Numerical Model Analysis
- 1. Numerical Models & Analysis Hydraulic & Hydrologic Considerations in Planning Chuck Shadie Jon Hendrickson Harry Friebel 2011 1
- 2. Objectives • Be able to: – Explain the importance of modeling & analysis in water resources management – Identify model types – Discuss model inputs/outputs 2
- 3. H&H Engineering Models • Purpose: To simulate and analyze physical processes, explore scenarios or do alternative analyses, and assist the decision makers in selecting from alternatives • H&H model results are used in many other types of models (e.g. sediment & nutrient transport, biological response models, statistical analysis) • Types: – Rivers – Coastal – Watersheds – Reservoirs & Lakes – Groundwater 3
- 4. 4 Planning Models • EC 1105-2-407 provides the following definition of a planning model: “any models and analytical tools that planners use to define water resource management problems and opportunities, to formulate potential alternatives to address the problems and take advantage of the opportunities, to evaluate potential effects of alternatives and to support decision-making.”
- 5. Certification of Models •Currently no certification of Engineering Software required •Engineering & Construction (E&C) – Science & Engineering Technology (SET) addressing engineering models & software •EC 1105-2-407, Planning Models Improvement Program: Model Certification •EC specifically for software used in USACE planning studies •Makes a distinction between “planning models” and “engineering models used in planning studies” 5
- 6. Computer models are based on • data • program algorithm (ie. model structure) • user experience & judgment 6
- 7. Garbage In/Out • Need interpretation - not blind acceptance of outputs “Models are for providing insight, not answers” - Tony Thomas
- 8. Internal analysis steps • calibration • validation • iteration 8
- 9. Physical representation by models • spatial variation • temporal variation Low Flow Velocities High Flow Velocities 9
- 10. Modeling Steps • Modeling Existing Conditions – about 50-70% of Effort – Data Collection – Initial Model Setup – Calibration/Verification to Existing Conditions • Modeling Future Without Project Conditions – about 10-20% of effort • Alternative(s) Modeling – about 10-20% of effort – Iterative Process to optimize designs – Involves initial hydraulic design of features – about 10-20% of effort • Comparison of With and Without Project Conditions – Stage Reductions from Alternatives – Impacts to Study Area/Watershed Hydrology & Hydraulics and Ecosystem 10
- 11. Model Sources • Corps’ Hydrologic Engineering Center in Davis, California – HEC series (HEC- HMS, HEC-RAS, etc.) • Engineering Research & Development Center – ERDC (ADH, CH3D, etc.) – System Wide Water Resources Program (SWWRP) https://swwrp.usace.army.mil/ • Private Sector and Academia (MIKE, ADCIRC, etc.) 11
- 12. Science & Engineering Technology (SET) Initiative •Engineering & Construction (E&C) – Science & Engineering Technology (SET) addressing engineering models & software •Conducting inventories & assessments of model software in use •Developing process to document quality of commonly used models •ERDC models •HEC models •Well known & widely used models •Focus is on application •Tool Selection •Quality of Input Data •Model Calibration •Verification of Assumptions •Validation Done thru Agency Technical Review
- 13. Science & Engineering Technology (SET) Initiative •Model Categories •Enterprise (Mandated, Required) •CoP Preferred (Preferred Software – Recommended) •Allowed for Use (Niche Software – Good Enough to Share) •Retired (limited functionality, allowed) •Not Allowed for Use (Obsolete or Technically Inadequate Software)
- 14. Enterprise Tools (Mandated, Required) • S&E Tools: ProjectWise, CWMS, RMS, MS Office Suite, CEFMS, P2. No other tool allowed for use. • Major resourcing requirements for support and funding • Business case to National Mgmt Board (NMB) for approval • Developed by CoP/HQ Sponsor • Implementation Plan • Exception to use needs to be approved by NMB
- 15. CoP Preferred (Preferred Software Option -- Recommended) • Example: Microstation • Software represents single-preferred solution as recommended by CoP-consensus; version should be specified. • Preferred software for use by virtual teams throughout USACE • Software assumed to be in use by a large percentage of USACE personnel with need for this requirement. • Software allowed for use without additional approval and documentation. • If software from this list is not selected, the alternate software selected should come from the "Allowed for Use" list or be coordinated with CoP Technology Team.
- 16. Allowed for Use (Niche Software -- Good Enough to Share) • Alternate software that provides similar capability to existing CoP endorsed package or provides unique analysis capability. •Supports specialized technical or local requirements, or required by customer. •Category included that permits flexibility to ensure people can accomplish their missions. •Software must be recognized as technically viable approach by industry acceptance or some certification/validation process. •The decision to select software from this list is made locally and rationale for selection should be described in study/design documentation.
- 17. Retired • Software determined by CoP technology review to have more limited functionality compared to similar tools listed on "Preferred" or "Allowed for Use". • Software does not best fulfill the needs of the technical functions or requirements. • Obsolete program but still needed for "niche" mission requirements; should be reviewed by CoP for upgrading. •Example: HEC-1 or HEC-2 models developed for previous studies but never converted to HEC-HMS or HEC-RAS Not Allowed for Use Software considered to obsolete or technically inadequate.
- 18. Summary Model Selection • When possible, use Mandated or Preferred Software • If other Software is being used, obtain consensus of Vertical Team ASAP during study • ATR and/or IEPR should review the use and applicability of non-standard software to confirm it is being used appropriately
- 19. Hydrology ____________ WQ & Sediment Land surface Hydraulics _________ WQ & Sediment Channel, floodplain Climatologic (pcp, ET, wind, temp, solar radiation) Topography (Watershed Terrain ) Hydraulic Connectivity (past, present, future) •Lakes •Wetlands •Streams •Drain tile density Ecosystem Flood Damage Analysis Land Use (past, present, future) Soils Surface Water Records Groundwater Data Navigation Hydrologic and Hydraulic Models For Watersheds, Rivers, Streams Data Transfer Data Transfer Channel & Floodplain Geometry Slope Roughness DSS GIS Planning for: Data Transfer Data Transfer 19
- 20. Hydrologic Modeling: Routing of rainfall and runoff through watersheds, reservoirs, channels Hydrograph Parameters Loss Rates (GIS Analysis) Hypothetical Rainfall Event Sub-basin Drainage Areas 20
- 21. Deterministic or Stochastic • Deterministic Models. These models try to represent the physical processes observed in the real world. Typically, such models contain representations of surface runoff, subsurface flow, evapotranspiration, and channel flow, but they can be far more complicated. Deterministic hydrology models can be subdivided into single-event models and continuous simulation models. • Stochastic Models. These models are black box systems, based on data and using mathematical and statistical concepts to link a certain input (for instance rainfall) to the model output (for instance runoff). Commonly used techniques are regression, transfer functions, neural networks and system identification. 21
- 22. Calibration of Hendrum flow gauge (2004) 0 50 100 150 200 250 300 Jan-04 Feb-04 Mar-04 Apr-04 May-04 Jun-04 Jul-04 Aug-04 Sep-04 Oct-04 Nov-04 Dec-04 Jan-05 Date Flow (cms) Final Q Cal Final Sed Calib Obs Hydrologic Modeling Over Periods of Time • Single rainfall event • Continuous simulation (for example, day after day for many years) Flow (Q) Time Rainfall Excess 22
- 23. Hydrologic Modeling Over Timesteps • Timesteps are used for single event and continuous simulation • The rainfall, runoff, and routing occurring during one time step is calculated 23
- 24. Main Channels Sub-watersheds (79) County_Ditch #1 Judicial Ditch #56 South Branch Coon Creek Mashaug Creek Fossum Spring Creek Marsh Creek White Earth Creek Twin Lake Creek WRR upstream of TLC 20 0 20 40 Miles N E W S Sub-watersheds Hydrologic Modeling Over Space Lumped by subwatershed or areas within a watershed having similar characteristics Discrete: Watershed is divided into discrete (distinct) areas 24
- 25. Lumped Hydrologic Modeling Software • HEC-1 • HEC-Hydrologic Modeling System (HMS) • Products--hydrographs 25
- 26. 26 Discrete Hydrologic Modeling Software • Gridded Surface Subsurface Hydrologic Analysis (GSSHA) developed at ERDC • MIKE SHE - Denmark
- 27. Hydraulic Modeling Rivers and Streams Water Surface Velocity Flow Distribution Discharge Channel Geometry Constrictions (bridges, levees) Roughness 27
- 28. Hydraulic Model Typical Uses (Open Channels) • Capacity /stability flowlines • Operation & Maintenance options • Ecosystem physical conditions • Input to sediment transport models 28
- 29. Hydraulic Modeling over Time (Steady-State vs Unsteady-State Modeling) Jan April July Oct Dec Steady Simulations are done using a constant discharge. For example the peak flood flow. No time step is used. Discharge (cfs) Unsteady – Simulations are done for varying discharge over time. For example the Spring flood. A time step is used. 29
- 30. One-dimensional (1D) models simulate the change in parameters in one direction (e.g. downstream to upstream) Two-dimensional (2D) models simulate the change in parameters in two directions (e.g. downstream to upstream and from one side of the channel or river valley to the other) Y Z X Hydraulic Modeling over Space (1D versus 2D) 30
- 31. 1D Model of Water Surface, Pool 5, Upper Miss Routine Regulation vs One Foot Drawdown Main Channel Water Surface, River Q = 90,000 cfs 655 657 659 661 663 665 667 736 738 740 742 744 746 748 750 752 754 River Mile Elevation (1912 Datum) Routine Regulation (90,000 cfs) One Foot Drawdown Elev 662.5 31
- 32. Elev 661 Elev 662.5 ≈ Elev 657 Water Surface Elevation Bathymetry 32
- 33. 1-Dimensional Hydraulic Modeling Software (Open Channels) • HEC-RAS (River Analysis System) • HEC-2 • WSPRO (Federal Highways) 33
- 34. 2-Dimensional Hydraulic Modeling Software (Open Channels) • ADH (Adaptive Hydraulics) developed at ERDC through the SWWRP • FESWMS (Federal Highways) 34
- 35. Subbasin Sediment Yield as % of + Hendrum Sediment Yield Coon Creek, 12.2% Mashaug Creek, 13.1% Marsh Creek, 14.7% Spring Creek, 6.8% WRR u/s Mahnomen, 16.1% South Branch, 21.9% White Earth, 20.8% WRR u/s Twin Lake Creek, 3.8% Twin Lake Creek, 2.3% J.D #56 / C.D. #45, 15.0% County Ditch #1, 3.8% *WRR u/s Twin Valley, 58.8% Fossum, 6.1% +Sub-basin Total above Hendrum = 131% *Sub-basin Total above TV = 64% Sediment Transport Modeling (Watershed Scale) • Hydrologic watershed runoff models that simulate both runoff and water quality parameters. Main Channels Sub-watersheds (79) County_Ditch #1 Judicial Ditch #56 South Branch Coon Creek Mashaug Creek Fossum Spring Creek Marsh Creek White Earth Creek Twin Lake Creek WRR upstream of TLC 20 0 20 40 Miles N E W S Sub-watersheds 35
- 36. Watershed Sediment Transport Modeling Software • GSSHA (USACE, ERDC) • HSPF (US EPA) • SWAT (USDA-ARS) • MIKE-SHE 36
- 37. Sediment Transport Modeling (River Scale) • Hydraulic models that simulate hydraulic parameters, sediment transport capacity, and bed displacement. Yearly Sediment Yield Analysis for WRR 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 450,000 500,000 0 10 20 30 40 50 60 70 80 90 100 River Mile Sediment Yield (Tons) HEC-RAS Sed. Cap. Gauge South Branch WRR Marsh Creek "WRR Project Reach" " SB Project Reach" 37
- 38. Sediment Transport Model Typical Uses in Rivers • Channel stability • Dredging requirements • Water quality • Pump station & diversion design • Maintenance • Ecosystem restoration 38
- 39. Channel Sediment Transport Modeling Software • HEC-6 • HEC-6T • HEC-RAS (with sediment) • ADH 39
- 40. Future Without Conditions: This 7-Year Simulation of Bed Displacement in Pool 5 using ADH Matches Observed Sediment Deposition in Backwater Delta. Forecast Future Without-Project Conditions 2D models are evolving to the point where patterns of erosion and deposition can be predicted. This allows the user to interpret future conditions given various hypothetical flow conditions. The effects of large floods can be analyzed also. 40
- 41. Statistical Model Typical Uses • Frequency • Duration • Timing • Stochastic flow simulation 41
- 42. Statistical Modeling Software • HEC-FFA (frequency) • HEC-EFM (Ecosystem Functions Model) • IHA (Indicators of Hydrologic Alteration, TNC) 42
- 43. Model Time and Cost Model Review • Time and Cost to do the modeling – Many factors (1D or 2D, overbank flows, structure complexity) influence this. Here is a few examples • UMRS 2D model for 2.5 mile wide river valley, 10 mile reach of river, took 3 to 4 months and 50K. • Small River 1D model for sediment budget on 90 mile reach took 2 months and 15K. • Culvert analysis 3 days and 2 or 3K – Model costs are about 1% of a typical project cost • Model Review – Usually a District Quality Control Function – The number of available reviewers is limited. 43
- 45. Coastal Engineering Modeling Considerations • $$$ • Site Characteristics • Data Gathering • Do we have enough engineering and science to make our results believable? • Cost of Tools • Appropriateness of Design • Stakeholder Issues • Environmental Constraints • State of the art modeling changes over life or large feasibility study 45
- 46. Remember, show me the $$$ 2. Sediment Budgets 3. Numerical Models 1. Data Collection & Analysis 4. Physical Models 46
- 47. H&H Coastal Engineers can model: • Sediment Budgets • Wave Climate • Cross-Shore Sediment Transport • Currents and Inlet Processes • Water Levels including Storm Surge • Long-shore Sediment Transport (Shoreline Change) 47
- 48. Sediment Budget Modeling Sediment Budget Models: SBAS, Excel 48
- 49. Trial Simulation details: Outer Grid = 200 x 200m Inner Grid = 100 x 100m Time Step = 3 hours Saved Spectra Locations: CDIP 26 m Waverider 17 m AWAC 11 m AWAC 08 m AWAC 06 m AWAC 05 m Inner Grid Outer Grid Nearshore Wave Modeling: STWAVE, CMS-WAVE, DELFT3D-WAVE USACE-FRF pier in Duck, NC Wave Modeling 49
- 50. Cross-shore Sediment Transport Modeling Cross-shore Sediment Transport Model: SBEACH Initial – 2 Nov 1991 Meas – 11 Jan 1992 Calculated Cross Section Existing Beach Dune Dune 50
- 51. Hydrodynamic (Currents) and Inlet Modeling Hydrodynamic Model: ADCIRC, CMS-Flow, DELFT3D-Current ADCIRC mesh Propagation of Tidal Wave Flood current patterns 51
- 52. Storm Surge Modeling Storm Surge Model: ADCIRC, DELFT3D http://www.nhc.noaa.gov/HAW2/english/surge/surge_big.jpg WATER LEVELS (STORM SURGES) 52
- 53. Longshore Sediment Transport Modeling Long-shore Transport Model: GENESIS 15 20 25 0 100 200 300 Atlantic Ocean Westhampton Shinnecock Inlet Shoreline Position (m) Distance Along Fire Island Baseline (km) 0 5 10 0 100 200 300 Atlantic Ocean Westhampton Moriches Inlet December 1979 Measured April 1995 Measured April 1995 Calculated 53
- 54. Question - What Storms to Model? Remember, each computer run can be time consuming (CPU time), thus not possible to model every possible storm scenario • SPH – Standard Project Hurricane • Probabilistic Approach – Monte Carlo simulation of all of the storm parameters 54
- 55. Frequency-Based Approach (Previous) Storm Erosion-Frequency Curve 100 110 120 130 140 150 160 170 180 190 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Exceedance Frequency Erosion Distance in feet Erosion Distance-Damage Erosion Distance (feet) Property Damage ($000) Flood = Erosion Erosion = $$$ 55
- 56. New Approach (Beach-fx) Event-based Monte Carlo Life Cycle Model Engineering-Economic Planning Tool for Hurricane and Storm Damage Reduction Developed to: • Address analytical shortcomings of traditional, frequency-based approach • More realistic estimates of life-cycle benefits and costs • Generate science-based information to aid decision making • Develop information to communicate plan performance to stakeholders 56
- 57. Modeling Limitations • Depth Integrated (2D) thus no variation with depth – no possible return flow scenario • Most circulation models do not include wave setup Grid cell size Programming constrains 57
- 58. Summary • model is a simplified representation of reality • common analysis steps 58
- 59. Questions “Models are for providing insight, not answers” - Tony Thomas 59