Hydrologic and hydraulic aspects of Corps planning/risk analysis studies: Life after Central Valley hydrology study

Transcription

1 Hydrologic and hydraulic aspects of Corps planning/risk analysis studies: Life after Central Valley hydrology study December 1, 2010 Contents: 1. Hydrologic and hydraulic integration flowchart (logic diagram) for development of risk analysis inputs, and discussion slides covering supporting information to flowchart 2. Discussion slides covering: 1. Review of risk analysis procedures to support planning studies 2. Overview of Central Valley hydrology study 3. Why move from Comp Study to CVHS, and how do CVHS products differ from Comp Study products? 4. Hydrologic and hydraulic aspects of a flood risk analysis study ( with a hydraulic slant) 5. Step-by-step procedure for developing hydrologic and hydraulic products to support a flood risk analysis study 6. Example application (no backwater) 7. Example application (with backwater) 8. Other risk analysis considerations 9. Selecting events for floodplain analysis 3. Document Comparison of engineering requirements for planning/risk analysis studies to products from the Central Valley hydrology study 4. Exhibit In-channel frequency curve options and development steps (2 diagrams)

2 Hydrologic and hydraulic aspects of Corps planning/risk analysis studies: Life after Central Valley hydrology study Summary of materials Discussion slides covering information to support the hydrologic and hydraulic integration flowchart Hydrologic and hydraulic integration flowchart (SPK s logic diagram) for development of risk analysis inputs Discussion slides covering: 1. Review of risk analysis procedures to support planning studies 2. Overview of Central Valley hydrology study 3. Why move from Comp Study to CVHS, and how do CVHS products differ from Comp Study products? 4. Hydrologic and hydraulic aspects of a flood risk analysis study ( with a hydraulic slant) 5. Step-by-step procedure for developing hydrologic and hydraulic products to support a flood risk analysis study 6. Example application (no backwater) 7. Example application (with backwater) 8. Other risk analysis considerations 9. Selecting events for floodplain analysis Document Comparison of engineering requirements for planning/risk analysis studies to products from the Central Valley hydrology study Exhibit In-channel frequency curve options and development steps (2 diagrams) 1

3 Other CVHS materials and documentation Materials and documentation available from CVHS team: Data management plan, Nov 2007 Procedure document, March 2008 Response to comments on Procedure document, September 2008 CVHS product uses, May 2009 Includes description of applications of products for both unsteady and steady flow Project management plan, June 2009 Illustrative example (CVHS procedures and application with the Yuba-Feather system), November 2009 Ungaged watershed analysis procedures, Sept 2010 Technical procedures document, Oct 2010 Includes a series of technical appendixes Various conference presentations from 2006 to 2010 Corps references on in-channel frequency curve analysis and specification in HEC-FDA EM : Hydrologic frequency analysis EM : Flood-runoff analysis EM : Risk-based analysis for flood damage reduction studies PR-71: Documentation and demonstration of a process for risk analysis of proposed modifications to the Sacramento River flood control project (SRFCP) levees HEC-FDA user s manual 2

4 Supporting information for hydrologic and hydraulic integration flowchart for development of risk analysis inputs Corps studies that require H&H info Operating (real time decision making) Permitting (section 408 permits, floodplain delineation for FEMA) Planning (flood risk management, environmental restoration/enhancement) Emergency action (dam and levee safety) Design (setting deterministic dimension for construction) 1

5 History The Central Valley hydrology study (CVHS) will produce updated system-wide hydrology for the Central Valley CVHS differs from the Comp Study The question has been asked: How will Corps hydraulic engineers use the CVHS products for various studies? To help answer this question, flow charts and information were gathered and developed These presentations and documents are a byproduct of this effort H&H integration flowchart 2

6 Key decision points (the asterisks) 1. Can we use existing 2011 CVHS information off-theshelf? 2. Is there a geographical gap? 3. Do any of the CVHS assumptions need to be modified? Which ones? 4. Do the hydraulics need to be updated too? 5. What is an index point (risk analysis point)? 6. What are the various options for defining in-channel frequency curves? And, what factors should be considered when selecting one? 7. What are the basic steps in completing the required products for "case 1" (stage influenced by flow) inchannel frequency curves? 8. What are the basic steps in completing the required products for "case 2" (stage influenced by both flow and backwater) in-channel frequency curves? 9. What are the basic steps in completing the required products for "case 3" (stage influenced by backwater only) in-channel frequency curves? 10.What events are used for floodplain mapping? Q1: Can we use existing 2011 CVHS information off-the-shelf? A1: Yes we can CVHS products include unreg flow time series, unreg flow-frequency curves, reservoir-routed event hydrographs, and the associated models CVHS will produce go by s : channel-routed event hydrographs, transforms, and in-channel regulated flowfrequency curves CVFED and other efforts will turn the go by s into products depending on study types The products and go by s include specific assumptions regarding, for example, flow delivery, reservoir operations, and channel routing If assumptions/conditions are the same for a given study, then they can be used off-the-shelf. If these differ from the needed assumptions / conditions, then analysis is needed to update the appropriate product A description of products and go by s is included in Discussion 2, the comparison of requirements to CVHS products document, CVHS product uses document, as well as other CVHS study documents 3

7 Q2: Is there a geographical gap? A2: Maybe The study reaches for CVHS include those with Federal- State levees in the Central Valley Maps of study reaches and analysis points (frequency analysis locations) are available. Doesn t include every stream in the Central Valley. Modeling boundary includes flood control reservoirs (project reservoirs) downstream to Delta CVHS has developed procedure documents for developing frequency curves for ungaged watersheds, this could support filling in unstudied streams The study area is included in Discussion 2 Q3: Do any of the CVHS assumptions need to be modified? Which ones? A3: Depends on type of study As noted with Q1, if the study assumptions and models used for CVHS and CVFED are appropriate for the study at hand, then no updates may be needed Factors to consider include (from a hydrologic perspective): Unreg frequency curve: Availability of additional years of record? Unreg to reg transform: Status of reservoir simulation model and channel model? Consider reservoir routing assumptions (i.e., starting storages) CVHS assumptions documented in Technical procedure document The hydrologic input development is included in Discussion 2, 4, and 5 4

8 Q4: Do the hydraulics need to be modified too? A4: Depends on study type As noted with Q1, if the study assumptions and models used for CVHS and CVFED are appropriate for the study at hand, then no updates may be needed Factors to consider include (from a hydraulic perspective): Unreg to reg transform: Consider the channel geometry (i.e., levee setbacks and alignment), flow delivery assumptions (performance of upstream levees), and other routing factors Flow to stage transform: Consider specification of channel roughness values and other hydraulic model inputs and changes in system timing Floodplain depths development: Consider methods used to develop those, and which events used The hydraulic input development is included in Discussion 4 and 5 Q5: What is an index point (risk analysis point)? A5: Representative location for a reach and floodplain area, used as the location in which study inputs are integrated to compute risk and risk reduction Index points are identified through coordination with the entire project team as they impact hydrology, hydraulics, geotechnical, economic, and planning activities The index point definition is included in Discussion 1 Index point Impact area Damage reach Structures or other damageable property in floodplain Stream 5

9 Q6a: What are the various options for defining in-channel frequency curves? A6: The option descriptions are included in Discussion 4 Q6b: And, what factors should be considered when selecting one? A6: Factors to consider are: Method used to fit frequency curve: a statistical distribution fitted to unregulated flow values (log Pearson Type III), a graphical curve fit to historical or simulated values, or use of rainfall-runoff models with design storms, for example Presence of backwater at the index point, and other factors that influence stage Mixed cross section approaches are available: Relate the stage in one stream to conditions at another, for example The factors to consider are described in Discussion 4 6

10 Q7: What are the basic steps in completing the required products for "case 1" (stage influenced by flow) in-channel frequency curves? A7: Follow option: Unregulated flow-frequency curve + unreg-reg transform curve + rating curve Use same cross section for development of each required input Exception: If frequency curve developed through rainfall-runoff analysis, use graphical flow-frequency curve Basic analysis steps are included in Discussion 4 and 5 Q8: What are the basic steps in completing the required products for "case 2" (stage influenced by both flow and backwater) inchannel frequency curves? A8: Follow option: Unregulated flow-frequency curve + unreg-reg transform curve + rating curve Use mixed cross section approach for development of each required input, sensitivity analysis needed to finalize the inputs Exception: Alternative would be use of an unreg flowfrequency curve + unreg-stage transform, entered in HEC-FDA as stage-frequency curve Basic analysis steps are included in Discussion 4 and 5 7

11 Q9: What are the basic steps in completing the required products for "case 3" (stage influenced by backwater only) in-channel frequency curves? A9: Use alternative frequency analysis method than that used for CVHS HEC-SSP available option for developing frequency curve to observed or simulated stages See EM Hydrologic frequency analysis Typical case for areas such as the Sac-SJ Delta Basic analysis steps are included in Discussion 4 and 5 Q10: What are the options for selecting events for floodplain mapping? What factors may influence your selection process? A10: Select event based on matching event flow, stage, or a combination to a flow or stage from a frequency curve; which one depends on study From CVHS, a large database of historical and scaled historical has been generated Event(s) from this database can be selected to delineate floodplains based on specified criteria For example, use an event that has the p=0.01 peak flow and the associated expected volume Software tools are being developed to assist with this process (semi-automated) The event selection process is described in Discussion 9 and Appendix I from the IPAST design document 8

12 Hydrologic and hydraulic integration flowchart for development of risk analysis inputs Start PDT: Establish key assumptions Assess the system configuration Select index point locations Q1. Can we use existing 2011 CVHS data off-theshelf?* No Is there a geographical gap? Yes Hydrology augments CVHS hydrology in study area Yes Review risk analysis index points*, complete sensitivity analyses to assess hydraulic conditions Consider both hydrologic and hydraulic assumptions and study area / CVHS frequency analysis locations No What CVHS assumptions need to be updated? Hydrologic OR hydrologic and hydraulic Yes Hydrology revises CVHS hydrology in study area to match feasibility study assumptions Do any of the CVHS assumptions need to be updated? No Hydraulics Hydraulics re routes hydrographs, this may include tranlation of flows or integration of additional reaches Yes Do the hydraulics need to be updated too? Q2. Select risk analysis approach for each index point* No Product: project specific hydrology* Select hydrologic event subset for floodplain development. * Case 1: stage influenced by flow only Follow standard CVHS procedure to develop 1) unreg flow frequency curve, 2) Unreg flow to reg flow transform, and 3) flow to stage transform. If frequency curve derived from rainfall runoff simulations, use graphical flow frequency curve (instead of curve 1 and 2) Case 2: stage influenced by both flow and backwater Consider a "mixed cross section approach" with either the unreg to reg flow transform or the flow to stage transform. Evaluate different options and select "controlling option". Finalize 3 components similar to Case 1. Case 3: stage influenced by backwater only (such as with Sac SJ Delta) CVHS frequency analysis may not be appropriate, consider alternitive frequency analysis methods described in EM Develop full suite of without-project floodplain depths (associated with channel stages) for each index point Define the exterior interior relationship (channel stage to floodplain stage/extent) Provide floodplains to economics for damage computations Develop descriptions of uncertainty about each of the inputs Develop descriptions of uncertainty about each of the inputs Hydraulics develops stagefrequency curve following selected method Notes: Specify in channel frequency curve in HEC FDA using the 3 curves noted Specify in channel frequency curve in HEC FDA using the 3 curves noted Specify in channel frequency curve in HEC FDA Information and factors to consider at decision points, specifically those tagged with an asteriks, are available and described in discussion slides Hydrologic and hydraulic aspects of Corps planning/risk analysis studies: Life after Central Valley hydrology study and the associated materials. Coordinate with hydrology and economics the completion of the HEC FDA model CVHS is the Central Valley hydrology study. A hydrologic analysis commissioned by DWR to support FloodSafe planning efforts. The focus of CVHS is to provide the necessary hydrologic inputs to assess the Federal-State levee system. v1.0 December 1, 2010

13 Discussion topics 1. Review of risk analysis procedures to support planning studies 2. Overview of Central Valley hydrology study 3. Why move from Comp Study to CVHS, and how do CVHS products differ from Comp Study products? 4. Hydrologic and hydraulic aspects of a flood risk analysis study ( with a hydraulic slant) 5. Step-by-step procedure for developing hydrologic and hydraulic products to support a flood risk analysis study 6. Example application (no backwater) 7. Example application (with backwater) 8. Other risk analysis considerations 9. Selecting events for floodplain analysis 1

14 Discussion 1. Review of risk analysis procedures to support planning studies Steps in flood risk management planning study 1. Identifying problems and opportunities 2. Inventorying and forecasting resources 3. Formulating alternative plans 4. Evaluating alternative plans 5. Comparing alternatives plans 6. Selecting recommended plan IWR report 96-R-21 Planning manual ( ts/96r21.pdf) 2

15 H&H analyses provide input to: Identify problems (flood risk) Current Future Evaluate effects of plans (risk reduction) Current Future Planning/risk analysis studies Identify the flood risk associate with each alternative, identify the alternative that maximizes the national economic development (NED) Flood risk is quantified by expected annual damage (EAD) From a hydrologic and hydraulic perspective: Develop the required in-channel frequency curves and channel-floodplain elevation relationships 3

16 Some of the basic engineering components of a flood risk management study In-channel frequency curves Fragility curves Floodplain depths related to inchannel frequency Consequences (damage, lives lost, etc.) System risk analysis guidance 4

17 Terminology: Damage reach, impact area, index point Divide floodplain into impact areas, channel into damage reaches Impact for area/reach related to hydraulic or hydrologic state at index point (cross section) for reach Damage reach Index point Impact area Structures or other damageable property in floodplain Stream a Unregulated discharge Annual exceedance probability b c d Regulated discharge Discharge altered by Yes reservoirs, diversions, etc.? Annual exceedance Unregulated discharge probability No Regulated discharge e f g Use appropriate No Reasonable to use unique stagedischarge Yes hydraulics model to estimate stage for discharge relationship? Channel stage Discharge h Annual exceedance probability Floodplain stage Exposure Channel water surface elevation Annual exceedance probability Damage Damage Channel stage i Floodplain stage j k l Consequence Floodplain depth Annual exceedance probability 5

18 Basics of risk [expected annual damage (EAD)] computations How much flow gets to the index point and what are the annual maximum flows? (Flow delivery) Given an annual maximum flow, what s the annual maximum stage? Given that annual maximum stage, does the levee fail? (yes/no) Given that annual maximum stage and that the levee fails, what s the annual maximum stage in the floodplain? Given that annual maximum stage in the floodplain, what is the consequence? Basics continued Key models: 1. Model of in channel hydrology and hydraulics (frequency curve) 2. Model of levee performance and failure 3. Model of hydraulic consequences (floodplain depths) 4. Model that relates depth to consequence (damage, loss of life, adverse environmental impact) Hydrologic and hydraulic probabilities are only significant in Model 1 (in channel frequency curve) 6

19 7

20 Discussion 2. Overview of Central Valley hydrology study Central Valley hydrology study CVHS commissioned by/paid for by California DWR to support CVFED Products useful for other tasks. How you use products requires policy decisions + professional judgment. Products include flow & volume-frequency curves & hydrographs at >200 locations + system hydrographs--all for the unregulated condition + 1 selected regulated condition CVHS uses Corps-approved procedures. It isn t a re-do of the Comprehensive Study. Results and tools shared, so if you need to make different assumptions, regulated case, etc., re-do the analysis to suit your study 8

21 Study area Basic analysis approach 1. Get historical record (reservoirs and downstream) 2. Remove regulation effects 3. Analyze frequency at 200+ points 4. Add regulation 9

22 What do we mean by unregulated? Remove effect of: Storage ( on stream and off stream ) Weirs Diversions Represents maximum potential flow to the analysis point Unregulated time series not necessarily the same as full natural flow or natural flow Review: Why do we need unregulated flows? Basis of flow frequency analysis Need subset of a natural population Transform non-homogenous record to a homogenous condition (unregulated) Apply statistical model to predict flood peak and volume frequency EM Hydrologic frequency analysis 10

23 Analysis procedure Process diagram to complete procedure 11

24 Products More details are available here: Data management plan, Nov 2007 Procedure document, March 2008 Response to comments on Procedure document, September 2008 CVHS product uses, May 2009 Includes description of applications of products for both unsteady and steady flow Project management plan, June 2009 Ungaged watershed analysis procedures, Sept 2010 Technical procedures document, Oct 2010 Includes a series of technical appendixes 12

25 Discussion 3. Why move from Comp Study to CVHS, and how do CVHS products differ from Comp Study products? 13

26 Comp Study overview SPK completed Sacramento-San Joaquin river system Comprehensive Study (Comp Study) in 2002 As a component of the Comp Study, a basinwide hydrologic analysis completed Detailed technical documentation available online at Per technical documentation, without further investigation, intended for pre-feasibility level applications At present, the hydrologic analysis completed considered best available for many locations in the Central Valley Review of Comp Study hydrologic procedure Developed unregulated flow time series (hydrologic routing) Developed unregulated flow frequency curves Created design storms (storm matrix, storm centering, design hydrographs) Based on iterative, subjective (based on matrix) distribution of events to upstream watersheds, balanced at target frequency curve Must be used within context of composite floodplains 14

27 Review of application of Comp Study hydrology (composite floodplains) Evaluate each storm centering for a specified probability (Mainstem: 5, Tributary: 22) Profile or water surface elevation for that probability = highest value amongst all centerings Centering location means that a statement of frequency can be made there Main Stem Issues raised with Comp Study hydrology Strong reliance on inferred historic storm matrix Looked at an average of what basins experienced, rather than evaluating explicitly Subjective and difficult to update Single temporal distribution of unregulated hydrographs (thus regulated hydrographs) 1997 for Sac, 1955 for SJ Number of storm centerings required to meet current project need (each reach would require entire set of system hydrographs) Simplified local flow estimates All reaches not included 15

28 Key points of CVHS procedure Follow basic procedures from Comp Study, but Replace design storm analysis with floods of record (and scaled historical floods) analysis Enhance estimation of ungaged flow (local flow below projects) Estimate directly from gage data Use rainfall runoff modeling only as needed Use unsteady-flow model for routing both unregulated and regulated flow Thus, specific enhancements: Unregulated flow frequency analysis Coincidence and timing of flows in regulated frequency curves (more objective and less subjective) Advantages of floods of record Build on data and information from Comp Study, but more direct usage of historic events Direct usage of long period of record Direct usage of historic events to analyze storm patterns and coincidence of flows Eliminates need to make assumptions on temporal distribution (flow and rainfall events) Eliminates need to hypothesize storm centerings Facilitates updated of results with increased record length 16

29 Comparison Common tasks Comp Study Floods of record Collect/augment data Yes Yes Unregulated flow time series Unregulated flowfrequency analysis Develop regulated flow time series Hydrographs/ centerings (timing) Ungaged watershed analysis Yes, hydrologic routing, simplified account of local flows Yes Yes, HEC-5 and UNET Storm matrix (averaging of selected historic events), composite floodplains Is not addressed Yes, hydraulic routing, include local flows Yes, coordinated with USGS Yes, HEC-ResSim and hydraulic routing Historical events and scaled historical events, analyze results Standard procedures developed CVHS and Comp Study: User s perspective Study procedures: CVHS = Comp Study + enhancements Study products: CVHS = Comp Study 17

30 What does this change mean to analysts? CVHS should not significantly alter the hydraulic tasks required to accomplish a feasibility study (or any other hydraulic study). In fact, with CVHS: Flow is tied to frequency at significantly more locations Event flows are known at a significantly finer spatial scale (contributing flows for small streams) Allows for more options/judgment in defining inchannel frequency curves (..but not required) What does change: Event selection for determining floodplain depths and extents Event selection What event is used for finding floodplain depths associated with the p=0.01 (100-yr) flow? 18

31 Comp Study centerings Results from Comp Study centerings are tabulated in HEC-DSS files The HEC-DSS filename and F-part label suggests a frequency SC yr CVHS and event selection CVHS results require event selection Events can be selected on a project-by-project basis Standard events can be selected (and accepted) as representative of a 100-yr flow 19

32 A new accepted standard /simplification needed? Where needed, new simplified assumptions may be required What do I do to get 100 yr flow at Sacramento Verona for floodplain mapping: Event magnitude Event hydrograph for Sacramento Verona 50 yr 1997 event, 1986 event X yr 1997 event X 1.2, 1986 event x yr 1997 event X 1.4, 1986 event X 2.0 The accepted practice will evolve with use, same as with the Comp Study Originally 27 centerings In practice, narrowed down per study based on initial sensitivity analysis Event selection guidance In short, select the event(s) from the large dataset that matches the frequency curve at the index point Large database of historical and scaled historical events Historical events scaled at 0.2 increments Database already developed, and events used in transform development Various descriptions of the event selection process in CVHS product uses and other separate technical memorandums (discuss process and criteria available) Software tool can be used to assist in process, but not required 20

33 Discussion 4. Hydrologic and hydraulic aspects of a flood risk analysis study ( with a hydraulic slant) 21

34 Some of the basic engineering components of a flood risk management study In-channel frequency curves Fragility curves Floodplain depths related to inchannel frequency Consequences (damage, lives lost, etc.) In-channel frequency curves What is needed: Given a specified annual exceedence probability, what stage do I expect? Options for defining curve: 1. Use analytical (unreg) flow-frequency curve + unreg-reg transform curve + rating curve 2. Use graphical flow-frequency curve + rating curve 3. Use stage-frequency curve directly 22

35 Option Referred to as analytical flow-frequency curve: Frequency curve defined by specification of parameters of a statistical distribution rather than specific flowprobability pairs Use when underlying frequency analysis completed with a fitted statistical distribution (log Pearson type III) Appropriate for most of the analysis points used in CVHS Facilitates the description of uncertainty in each component, and consistent with EMs Approach can be adapted for modeling complex hydraulic scenarios by using mixed cross section approach (described later) Option 2 + Referred to as graphical flow-frequency curve: Specific flowprobability pairs are used to describe the frequency curve rather than parameters of a statistical distribution (not required to be regulated flows) Used when underlying frequency analysis not the result of a fitted statistical distribution (log Pearson type III) Most commonly used when the flow-frequency curve was developed from rainfall-runoff model simulations with design storms, as with the ungaged watersheds of CVHS Available option when flow or stage can not be predicted a single source. Source = watershed or stream. However, the previous option could also be used. 23

36 Option 3 Appropriate when underlying frequency analysis based on fitting a graphical frequency curve to observed stages, such as the recent Delta stage-frequency analyses. Note that this is a frequency analysis option not used for CVHS Used when no direct relationship of flow to stage is found, such as in the Sac-SJ Delta (tidally driven stages) In practice, as it relates to CVHS, this is equivalent to choosing to use an unreg flow to stage transform. However, this can t be input directly into HEC-FDA. Thus, external to HEC-FDA, this stage-frequency curve would be developed by combing the unreg flow-frequency curve and the unreg flow to stage transform. The resultant stage-frequency curve would then be put in HEC-FDA. What factors that should be considered when developing the functions? Factors that affect flow delivery to the index point (these in channel inputs) Upstream levee performance Flow diversions, weirs Channel hydraulics Evaluation condition (project and alternative specific) Reservoir operations Set back levees Future conditions and land use changes 24

37 Evaluating complex hydraulic conditions Flexibility offered through judgment: Frequency curve should be developed at a location driving the hydraulic conditions, not necessary at the same cross section as the index point Unregulated to regulated transform can relate the unregulated flow at the frequency curve location to the regulated flow at the location of interest Stage-discharge curve relates the regulated flow at the location of interest to a regulated stage at the location of interest Who makes the decision? Coordination needed between hydrologic engineers and hydraulic engineers Considerations and work that goes into development of the in-channel frequency curves include: Unregulated flow time series development Flood-flow frequency analysis Reservoir simulation Channel routing of reservoir releases, computation of stages 25

38 Here s one way to break out the development effort Team: Hydrology Develop unregulated flow time series, fit a flowfrequency curve Team: Hydrology Configure HEC-ResSim with chosen study assumptions. Route events (historical and scaled historical) through the reservoir system and to downstream handoff points. Team: Hydraulics Configure HEC-RAS with chosen assumptions. Route flows (historical and scaled historical) through channel system and compute stages. Use developed tools to extract event datasets and fit transforms. Confirm transform results with Hydrology. Team: Hydraulics Using same tools and extracted datasets from above, fit transforms. What s available off the shelf for in-channel frequency curve development CVHS is completing a go by set of frequency curves Reservoir releases are being routed through the system using best available channel models For flow delivery assumptions, levees are assumed to overtop, but not fail For a planning/risk analysis study The go by can be reviewed and be judged sufficient for a given application Factors to consider include study specific assumptions for: flow delivery, reservoir routing, channel geometry, and channel configuration If a specific study assumption differs, then products must be re-developed as necessary 26

39 Locations for which off-the-shelf hydrologic information are available? CVHS study area does not include every stream in the Central Valley CVHS modeling extents Modeling extents go from flood control reservoirs to Delta for both Sacramento and San Joaquin river systems Focused on channels with State-Federal levees Hydrographs are available any location within modeling extents Maps available from CVHS team showing modeling extents CVHS frequency analysis Over 200+ locations have frequency curves Set at locations within modeling extents where significant changes in the volume-frequency curve is expected References on in-channel frequency curve analysis and specification in HEC-FDA EM : Hydrologic frequency analysis EM : Flood-runoff analysis EM : Risk-based analysis for flood damage reduction studies PR-71: Documentation and demonstration of a process for risk analysis of proposed modifications to the Sacramento River flood control project (SRFCP) levees HEC-FDA user s manual 27

40 Some of the basic engineering components of a flood risk management study In-channel frequency curves Fragility curves Floodplain depths related to inchannel frequency Consequences (damage, lives lost, etc.) Floodplain information What is needed? Relationship of channel elevation to floodplain extents and depths (floodplains tied to channel conditions) Options: 1. Channel elevation = floodplain elevation 2. Channel elevation related to floodplain elevations (pooled flooding / level floodplain) 3. Channel elevation related to floodplain surface (analyzed with FLO2D, for example) 28

41 Floodplain considerations What s the volume available to fill the floodplain? And, where is it coming from? Given a flood event and levee failure where does the water go? And, how deep is it? Topography of the floodplain, i.e. a level pool or sloped surface The resulting floodplain must tie back to a channel elevation Decisions that need to be made for this Factors relevant to channel-floodplain elevation relationship development Which events (historical or selected historical events) to use? Events must span range of flood events, of interest, to adequately define relationship For example, must define to at least extents of the levee fragility curve (channel elevation of 0.0 probability of failure to elevation of 1.0 probability of failure) Levee failure properties (breach width, height) 29

42 Here s one way to break out the development effort Team: Hydraulics Decide on appropriate method to delineate floodplain and the breach characteristics. Coordinate with Hydrology on events used for floodplain delineation. Select 7 events that span, at least, the range of the defined levee fragility curve. Delineate floodplain and complete channel-floodplain elevation relationship. 30

43 Discussion 5. Step-by-step procedure for developing hydrologic and hydraulic products to support a flood risk analysis study 1. Are previous system analyses available for use in current study? Are assumptions used for CVHS go by appropriate? Decision should be coordinated with study team. Flow delivery assumptions (upstream levee failure) Reservoir routing assumptions Channel routing assumptions Channel geometry and configuration Consistency with study without-project condition If no, must re-develop in-channel relationships Consult CVHS study documentation (flow chart) 31

44 2. Are previous studies available for the current study? Study area is an extension of the previous step: Are previous analyses available? Identify the required study reaches and index point for the study Are those reaches/index points covered by previous studies, such as CVHS CVHS focused on reaches that are part of the Federal-State levee system, it doesn t address every stream in the Central Valley CVHS study team has maps of the CVHS study reaches and frequency analysis locations 3. Decide on appropriate hydraulic analysis approach for study area, and modeling extents Consider the need for a steady flow or unsteady flow hydraulic analysis Consider the need of the modeling extents Hydraulic model should extend through the area of interest, and ideally downstream far enough that the specific downstream boundary condition does not affect water surface elevations in the study area Is a multi-reach analysis required? 32

45 4. Select option for defining in-channel frequency curve Complete sensitivity analysis of the hydrologic and hydraulic conditions at the required index points For sensitivity analysis, complete water surface profile computations for a range of likely flows and downstream boundary conditions Are stages at the index point influenced by the downstream boundary condition or a downstream confluence (backwater)? Decide on risk analysis approach (option for defining in-channel frequency curve). If significant backwater is found, go to step 5. Otherwise, proceed to step If significant backwater conditions are found Consider the following: What location (unregulated frequency curve) is the best predictor of stage at the index point? If needed, try different options and select the location that controls for the range of stages of interest Build a mixed cross section transform, options include: 1. Unregulated to regulated flow transform 2. Regulated flow to stage transform Need for an alternative approach (other than CVHS) such as a direct fitting of a frequency curve to observed stages (likely scenario in the Delta) 33

46 6. Select events for floodplain delineation Identify a set of historical and scaled historical events with maximum channel water surface elevations that span the required range of channel stages Required range dictated by range to which levee fragility curve defined Consider the volume of each event hydrograph to the expected volume from the CVHS routings If multiple events have the same maximum stage, select an event with the most likely volume associated with a given maximum water surface elevation. If needed, select additional events and develop a most likely channel to floodplain elevation relationship 34

47 CVHS illustrative example document provides a detailed write up on an application of CVHS procedures to the Yuba-Feather system. The example focuses on development of the inchannel frequency curve and evaluation of alternatives. It also provides an example of selecting events for floodplain development. This document is available from the CVHS team. Discussion 6. Example application (no backwater): HEC-FDA model development for American River 35

48 Situation and application An index point on the American River downstream of Folsom Dam For American River ERR risk analysis files, the in-channel frequency curve was defined with: Analytical frequency curve (log Pearson type III) Unregulated flow to regulated flow transform Regulated flow to stage transform In-channel frequency curve input to HEC-FDA shown in next few slides Specification of analytical frequency curve 36

49 Confidence bounds about curve Specification of unreg to reg flow transform 37

50 Specification of flow to stage transform Geotechnical information Levee properties and performance assumptions are specified by user for HEC-FDA Definition in HEC-FDA shown in next few slides 38

51 Specification of levee properties Specification of levee fragility curve Note that fragility curves are now typically defined from 0.0 probability to 1.0 probability. As defined above, below the 0.15 probability point, HEC-FDA presumes there is a 0.0 probability of failure. 39

52 Floodplain depths related to channel elevations Floodplain depths and extents developed using 2D hydraulics model Floodplain surfaces then tied back to channel elevations Specification for HEC-FDA shown in next slide 40

53 Floodplain consequence information Structure database developed, including information on: Structure elevation Structure value Content value Structure type Structure damage as a function of flooding depth HEC-FDA computes an aggregated elevationdamage relationship for a reach Example elevation-damage relationship in HEC- FDA shown in next slide Specification of structure properties 41

54 Discussion 7. Example application (with backwater): HEC-FDA model development for lower Bear River 42

55 Situation and application An index point on the Bear River upstream of the Feather River confluence Based on sensitivity studies with available channel models, it was found that the stages are influenced by both flows in Bear River and Feather River Even with backwater, there is only 1 true frequency curve at given index point To decide on an in-channel frequency curve definition approach, a sensitivity analysis was completed Index point location Feather River Yuba River Bear River 43

56 Sensitivity analysis Index point WS Max WS - EWO100 WS Max WS - HWO100 WS Max WS - MWO100 WS Max WS - LWO100 Elevation (ft) Main Channel Distance (ft) Graphic for illustrative purposes only, not based on actual Bear River model runs. Complexity in selecting option to describe in-channel frequency curve Presumption is that an entirely different approach will not be utilized, i.e., switch from an unreg flowfrequency analysis (Bulletin 17B) to a graphical frequency curve fitted to historical or simulated data (such as for Delta frequency analysis) Thus, must decide which unregulated flowfrequency curve dominates at the index point Since stages at index point influenced by both Bear River and Feather River, consider both frequency curves as candidates with respective mixed cross section transform (response surface) For transform development, use flows above area influenced by backwater 44

57 Stage at index point function of 2 flows (response surface) Graphic for illustrative purposes only, not based on actual Bear River model runs. Rotated response surface: Stage as function of Bear River flows Graphic for illustrative purposes only, not based on actual Bear River model runs. 45

58 Plot (same values): Stage as function of Bear River flows Graphic for illustrative purposes only, not based on actual Bear River model runs. Options considered and option selected Two options considered (using flow to stage transform mixed cross section approach): 1. Unreg freq curve Feather River + Unreg flow Feather River to regulated flow Feather River transform + flow Feather River to stage Bear River transform 2. Unreg freq curve Bear River + Unreg flow Bear River to regulated flow Bear River transform + flow Bear River to stage Bear River transform Note that with this second option, the backwater from the Feather River is accounted for in the dataset (response surface) used to develop the flow Bear River to stage Bear River transform Using both options noted, find which dominates, i.e. gives the higher stage for the range of events of interest Select the dominant option for use at the index point Flow locations used above are upstream of backwater area, i.e. not the same cross section as the index point Remember that flow to stage transform is a response matrix based off of many simulations of historical events considering the actual coincident events for each 46

59 Events selected for floodplain development Process used from CVHS event selection procedure Because the area was governed by backwater, the criteria used for selecting the event was peak stage Alternatively, peak flow on the Bear River could have been used as well 47

60 Discussion 8. Other risk analysis considerations Other considerations Interior drainage System levee performance decision Interaction with Sac-SJ Delta Multiple source flooding to impact areas Future conditions, climate change River F River F Index point 2 River Y Impact area Impact area Index point 1 Index point 1 River I Index point 4 River B Index point 3 (A) (B) 48

61 Discussion 9. Selecting events for floodplain analysis 49

62 Event selection From CVHS, a large database of events developed to support transform development Same database used for transform development Database focused on largest system events experienced, and with those, scaled versions as well Scaling factors include: 1.2, 1.4, 1.6, 1.8, 2.0, etc. Event selection = query database of events and find one that meets the specified criteria Match peak regulated flow and/or volume Match peak stage and/or duration Match unregulated flow for a given duration Or, a combination of above Most likely, match peak flow or stage Guidance Technical memo on event selection and various illustrations of the process in CVHS documentation For combing through database of event properties and selecting events, software tools can be utilized to facilitate process Design of such a document completed Nov

63 Practical guidance For most applications, an event is needed that has a specified probability peak and the associated volume First, identify candidate events with the appropriate peak flow Then, from that subset, identify the events that have volume characteristics consistent with the expected hydrograph (as described in the CVHS product uses document) For event selection criteria Match regulated flow when you want the 100yr flow down a given reach Match stage when you want the 100yr stage near a confluence subject to backwater See attached excerpt for further guidance: Requirements and design document for software tool for data processing 51

64 Comparison of engineering requirements for planning/risk analysis studies to products from the Central Valley hydrology study 1

65 Table of contents Overview... 3 Why does the Corps do a feasibility planning study?... 3 How does the Corps do a feasibility planning study?... 3 What happens in a feasibility planning study?... 3 What information is required to complete a feasibility study?... 3 What are the hydrologic outputs of the CVHS?... 4 In which components of a feasibility planning study can CVHS products be used?... 5 Can products of the CVHS be used off-the-shelf as inputs to any of the components of a feasibility planning study?... 5 How does the study team determine whether CVHS procedures or products can be used in a feasibility planning study?... 5 What is in the rest of this document?... 6 Requirements for risk analysis to support flood damage reduction feasibility studies... 7 Excerpt from draft EM Comparison of hydrologic and hydraulic requirements to CVHS products

66 Overview Why does the Corps do a feasibility planning study? The Corps completes feasibility studies to find the flood risk reduction measure (or combination of measures) that best meets the economic, performance, and environmental objectives. It is primarily an examination of economic impact of flooding with and without the proposed alternative(s) under current and future development conditions, from an engineering perspective. How does the Corps do a feasibility planning study? The Corps uses a hydroeconomic model that combines the characteristics of the flooding and the development in the floodplain to characterize the economic dimensions of the flood problem. These economic dimensions are expressed in terms of expected annual damages (EAD). (Source: nmdl.asp, hereinafter referred to as USACE NED online manual.) What happens in a feasibility planning study? In a feasibility study, the set of feasible alternatives identified in the reconnaissance phase is refined and the search narrowed. Plans are nominated with specific locations and sizes of measures and operating policies. Detailed hydrologic and hydraulic studies for all conditions are completed as necessary to establish channel capacities, structure configurations, levels of protection, interior flood-control requirements, residual or induced flooding, etc. (ER ). Then, the economic objective function is evaluated, and satisfaction of the performance and environmental standards is tested. Feasible solutions are retained, inferior solutions are abandoned, and the cycle continues. The NED and locally preferred plans are identified from the final array of alternatives. The process concludes with a recommended plan for design and implementation. (Source: HEC-HMS app guide, Table 1, page 2.) What information is required to complete a feasibility study? A feasibility study aims to quantify the economic effect of flooding in a floodplain. The measure of economic effect is expected annual damages (EAD). To arrive at that answer requires an understanding of (1) the hydrologic and hydraulic behavior of the stream, (2) the hydraulic behavior of the floodplain, (3) the hydraulic relationship between the stream and the floodplain, including the behavior of flood risk reduction measures that influence that relationship, and (4) the people and property who may be subject to the flood hazard in the floodplain. In simple terms, the Corps hydroeconomic model used to estimate EAD has 3 basic inputs (USACE NED online manual): Flow-frequency relationship Channel stage-flow relationship Floodplain stage-damage relationship 3

67 And, the output of the model is a damage-frequency relationship. A very simple representation of the inputs and outputs is shown in Figure 1. Flow-frequency Stage-flow Stage-frequency Stage-damage Damage-frequency Figure 1. The Corps hydroeconomic model has 3 inputs: flow-frequency, stage-flow, and stage-damage relationships. Considering the types of inputs needed and the goal of a feasibility study, a feasibility study can be divided into 4 components: Development of in-channel frequency curves. Development of geotechnical information. Development of floodplain information. Development of flood damage information. What are the hydrologic outputs of the CVHS? CVHS products include both hydrologic output and accompanying technical documentation. The CVHS products are described in CVHS products uses document, CVHS procedures document, and CVHS technical procedures document. For completeness, we include here a list of the study products: Unregulated flow-frequency curves at key locations in the Central Valley Unregulated flow times series (serves as basis of the frequency analysis and transform development) HEC-ResSim models of the Central Valley Identification of the critical duration for key locations in the Central Valley Partial regulated flow times series (Unregulated flows routed through the reservoir simulation model, but not the channel routing model) In addition to the CVHS products above, included in the study effort is development of go by products that complete the process described in the 4

68 CVHS procedures document and the Technical procedures document. Although these go by products may be useful for other applications (such as a first look if the regulated frequency curves may change), the intended purpose is to provide guidance on the appropriate use of the CVHS study products. The go by products include: Completed regulated flow times series (Unregulated flows routed through the reservoir simulation model and the channel routing model, given a specific set of routing assumptions.) Unregulated to regulated flow transform Regulated flow to stage transform Beyond the actual results of the CVHS, the CVHS procedures can inform hydrologic analyses required for feasibility studies because the CVHS procedures are Corps procedures. In addition, the technical documentation being developed can provide valuable guidance to analysts because it describes issues that arose in the course of the CVHS and how those issues were resolved. In which components of a feasibility planning study can CVHS products be used? CVHS procedures and products are applicable to developing in-channel frequency curves and floodplain stage-channel stage transforms. Can products of the CVHS be used off-the-shelf as inputs to any of the components of a feasibility planning study? Yes, products can under the right circumstances and after a review of the study needs and the assumptions used in development of the products. CVHS is not producing a library of models, procedures, and/or functions that serve as inputs for every feasibility planning study in the Central Valley. CVHS is a study funded by DWR to meet very specific needs and a specific study area. For a given study, it may be found that the models are appropriate, but specific analysis assumptions differ thus additional hydrologic and/or hydraulic analysis is needed to modify the products to match those assumptions. How does the study team determine whether CVHS procedures or products can be used in a feasibility planning study? Because every flood problem is unique, the first step is to articulate the exact question(s) the Corps needs to answer. Everything about the study the objectives, inputs, strategies, procedures, and tools is determined by the question(s). Thus, the first box in a feasibility flow chart is to articulate the study question. After defining the study question, the analyst must determine whether the assumptions built into the CVHS go by products appropriate for the study at hand. To determine the validity of the CVHS assumptions, the study team must ask, Is the study area in the existing baseline CVHS analysis? 5

69 If the answer is no, CVHS products are not going to be of use, and the hydrologic and hydraulic inputs for the study at hand must be developed. If the study area is in the existing baseline CVHS analysis, the study team must examine the following: Models and computer programs used, including reservoir simulation and channel routing Time: Whether there are additional years of record since CVHS and the current study Study area and study reaches Reservoir routing assumptions Channel routing assumptions, include flow delivery assumptions (upstream levee performance criteria used) and levee alignments. If any of these assumptions are invalid for the present study, new in-channel relationships must be developed. The accompanying set of discussion slides provides background and information when making the decisions. Further, various CVHS documents are available that describe the assumptions made in developing the products for that study. What is in the rest of this document? In the remaining portion of this document, we describe further the required hydrologic and hydraulic inputs required for a feasibility study. In the last chapter, we compare the products of CVHS to the feasibility study requirements. 6

70 Requirements for risk analysis to support flood damage reduction feasibility studies The required inputs for risk analysis studies are described in EM , Risk-based analysis for flood damage reduction studies, dated August 1996 and available at: HEC is currently in the process of updating EM Below is an excerpt from the draft document that describes the requirements for a risk analysis. Excerpt from draft EM Computation of Economic Risk. a. As noted above, one goal of a Corps flood risk management planning study is to find the plan that yields the greatest contribution to national economic development (NED) consistent with the environmental and social objectives of the project. In the simplest terms, that NED contribution of a proposed plan is the net benefit of the plan, which is computed as NB = B + B + ( E[ X ] E[ X ]) C L I without with in which NB = net benefit; B L = location benefit, the value of making floodplain land available for new economic uses, such as shifting from agricultural to industrial use; B I = intensification benefit, the value of intensifying use of the land, such as shifting from lower-value to higher-value or higher-yield crops; E[X without ] = expected value of without-project economic flood-inundation damage; E[X with ] = expected value of economic damage if the plan is implemented; and C = total cost of implementing, operating, maintaining, repairing, replacing, and rehabilitating (OMRR&R) the plan. The terms E[X without ] and E[X with ] are the economic risk without and with the plan, and the difference is the risk reduction, also known as the inundation-reduction benefit. b. The economic risk values are computed by deriving and integrating a probability model of flood damage. Practically, this computation may be accomplished as illustrated in Figure 2-1. For project life cycle risk analyses, risk is computed considering likely temporal changes of watershed conditions. Thus the computations illustrated are completed without and with project conditions, for the base year and for future years, considering and simulating the most likely conditions in those future years. c. Tasks illustrated in Figure 2-1 are as follows: 7

71 (1) A statistical analysis of the unregulated discharge is performed, creating the discharge-probability function (item a in the figure). This function is derived at the location for which risk is to be assessed. Such a location is referred to herein as an index location or index point. Development of the dischargeprobability function is described in more detail in Chapter 3 of this manual. (2) The impact of reservoirs, diversions, and other natural or engineered water control features in the system is considered (item b). If those features alter flows significantly by storing or diverting water, the unregulated discharge-probability function must be transformed to a regulated function. To do so, a discharge transform relationship (shown as item c) is developed with historical data or a model of the system. Derivation and application of this transform is described in more detail in Chapter 3 of this EM. Using the transform, the unregulated probability function is transformed to a regulated discharge-probability function (item d). (3) Damage is correlated principally with depth of inundation, so the discharge-probability function (regulated or unregulated) must be transformed to a stage-probability function. For this step, a hydraulic model is used, and a decision must be taken (in item f) regarding the appropriate type of model to use. For coastal flooding situations or riverine flooding in which backwater has a significant impact, a detailed hydraulics model (item e) is used, considering coincidence of downstream boundary conditions with upstream flows. For simpler cases for which a unique relationship between discharge and stage is appropriate, a simpler transform function can be developed and used. This transform (item g) is similar to a discharge-stage function (rating curve). Chapter 3 of this manual describes development of the transform. (4) With the appropriate stage-discharge transform, the discharge-probability function is transformed to a stage-probability function (item h). This function describes the exposure to flooding at the index point. It predicts the stage in the channel at the index location for the full range of likely flood events. 8

72 a Unregulated discharge Annual exceedance probability b c d Discharge altered by reservoirs, diversions, etc.? No Yes Regulated discharge Unregulated discharge Regulated discharge Annual exceedance probability e f g Use appropriate hydraulics model to estimate stage for discharge No Reasonable to use unique stagedischarge relationship? Yes Channel stage Discharge h Annual exceedance probability Floodplain stage Exposure Channel stage Annual exceedance probability Damage Damage Channel stage i Floodplain stage j k l Consequence Floodplain stage Annual exceedance probability Figure 2-1. Schematic of Risk Computation (5) The exposure of people and property in the floodplain adjacent to the channel is assessed. This assessment is accomplished by developing and applying a relationship of stage in the channel to stage at various locations within the floodplain (item i in the figure). This relationship is often referred to as an interior-exterior relationship, differentiating the interior (landside) from the exterior (waterside) of the hydraulic system. For a simple case in which water overflows the channel and moves across a floodplain when the channel capacity is exceeded, this function is influenced principally by the terrain in the floodplain. But if the floodplain is protected by a levee, for example, the relationship will explicitly represent the protection afforded by the levee when it functions as designed. That is, interior stage will be zero until exterior stage exceeds the top of the levee, as shown in item i in the figure. Chapter 7 of this manual addresses development of this function. 9

73 (6) With the interior-exterior function, the channel stage-probability function (item h) can be transformed to develop an interior (floodplain) stage-probability function, as illustrated by item j in the figure. This function characterizes the likelihood of people and property in the floodplain being exposed to the flooding hazard. (7) The final task in risk analysis and computation of the expected value of inundation damage is assessment of the consequence of inundation. This assessment is accomplished by developing a stage-damage function (item k) for property in the interior floodplain. Chapter 6 of this manual describes development of this function. With the stage-damage function, the stageprobability function (item j) is transformed to develop a damage-probability function (item l). Integrating (finding the area under) this curve yields expected annual damage (EAD), or the economic risk associated with the condition of interest. d. This description presumes that the entire discharge-probability function with which this analysis begins will be transformed to a channel stage-probability function, that the channel stage-probability function will be transformed to a floodplain stage-probability function, and so on until a damage-probability function is derived. That damage-probability function is integrated to compute economic risk the long-term average value of damage. 10

74 Comparison of hydrologic and hydraulic requirements to CVHS products Table 1 compares the feasibility study requirements described above to the products of the CVHS. In column 3, we describe to what extent the given requirement is met with CVHS. And, in column 4, we describe some of the factors that should be considered when evaluating each specific component. As noted above, for any application of CVHS products, a comparison between the current study needs and assumptions should be made with the development of the CVHS products. This comparison may require input from several members of the study team including hydrologic engineers, hydraulic engineers, geotechnical engineers, and project planners. 11

75 Table 1. Comparison of feasibility study requirements to CVHS products ID (1) Product (2) Developed from CVHS (3) A Yes The unregulated flow-frequency curve is developed based on the derived unregulated flow time series and statistical analysis. Notes (4) Factors to consider in evaluating this requirement are the method of analysis and whether or not additional years of record are available to refine the analysis. CVHS developed frequency curves at over 200 locations in the Central Valley. The frequency curves are for streams with a Federal-State levee. The hydrology section typically takes the lead on developing the unregulated flow-frequency curve. C Partially The unregulated to regulated flow transform is partially developed from CVHS. The transform is a product of matching the unregulated flow time series and the regulated flow time series. The unregulated flow time series is a product of CVHS. The regulated flow time series is only a partial product of CVHS, but is completed as a go by. Factors to consider in evaluating this requirement are: Is the reservoir simulation model appropriate? Are the reservoir routing assumptions (starting storage, operating criteria) appropriate? Has a newer channel routing model been developed? Are the representations of the channel geometry, levees, and floodplains appropriate? Is the assumption of flow delivery (upstream levee performance) appropriate? Both the hydrology and hydraulic section typically have input on developing the unregulated to regulated flow transform. D Developed through combining of A and C The regulated flow-frequency curve is developed by combining the unregulated flowfrequency curve, for the identified critical duration, with the unregulated to regulated flow transform. The critical duration is a function of the storage in the system: as the storage increases, the critical duration is likely to increase. Typically the hydrologic engineer would complete the critical duration analysis. 12

76 G Partially The flow to stage transform is developed by analyzing the regulated flow time series. The flow to stage transform is similar to a typical rating curve in that it can convert a given flow to stage based on the channel geometry. However, because it is derived based on analysis of historical events, it also contains information on the timing of coincident events. This is especially important in areas of backwater, where flows from one stream affect stages on another. H Partially May be developed through combining of D and G CVHS developed go by flow to stage transforms based on the go by regulated flow time series. Note that in application, the cross section used to extract a given flow does not necessarily need to be the same as the cross section used to extract the stage. This mixed cross section approach may be needed for hydraulically complex areas. Typically the hydraulic engineer would complete the flow to stage transform after completing the regulated flow time series. The regulated flow-frequency curve is developed by combining the unregulated flowfrequency curve, for the identified critical duration, with the unregulated to regulated flow transform. Alternatively, a stage-frequency analysis can be completed directly. With this type of analysis approach, a graphical frequency analysis is completed by direct analysis of observed or simulated stage values. This is typical, for example, in analysis of frequency in the Delta. This type of analysis was not completed in the CVHS. I No The channel stage to floodplain stage relationship (exterior to interior relationship) relates the floodplain extent and depth to conditions in the channel. Factors to consider when developing the relationship include: What events are used for floodplain development, and are these sufficient? Are the at site levee break assumptions (breach type, formation time and size) appropriate? What method was used to delineate the floodplain? For complex cases, several points on the exterior to interior relationship could be developed and then a best fit line used to establish the relationship used for the risk analysis. Also, note that the probability associated with the events used to develop this relationship is not significant. However, the range of events should be large enough to span the definition of the levee fragility curve or the range of events of interest. Typically the hydraulic engineer would complete the floodplain analysis. However, the hydrologic engineer may provide guidance on selecting which events to use for the floodplain analysis. 13

77 J No Developed through combining of H and I Typically the floodplain stage-frequency curve is not developed explicitly. The development and use of this curve is handled by HEC-FDA. K No The floodplain stage to damage relationship is developed by developing and analyzing the structures subject to flooding in the study area. The structure inventory includes information on the number, type, and value of structures in the study area. Also, relationships of depth of flooding to damage for each structure are needed to develop the aggregate (total) floodplain stage to damage relationship. Typically the economist would complete the floodplain stage to damage relationship. I No Developed through combining of H and I Typically the floodplain stage-frequency curve is not developed explicitly. The development and use of this curve is handled by HEC-FDA. 14