Presentation_Niraj_Final
Download as PPTX, PDF1 like190 views
This document summarizes a master's thesis defense presentation on predicting travel time and developing flood inundation maps for flood warning systems. The presentation covers: (1) background on floods and need for warning systems; (2) objectives to quantify effects of elevation data and roughness on travel time/inundation area and develop maps; (3) methodology using HEC-RAS modeling; and (4) results on effects of topography/roughness and developing maps for various flood stages along the Grand River, Ohio.
Presentation_Niraj_Final
- 1. Department of Civil/Environmental and Chemical Engineering Youngstown State University Master’s Thesis Defense April 25, 2016 Niraj Lamichhane Committee: Dr. Suresh Sharma Dr. Tony Vercellino, and Dr. Bradley A. Shellito Prediction of Travel Time and Development of Flood Inundation Maps for Flood Warning System Including Ice Jam Scenario : A Case Study of the Grand River, Ohio
- 2. Thesis Overview Background/Need of study Objectives Materials Methodology Results/Discussions Conclusion Acknowledgements 2April 25, 2016
- 3. Background/Need of study Flood/Flood Warning System: • Most common form of natural disaster • ~ 520 millions people affected every year worldwide • Property loss ~ 50 - 60 billions USD • In US: loss of 140 people lives/year & property loss ~ 8 billions USD • Ice jam flooding - common event in Northern region of the US that result the loss of lives and millions’ dollar worth properties • Flood warning system - warn people sufficient time ahead 3April 25, 2016
- 4. Background General Terms No Flood 2006 Flood Location: ECKART America Corporation, 830 E Erie St., Painesville OH Source : USGS Report 2006 Flood Flood travel time : Time taken by flood to travel from one place to another Inundation area : The areas covered by flood during peak flooding time cont. 4April 25, 2016
- 5. I. To quantify the effects of resolution of elevation datasets and Manning’s roughness in travel time and inundation area prediction II. To develop an approach for flood warning system and generate flood inundation maps for the Grand River III. To assess the potential impacts in river stage and hydraulic structures due to winter ice cover/ice jam Flood maps could be uploaded online. Objectives 5April 25, 2016
- 6. Theoretical Description HEC-RAS (Hydraulic Engineering Center - River Analysis System) Widely accepted hydraulic tool, developed by United States Army Corps of Engineers (USACE) Two types of simulation in HEC-RAS • Steady Flow • Unsteady Flow Chapter 2 6April 25, 2016
- 7. Theoretical Description Unsteady flow • Flow velocity changes with time Continuity Equation Momentum Equation Steady flow • Flow velocity do not change with time Energy Equation cont. Chapter 2 7April 25, 2016
- 8. • Located in Northeastern region of Ohio • Watershed Area : 705 mi2 • Length of Grand River : 102.7 mi Materials Study Area: Map of USA Grand River Watershed Grand River Outlet City of Painesville Harpersfield Ohio Chapter 2 8April 25, 2016
- 9. • Harpersfield to mouth at Fairport Harbor: 32.2 mi. in length • Three major tributaries: Mill, Paine, and Big Creek Materials Study Area: cont. Source: Google Map Harpersfield City of Painesville Fairport Harbor 04211820 04212100 Chapter 2 9April 25, 2016
- 10. Methodology Collection of input data Preparation of geospatial data in HEC-GeoRAS Simulation in HEC-RAS Flood maps generation in HEC-GeoRAS 1. Pre -processing 2. Processing 3. Post -processing Overall Methodology Chapter 2Objective I 10April 25, 2016
- 11. Methodology cont. 1. Pre-processing: Collection of input data • Elevation datasets (LiDAR, 10m DEM, 30m DEM) • Detail topographic survey of the river • Land cover • Discharge/Stage • Aerial photographs • Detail drawing of bridges Collection of input data Preparation of geo-spatial data in HEC-GeoRAS Simulation in HEC-RAS Flood maps generation in HEC-GeoRAS Chapter 2Objective I 11April 25, 2016 LiDAR – Light Detection and Ranging DEM – Digital Elevation Model Process for obtaining LiDAR data
- 12. Methodology cont. 1. Pre-processing: Geo-spatial data preparation Collection of input data Preparation of geo-spatial data in HEC-GeoRAS Simulation in HEC-RAS Flood maps generation in HEC-GeoRAS • Create geo-spatial data (river, cross sections, bank stations, flow path) for all datasets (LiDAR DEM, 10m DEM, and 30m DEM). • Export to HEC-RAS Chapter 2Objective I Geo-spatial data preparation in HEC-GeoRAS 12April 25, 2016 Surveyed points Points where cross section elevation are compared for various elevation datasets
- 13. Methodology cont. 2. Processing: Simulation in HEC-RAS • Import datasets from HEC-GeoRAS Collection of input data Preparation of geo-spatial data in HEC-GeoRAS Simulation in HEC-RAS Flood maps generation in HEC-GeoRAS LiDAR DEM 10m DEM 30m DEM LiDAR with survey 10m DEM with survey 30m DEM with survey • Surveyed in 77 cross sections • Integration of survey data in all datasets. • Elevation datasets for modeling Chapter 2Objective I 13April 25, 2016
- 14. Methodology cont. 2. Processing: Simulation in HEC-RAS Collection of input data Preparation of geo-spatial data in HEC-GeoRAS Simulation in HEC-RAS Flood maps generation in HEC-GeoRAS • Apply boundary condition (discharge, stage data, etc.) • Run simulation for 10, 50, 100, 500 return floods and 2006 floods using6 datasets • Export data to HEC-GeoRAS Chapter 2Objective I HEC-RAS Model D/S streamgage station U/S streamgage station 04212100 04211820 14April 25, 2016
- 15. Collection of input data Preparation of geo-spatial data in HEC-GeoRAS Simulation in HEC-RAS Flood maps generation in HEC-GeoRAS Methodology cont. 2. Processing: Simulation in HEC-RAS Calibration/Validation • Manning’s roughness - only variable in HEC-RAS • Calibrated/validated for 8 events between 1996-1998 for stage and discharge • Model evaluation - Using 4 statistical criteria • Calibrated roughness value : 0.035 - channel, 0.15 - flood plain Chapter 2Objective I 15April 25, 2016 R2=0.99 R2=0.99
- 16. Methodology cont. Collection of input data Preparation of geo-spatial data in HEC-GeoRAS Simulation in HEC-RAS Flood maps generation in HEC-GeoRAS 3. Post-processing: Flood maps generation • Hydraulic computations imported from HEC-RAS • Flood maps generation using HEC-GeoRAS Chapter 2Objective I Comparison of predicted flood inundation maps for 2006 Flood when different elevation datasets were used Area in red color covers all the area of other colors. 16April 25, 2016
- 17. Results & DiscussionsObjective I Effect of topography (Resolution of elevation datasets) : • Cross section greatly vary for different resolution datasets • 10m DEM - better in channel than LiDAR Chapter 2 • Topographic survey performed. • LiDAR with integration of survey data was taken for comparison and generating maps 17April 25, 2016
- 18. Results & DiscussionsObjective I Effect of topography: Flood travel time (to City of Painesville) • Travel time – calculated for 5 different floods using 6 datasets Travel time difference: Chapter 2 Note: Comparison was done with the result obtained from LiDAR with survey • % difference in travel time was highest for coarse elevation dataset. 18April 25, 2016 11.03%-15.01%1.19%-3.35% 10.24%-11.75% 3.67%-4.87%8.73%-10.52%
- 19. Results & DiscussionsObjective I Effect of Topography: Inundation area • Inundation maps – generated for 5 different floods using 6 datasets Inundation area difference Chapter 2 Note: Comparison was done with the result obtained from LiDAR with survey data • % difference in inundation area is highest for coarse elevation dataset 19April 25, 2016 32.56%-44.52%3.55%-7.80% 8.08%-15.48% 17.36%-18.71%10.85%-18.71%
- 20. Results & DiscussionsObjective I Effect of Manning’s roughness “n” : Flood travel time (Table 2-6) Chapter 2 In channel sections • 0.035, 0.030, 0.025, and 0.020 • Varied in channel & constant in flood plains • Travel time: highest for higher “n” (0.035) • Decrement in travel time: 7.48% - 22.35% In flood plains • 0.15, 0.10, 0.09, and 0.7 • Varied in flood plains & constant in channel • Travel time: lowest for higher “n” (0.15) • Increment in travel time: 0.60% - 3.45% 20April 25, 2016
- 21. Results & DiscussionsObjective I Effect of Manning’s roughness: Inundation area Chapter 2 In channel section In flood plain Note: Comparison was done with the result obtained when roughness value is 0.035 in channel & 0.15 in flood plain • The sensitivity of Manning’s roughness - found to be more in channel than in floodplain 21April 25, 2016
- 22. Results & DiscussionsObjective I Effect of Manning’s roughness: Inundation area Chapter 2 Difference in predicted inundation area for various sets of Manning’s roughness values 0.020_0.15 0.025_0.15 0.030_0.15 0.035_0.07 0.035_0.09 0.035_0.10 0.035_0.15 World Imagery Ü 22April 25, 2016
- 23. To develop an approach for flood warning system and generate inundation maps for various flood stages in the Grand River Objective II 23April 25, 2016
- 24. MethodologyObjective II Chapter 3 Overall Flood Warning Approach Development of hydraulic model Calibrate steady model and run the simulation Preparation of digital flood inundation maps Installation of siren system Evacuation time Recommendation for better warning system 24April 25, 2016
- 25. Results & DiscussionsObjective II Chapter 3 Rating curve for streamgage at City of Painesville Validation of rating curve Steady flow : Calibrated using high-water marks of 2006 flood calibration Rating curve : Developed using 75% exceedance flow of 1988-2005 Q=166.67H1.79 Validation : 2006-2015 R2=0.93 25April 25, 2016
- 26. Results & DiscussionsObjective II Chapter 3 Calculation of travel time & development of flood-inundation maps • Flood travel time for 12 different flood stages • Developed digital inundation maps can be uploaded online. ( E.g. http://water.weather.gov/ahps/inundation.php ) Travel time and inundation area for various flood stages 26April 25, 2016
- 27. Results & DiscussionsObjective II Chapter 3 Flood damages along the Grand River • Major affected places: Hidden-Valley Park near S. Madison Rd, Helen Hazen Wyman Park, Mill Stone Drive, Steel Ave. and Grand River Ave, Kiwanis Recreation Park, Western Reserve & Fairport Harbor Yacht Clubs, Ram Island, Hidden Harbor Drive area, Fairport Harbor • Affected bridges: Bridges at Vrooman Rd., Lakeland Freeway & Fairport Rd. Vrooman Bridge - more critical (water level increased>3 ft above road level) from the analysis as well as from the historical data • More than 100 houses, many roads and parks are susceptible to 500 year return period or greater floods 27April 25, 2016
- 28. Results & DiscussionsObjective II Chapter 3 Flood inundation map for 19.35 ft stage (2006 flood) near City of Painesville S.MadisonRd Huntington Road Western Reserve & Fairport Yacht Club 28April 25, 2016
- 29. Results & DiscussionsObjective II Chapter 3 1D and 2D animation of 2006 flood in the Grand River Flood Animation in HEC-RAS 29April 25, 2016
- 30. Results & DiscussionsObjective II Chapter 3 Water level at various bridges for 19.35 ft flood stage (2006 Flood) 30April 25, 2016
- 31. Results & DiscussionsObjective II Chapter 3 Recommendation for more effective automated flood warning system • Reestablish discontinued streamgage (04211820) at Harpersfield • Establish new streamgages for major creeks like Mill, Paine and Big Creeks • Install automated warning system containing a rain gauge, a Geostationary Operational Environmental Satellite (GOES) transmitter, a Radio Frequency transmitter using Automated Local Evaluation in Real Time (ALERT) protocol and a voice modem 31April 25, 2016
- 32. Objective III To assess the potential impacts in river stage and hydraulic structures due to winter ice cover/ice jam 32April 25, 2016
- 33. Objective III Near Fairport Harbor Near Fairport Rd. bridge 33April 25, 2016 Objective III Chapter 4 Historical Source: CRREL Ice Jam Database, USACE (2015)
- 34. 34 MethodologyObjective III Chapter 4 Overall Approach April 25, 2016 Historical analysis and estimate ice thickness Prepare input data and run the simulation for various scenarios Compare and analyze results of those scenarios Generate flood inundation maps Maximum estimated ice thickness= 10 inches for 1977/1978 period Where ti = thickness of ice 𝛼 = coefficient for wind and snow cover AFDD = Accumulated Freezing Degree Days Ta = Temperature of air Historical analysis and ice thickness estimation 𝑡𝑖=𝛼√𝐴𝐹𝐷𝐷 𝐴𝐹𝐷𝐷 = (32−𝑇 𝑎) Modified Stefan’s equation
- 35. MethodologyObjective III Chapter 4 Ice jam simulation was run for 3 different scenarios using 5 winter discharge a) without ice cover & jam with bridges b) with ice cover and ice jam with bridge c) with ice cover & ice jam but no bridges 35April 25, 2016 cont. 5 different winter discharge values a) 25 Percentile b) 50 Percentile c) 75 Percentile d) 90 Percentile e) 100 Percentile
- 36. Results & DiscussionsObjective III Chapter 4 Relationship discharge, AFDD & precipitation • Increase in river discharge due to ice melting not due to precipitation. Increase in discharge 1976/1977 1984/1985 36April 25, 2016
- 37. Results & DiscussionsObjective III Chapter 4 Comparison of water level • Overall increase in river stage (~2 ft.) due to ice cover/jam • Increase in stage was higher mostly at the upstream part of bridge (Table 4-6) Longitudinal profile of bridge at South Madison road Longitudinal profile of bridge at Blair Road Effect of bridge: Increase in river stage by 4.16 ft Increase in river stage by 1.21 ft 37April 25, 2016 4.16 ft 1.21 ft
- 38. Results & DiscussionsObjective III Chapter 4 Vrooman bridge - more crucial - water level crossed deck level Affected several times - 2007, 2010, 2011, and 2014 Large volume of ice blocks in jam location - can result flooding when it melts 38April 25, 2016
- 39. Results & DiscussionsObjective III Chapter 4 Flood inundation maps - Generated for 5 different winter discharge Increase in inundation area for all return period floods 22 % for 100 percentile & 52% for 25 percentile Flood plain map for highest winter flow (100 percentile flow) 39April 25, 2016
- 40. Conclusions/Recommendations Chapter 5 • Travel time and inundation area: high for coarse elevation datasets • Channel is better represented by 10m DEM than by LiDAR data • Manning’s roughness – more sensitive in channel than in flood plains • > 100 houses, many roads, parks and bridges affected by 500 years return period flood or the higher ones • Significant effects of ice cover/jam observed in river stage and flood • Installation of siren system in suitable location to warn people 40April 25, 2016
- 41. Conclusions/Recommendations Chapter 5 • Results can be uploaded in National/Regional Portal System • Further calibration and validation with detail recording of high-water marks and ice jam recordings might help develop more reliable model 41April 25, 2016
- 42. Acknowledgements Advisor: Dr. Suresh Sharma Committee Members: Dr. Tony Vercellino and Dr. Bradley A. Shellito • Dr. Anwarul Islam, Department Chair • Ohio Sea Grant – for providing research grant • Christopher R. Goodel • My parents and families • Friends and well wishers: Binod, Kamal, Shobha, Aashish, Janga, Bhishan, Sabin, Sanjay, Abhijeet and many others • Department of Civil/Environmental and Chemical Engineering April 25, 2016
- 43. Thank you ! April 25, 2016