are included : 1. Determination of grid shape 2. Roughness coefficients of the floodplain 2
- 4. Current Progress Modules Evaluation tasks Hydrological Analysis Flow hydrograph shape Peak discharge estimation water-elevation and distance relationship curve Rating-curve rating curve estimation Flood routing Flow hydrograph shape Flood inundations were simulated using the flow conditions on the past flood events between January to March in 2014, 2015 and 2016 Reported flood events obtained from National Agency for Disaster Management Data Date Inundation Height (cm) Duration 23-01-2014 (event 1) 50-70 2 to 3 days 20-02-2015 (event 2) 10-30 2 days 24-02-2016 (event 3) 20-30 2 days The annual maximum peak discharge during the flood events are observed in 3 weeks around flood peak • Initial condition : 2 weeks before the flood occurs • Simulation period : 100 hours 3
- 5. Calculation setting River length 8.64 km Size of W 3000 meter Grid size 30 m x 30 m number of river cross section data 9 points upstream downstream inflow Name of polygon Description Roughness Coefficient Agricultural area Grass, agricultural site 0.030 Low density area Citizen house 0.040 High density area Office, school and public facility 0.070 – 0.80 River River 0.01 (*)IRiC User Manual 4
- 6. Initial Condition (2 weeks before flooding inundation simulation) based on the observation data Date Average discharge (m3.s-1) Maximum Discharge (m3.s-1) 9-01-2014 to 22-01-2014 (IC 1) 79.54010976 181.22 5-02-2015 to 19-02-2015 (IC 2) 66.41162405 182.22 9-02-2016 tp 23-02-2016 (IC 3) 60.37975238 180.1 -50.00 0.00 50.00 100.00 150.00 200.00 0 50 100 150 200 250 300 350 400 Discharge(m3.s-1) Period (Hours) IC 1 IC 2 IC 3 Calculation setting (initial condition) The discharge on event 1 is relatively low before the flood events 5
- 7. Result (initial condition) 1. The value of river depth are varies from 0.5 m to 4 m -> the depth value need to calibrate with the datum to produce water elevation. 2. In the middle stream area, the water depth become higher, probably due to the effect of topographic changes or narrow changes in river cross-section width. Water depth 6
- 8. Result (initial condition)-velocity 1. The velocity magnitude are varies between 0.784 m/s to 1.57m/s along the center of the stream line. 2. Higher velocity occurs due to the meandering of the river Velocity 7
- 9. Discharge routing method The Muskingum channel routing method is employed to route the flow through the streams 𝑄 𝑢 − 𝑄 𝑑 = ∆𝑥 𝑐3 𝜕 𝜕𝑡 𝑄 𝑑 + 1 2 − 𝐿 ∆𝑥 (𝑄 𝑢 − 𝑄 𝑑) 𝑄 𝑑,𝑗+1 = 𝐶1 𝑄 𝑢,𝑗+1 + 𝐶2 𝑄 𝑢,𝑗+1 + 𝐶3 𝑄 𝑢,𝑗+1 Where the coefficients C1,C2,C3 are expressed as 𝐶1 = −𝐾𝜃 + 0.5∆𝑡 𝐾 1 − 𝜃 + 0.5∆𝑡 𝐶2 = −𝐾𝜃 + 0.5∆𝑡 𝐾 1 − 𝜃 + 0.5∆𝑡 𝐶3 = 𝐾 1 − 𝜃 − 0.5∆𝑡 𝐾 1 − 𝜃 + 0.5∆𝑡 K= travel time θ = weighing parameter ( 0 to 1) Time needed from the water at the upstream reach the downstream end at the simulation ∆𝑡 > ∆𝑥𝑗 𝑐𝑗 𝑛 + 𝑢𝑗 𝑛 ∆𝑥𝑗 = 𝑙𝑒𝑛𝑔𝑡ℎ 𝑜𝑓 𝑡ℎ𝑒 𝑟𝑖𝑣𝑒𝑟 𝑟𝑒𝑎𝑐ℎ 𝑐𝑗 𝑛 = 𝑔𝑟𝑎𝑣𝑖𝑡𝑦 𝑤𝑎𝑣𝑒 𝑐𝑒𝑙𝑒𝑟𝑖𝑡𝑦 𝑢𝑗 𝑛 = 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑟𝑒𝑎𝑐ℎ 𝑣𝑒𝑙𝑜𝑐𝑖𝑡𝑦 8
- 10. Calculation setting (2) Inflow hydrographs used for the simulations boundary conditions Hourly discharge data are provided at the upstream area of Surabaya River. Date Maximum Discharge (m3.s-1) Tp (hours) T0.3 (hours) 23-01-2014 to 27-01-2014 (event 1) 367.55 28 18 20-02-2015 to 24-02-2015 (event 2) 337.53 46 32 24-02-2016 to 28-02-2016 (event 3) 335.9 35 27 Tp (lag time) = the amount of time it will take for a river to flood after a period of heavy precipitation T0.3 = the amount of time need from the peak of discharge to 30% peak of discharge Event Initial Water Surface (m) time step (sec) Initial manning roughness channel floodplain slope 1 17.48514757 120 0.01 0.001 2 17.71327015 180 0.01 0.001 3 17.672942 120 0.01 0.001 A uniform routing time interval, Δt = 30min and Δx = 30 m was used for all the test runs. 9
- 11. Routing results (event 1) 1. The hydrograph peak theoretically took 72 minutes to travel between 2 gauges and relatively little attenuation occurred. With the estimation of velocity between 0.74 m/s to 1.8 m/s 2. Peak flow occurred at the upstream end at 30 hours from the initial condition, and at downstream end on the same day within 124 minutes 3. The downstream discharge is slightly unstable on the peak of the hydrograph 0 50 100 150 200 250 300 350 400 0 20 40 60 80 100 120 Discharge(m3/s-1) Time (hours) upstream boundary downstream discharge downstream discharge (simulation) 10
- 12. Routing results (event 2 and 3) 0 50 100 150 200 250 300 350 400 0 20 40 60 80 100 120 Discharge(m3/s-1) Time (hours) 0 50 100 150 200 250 300 350 400 0 20 40 60 80 100 120 Time (hours) upstream boundary downstream discharge downstream discharge (simulation) 1. Event 2 and 3 relatively have long peak hydrograph with duration of 4 to 6 hours 2. Using the same travel time estimation (K) for the routing calculation on event 2 and 3, the downstream simulation result has a close relation with the calculated downstream discharge 3. But, we still need to evaluate the relative error in the time- to peak of the routed discharge hydrograph 11
- 13. Rating-curve rating-curve estimation (tentative) Flood dynamics Parameter estimation Grid-shape accuracy Damage estimation Probabilistic flood Map water elevation-damage relations building and land-use value Methodology Hydrological Analysis Flood routing Damage estimation uncertain 12 Setting of initial condition Analysis of hydrograph shape and water elevation profile
- 14. Model Parameters The table lists the factors which are used as calibration factors in each model implementation. Parameter Sampling range Manning roughness channel 0.01-0.04 Floodplain slope 0.001-0.015 Initial water surface at downstream boundary 17.5 m – 17.8 m Eddy viscosity coefficient 0 to 1 Calibration parameters 1. Capability of the model reproduce hydraulic behaviour of the river and inundated floodplain 2. Time stationary of model parameter (i.e. roughness parameter) 3. Hydraulic information (i.e. flow hydrograph) 13
- 15. Future Tasks 14
Editor's Notes
- #3: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #5: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #6: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #7: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #8: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #9: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #10: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #11: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #12: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #13: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #14: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #15: kinetic energy analysis will be irrelevant because flow velocity is considerably low
- #16: kinetic energy analysis will be irrelevant because flow velocity is considerably low