Mapping the Zone: Improving Flood Map Accuracy (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

2 Flood Mapping and Flood Insurance P eople have always settled near rivers and coasts, ance and floodplain management purposes, FEMA’s but population growth and the commensu- Flood Insurance Rate Maps (FIRMs) are now used rate expansion of the built environment have for many other purposes, including disaster mitigation, increased their risk of losses to flooding over time. land use planning, and emergency response. This chap- From the mid 1930s to the late 1960s, the federal gov- ter describes how FIRMs are created and maintained ernment dealt with flood hazard primarily by building and how information technology is used to update and flood control structures, such as dams and levees. Flood share flood-related data. insurance was not available because (1) the people most likely to buy it were those most prone to flood- FLOOD INSURANCE RATE MAPS ing, which meant that private companies could not profitably provide coverage at an affordable rate, and Flood Insurance Rate Maps delineate flood hazard (2) existing data about flood extent were insufficient to areas, identify flood insurance rate zones within these accurately assess flood risk. areas, and may show elevation and other data related to Escalating flood losses and disaster relief costs, par- flooding. The information that appears on individual ticularly the widespread damage caused by ­Hurricane maps (and the accuracy of those data) depends on the Betsy, led to the creation of the National Flood Insur- type of flood hazard (e.g., riverine, coastal) and the way ance Program (NFIP) in 1968. The objectives of the the flood hazard was studied. The primary information NFIP, which is administered by the Federal Emer- portrayed on FIRMs is discussed below. gency Management Agency (FEMA), are to identify and map floodprone communities and to make flood Flood Hazard Areas insurance available in communities that adopt and enforce floodplain management regulations (e.g., zon- Three types of flood hazard areas are shown on ing, building requirements, special-purpose floodplain FIRMs: ordinances). More than 20,400 communities currently participate in the NFIP. Although created for insur- 1. Special Flood Hazard Areas (SFHAs) subject to a 1 percent or greater chance of flooding in any given The private sector stopped covering flood losses in 1929 ­after a year (44 CFR 59.1). The 1 percent annual chance flood, series of devastating floods, including a 1927 flood of the ­Mississippi River, which inundated 13 million acres and killed several hundred also known as the base flood or 100-year flood, is the people. See American Institutes for Research, 2002, A Chronology NFIP standard for regulating new development in the of Major Events Affecting the National Flood Insurance Program, floodplain and determining where mandatory flood 78 pp., available at <http://www.fema.gov/­library/viewRecord. insurance coverage is required. do?id=2601>.  See <http://www.floodsmart.gov/floodsmart/pages/about/­ 2. Moderate flood hazard areas, including areas community_preparedness_ratings.jsp>. subject to a 0.2 percent annual chance (500-year) 13 

14 MAPPING THE ZONE flood (44 CFR 64.3) and SFHAs that are either small (subject to storm surge where wave heights for the (drainage areas of less than 1 square mile), expected 1 percent annual chance flood are 3 feet or greater). to flood less than 1 foot, or protected by levees from Moderate flood areas are designated as shaded Zone X, the 1 percent annual chance flood. Flood insurance is and areas of minimal flood hazard include unshaded voluntary, although lenders may require flood insurance Zone X and zones for which flood hazard has not been for structures. In addition, communities may choose determined. Example portions of FIRMs showing to regulate land use and siting of critical services and some of these zones in a riverine area and a coastal area emergency response facilities in these areas. are shown in Figures 2.1 and 2.2. 3. Areas in which flood hazards are minimal (e.g., FEMA’s Map Modernization Program was less than a 0.2 percent annual chance of flooding) or intended to produce digital FIRMs for all of the undetermined, but still possible. These areas are not nation’s 1 percent annual chance floodplains, but a subject to federal regulations on insurance or land use, midcourse adjustment gave priority to densely popu- although communities and lenders may impose such lated areas, where more lives and property are at risk requirements. (FEMA, 2006a). Risk-related priorities were based on total population, rate of population growth, number Each of these areas is divided into flood insurance of housing units, number of flood insurance policies rate zones, which designate the level and type of flood and claims, number of repetitive loss properties and hazard (Box 2.1). The majority of SFHAs are either claims, and number of declared flood disasters. This riverine and lacustrine (area along the shore of a lake decision shifted emphasis from the risk of occurrence or closed water basin) A zones (subject to a 1 percent of a 1 percent annual chance flood to the risk of more annual chance flood) or coastal A zones and V zones significant flood damage. BOX 2.1  Definitions of the Most Common Flood Insurance Rate Zones Zone A: Special Flood Hazard Area (SFHA), defined as land subject to a 1 percent annual chance of flooding. The zone is divided into several sub- types, including • A (or unnumbered or approximate A): SFHA in which detailed analyses were not carried out and the base flood elevation is not shown. • AE, A1 through A30: SFHA in which the water surface elevation has been determined and is shown on the map. Zone V: Coastal SFHA subject to high velocity wave action from storms or seismic sources. The zone is divided into several subtypes, including • V (or unnumbered V): Coastal SFHA for which water surface elevations are not shown. •  1 through V30, VE: Coastal SFHA with velocity hazard and water surface elevation determined and shown on the map. The VE designation V is replacing the earlier numbered V designations. Shaded Zone X, Zone B: Area of moderate flood hazard or future conditions flood hazard, generally defined as the 0.2 percent annual chance flood. Unshaded Zone X, Zone C: Area of minimal flood hazard, commonly understood to have a lower probability of flooding than the moderate hazard area. The numbers for zones A1 through A30 were determined by computing the difference between the 1 percent annual chance and 10 percent annual chance flood elevation, multiplying by 10, then applying a conversion factor (FEMA, 1983). The process was similar for numbered V zones, although different multiplication and conversion factors were used. Modernized maps have replaced the A1 through A30 designations with an AE designation, and the B and C designations with an X designation. ________ SOURCE: 44 CFR 59.1 and 44 CFR 64.3. 

FLOOD MAPPING AND FLOOD INSURANCE 15 COASTAL BASE FLOOD ELEVATIONS APPLY ONLY LANDWARD OF 0.0 NGVD Figure 2-1 River.eps FIGURE 2.1  Extracted image from a paper map (FIRMette) bitmap image Figure 2-2.eps FIGURE 2.2  Example of a FIRMette for a coastal area near for a riverine area in Greenville, South Carolina. The left side bitmap image Myrtle Beach, South Carolina. The figure shows VE zones (SFHAs shows an approximate A zone (SFHA, shaded dark gray), subject to coastal wave action) and associated elevations at the where no elevation or floodway information is provided. The point on the ground to which the wave runs up during the 1 per- right side of the image shows an AE zone (SFHA, shaded dark cent annual chance flood. Landward, the flood zones transition gray) with lettered cross sections, base flood elevations (wavy to Zone AE with their associated base flood elevations. SOURCE: lines with elevation), and floodway (hatched area bounded by FEMA’s Map Service Center, <http://msc.fema.gov/>. heavy dashed lines), and a shaded Zone X (moderate flood hazard area, shaded light gray). The other areas are classified as unshaded Zone X (minimal flood hazard). SOURCE: FEMA’s Map Service Center, <http://msc.fema.gov/>. Base Flood Elevations both regulating and insuring properties commensurate with the true risk of flooding. The base flood elevation (BFE) is the computed Despite the importance of accurate BFEs in elevation of a flood having a 1 percent chance of being S ­ pecial Flood Hazard Areas, in unnumbered A and V equaled or exceeded in a given year (base flood). It zones they are generally only estimated using approxi- accounts for the volume and velocity of water mov- mate methods (see “Types of Flood Studies” below), ing through the watershed and reflects the cumulative which estimate key variables such as water volume. effects of topography, soils, vegetation, surface perme- The determination of flood risk is less certain in these ability, and other factors. The BFE is the regulatory areas, so local communities may require a safety factor standard for the elevation or floodproofing of structures, (known as freeboard) above the estimated BFE for and the relationship between the BFE and the eleva- additional financial protection. However, even where tion of a structure also determines the flood insurance BFEs are established with more certainty, communities premium. In general, the higher the first floor elevation, may impose freeboard to help protect against damage the lower the insurance premium. Consequently, the resulting from multiple 1 percent annual chance floods accuracy of BFEs on the flood maps is important for in a given year or higher than expected flood waters. 

16 MAPPING THE ZONE Future Hydrologic Conditions 1. Scoping, including identifying flood risk, assess- ing immediate and future needs (e.g., development of Flood hazard information presented on FIRMs floodprone areas), and determining what type of flood is typically based on conditions in the floodplain and study is feasible with available resources. This step is watershed that existed when the map was made. In carried out by FEMA in conjunction with state and recent years, however, some growing communities local officials. have become interested in projecting how future land 2. Development, including collecting techni- use and development in the watershed will affect the cal data, modeling, creating a preliminary map, and extent of the floodplain, and using those projections to performing quality control and quality assurance. regulate floodplain development. In response, FEMA Modeling and map production are carried out by a issued a final rule in November 2001 that allows com- FEMA mapping partner (e.g., contractor, state or local munities the option of showing future conditions flood- government employee). Once the technical work has plains based on land use change on the FIRM, along been completed, it is reviewed by a FEMA contractor, with the required existing conditions floodplain. The then preliminary maps are prepared and released to the decision about how to use information on future condi- relevant communities for review. tions for regulatory decisions is left to the community. 3. Adoption, including periods for public com- FEMA continues to use data on existing conditions for ment and appeal. FEMA, contractors, and state and flood insurance purposes and has yet to consider the local government agencies involved in the process must effects of climate change, long-term erosion of coastal respond to comments made within the appeal period. areas, or long-term trends in hydrologic records on the Once the protest and appeal process is completed and determination of future conditions. By mid-century, any outstanding issues are resolved, the maps are final- the absolute flood elevations on structures along the ized and FEMA issues a Letter of Final Determina- Gulf Coast will be higher than at the time of their con- tion. The local community then has up to six months to struction because of sea level rise and subsidence. The adopt the new map and update its floodplain manage- U.S. Army Corps of Engineers is including ­location- ment ordinances, if necessary, before the map becomes dependent adjustments in the design of structures to effective (i.e., the most current legal map for regulatory compensate for the expected rise. and insurance purposes). FLOOD MAP PRODUCTION Data for Digital FIRMs The process for producing flood maps involves Digital Flood Insurance Rate Maps (DFIRMs) three main phases (Figure 2.3): are built from three layers of information (Figure 2.4). FIGURE 2.3  Flood map production process. SOURCE: Courtesy of Michael Godesky, FEMA. Figure 2-3.eps bitmap image 

FLOOD MAPPING AND FLOOD INSURANCE 17 FIGURE 2.4  Major components of DFIRMs. SOURCE: Modified from Maune (2007). Reprinted with permission from the American Society for Photogrammetry and Remote Sensing. Figure 2-4.eps bitmap image The base map imagery (orthophoto or vector) shows 2003, V. 1 and Appendix C). Cross sections, based on planimetric features such as roads, rivers, and buildings. topographic data collected in the field or scaled from Digital elevation data are overlain to give each feature U.S. Geological Survey topographic quadrangle maps, in the base map image a vertical position. Finally, flood are taken to define the floodplain. The locations of hazard data, collected and modeled by surveyors and these cross sections are chosen to capture variations in engineers, are overlain to produce the DFIRM. t ­ opography and possible obstructions to flow. Methods for Mapping Flood Hazard Coastal Flooding. The coasts of the Great Lakes and the oceans are subject to severe flooding from storm FEMA’s methods for mapping the most common surge, the result of high winds and air pressure changes flood hazards are summarized below and discussed in that push water toward the shore. Coastal flood studies more detail in Chapters 4 and 5. assess the effects of storm surge and wave action and determine base flood elevations (FEMA, 2003, V. 1 and Riverine Flooding. Overbank flooding, the most Appendix D). The study process is similar to that for common type of flooding in our nation, occurs when riverine flooding, except that instead of cross sections, downstream channels receive more water than they can transects are surveyed perpendicular to the coastline, accommodate due to rain, snowmelt, blockage of chan- yielding onshore and offshore ground elevations. The nels by ice or debris, or dam or levee failure. Mapping elevations are then used to compute the expected height riverine flood hazards requires hydrologic and ­hydraulic of wave crests and wave runup that are added to the studies to determine ground elevations, the depth storm surge as it approaches the shoreline. of floodwaters, the width of floodplains, the amount of water that will be carried by watercourses during Shallow Flooding. Even a minimal rise in water level flood events, and obstructions to water flow (FEMA, can lead to extensive inundation in relatively flat areas 

18 MAPPING THE ZONE TABLE 2.1  Types of Flood Study Methods Detailed Detailed (Riverine) (Coastal) Limited Detailed Approximate Redelineation Base mapa Orthophotography or Orthophotography or Orthophotography or Orthophotography or Orthophotography or vector vector vector vector vector Hydrology Regression equations, Historical water marks Regression equations or Analysis not technically Uses previously published (flows) stream gage data, or and tide gage data stream gage data reviewed flow information rainfall-runoff models Hydraulics Modeled (steady state or Modeled storm surge, Modeled (steady Analysis not technically Uses previously determined (flood dynamic) with detailed waves, erosion, and state) without survey reviewed elevations elevations) structure survey data wave runup information on bridge or culvert structures Mapping Typical zone Typical zone Zone representation Typical zone New floodplain boundaries presentation representations include representations include limited to AE representations include matching new base map AE with floodway AE and VE A and V information; Letters of Map Change (LOMCs) Study report Provides flow estimates, Provides shoreline Provides flood elevation Not applicable Republishes flood study floodway data tables, profiles and stillwater and profile information and flood elevation data tables profiles Cost per $10,000-$25,000 Approximately $9300 $1500-$5000 $250-$2000 (typically mileb (typically $13,500) (typically $3000) $900) aAll flood study methods use best available base map at the time of production; the current FEMA minimum standard is digital orthoquarter quadrangles. bSOURCE: Paul Rooney, FEMA. such as Florida. The low relief and absence of channels studies, (3) approximate studies, and (4) redelineation. in these areas can cause water to flow in sheets across Each approach yields different information, and the the land surface, often in unpredictable directions. decision about which to use depends on the type of Drainage ditches and stormwater management facili- flood hazard, the resources available, and the risk of flood ties may be overloaded by storms more severe than the damage. Coastal flood mapping is currently done using 10 percent annual chance floods for which they are the equivalent of detailed studies. Table 2.1 compares usually designed. Ponding of rainfall in depressions the information used and presented in the four study often creates local floods, which may be alleviated types. by infiltration, evaporation, or mechanical pumping. Detailed studies are most expensive and provide S ­ hallow flood studies yield a uniform depth of flood- the most information about flood hazards, establishing ing, which is either added to the ground elevation or base flood elevations, special and moderate flood haz- used to determine a single base flood elevation for ard areas, and where appropriate, floodways. Limited a large area (FEMA, 2003, V. 1 and Appendix E). detailed studies provide a reasonable representation of When adequate topographic data are not available, the floodplain limits and often a base flood elevation. cross sections may be taken to determine storage vol- Structures such as bridges or culverts are represented ume for areas subject to ponding and average flood in the models, but their dimensions and elevations are depths for areas subject to sheet flow. not verified in the field. Approximate studies yield Types of Flood Studies A floodway is the river channel and adjacent land areas required to discharge the base flood without significantly increasing flood The four main approaches used to study riverine heights. Coastal high hazard areas and tidal rivers, which experi- ence regular fluctuations in water surface elevations, do not have flood hazard are (1) detailed studies, (2) limited detailed designated floodways. 

FLOOD MAPPING AND FLOOD INSURANCE 19 an approximate outline of the floodplain, but no base the maps to establish zoning and building standards; flood elevations, floodways, moderate hazard areas, or insurance companies, lenders, real estate agencies, other details. Although comparison of the floodplain and property owners who use the maps to determine boundaries to a topographic map provides an estimate whether flood insurance is required; and government of the base flood elevation, this estimate is inadequate officials who use the maps to support infrastructure, for regulatory purposes. FEMA provides written transportation, and other planning and to prepare for guidance (FEMA, 1995) and a computer program for and respond to flooding. calculating approximate water surface elevations on Mapping costs and map accuracy are directly related, open channels based on specified field measurements and funding for the Map Modernization Program was (see Appendix A for a list of methods used to estimate insufficient to produce high-quality maps of the entire BFEs in approximate studies). nation (GAO, 2004). Moreover, the Government Redelineation studies are aimed at producing digi- Accountability Office, Congress, and stakeholders were tal representations of flood maps as part of a national concerned about the accuracy of the mapped floodplain digital flood layer. Redelineation uses existing flood boundaries that were to be digitized (FEMA, 2006a). In elevation information and redraws the flood ­boundaries response, FEMA made a midcourse adjustment to the on new or updated topographic maps. All approved Map Modernization Program. Two criteria were used to changes to the flood maps (see “Map Maintenance” quantify map and engineering accuracy: (1) a floodplain below) are incorporated, resulting in an updated map boundary standard and (2) validation guidelines for flood that reflects the most current effective flood elevation data and engineering analyses used to delineate flood- and hazard information. In contrast, the digital conver- plains. The floodplain boundary standard is a ­statistical sion method simply scans the flood boundaries shown measure of the vertical discrepancy between the water on paper maps and transfers them to a new digital map. surface elevation at the boundary of the floodplain and Fifty-four percent of the stream miles mapped until the land surface elevation at that location (FEMA, 2007 were the result of the digital conversion process. 2007c). The measure is computed at a sequence of points This approach was discontinued for new studies follow­ along the floodplain boundary and a specified percentage ing FEMA’s midcourse adjustment (FEMA, 2006a) of these points must lie within defined error ranges that and prior to issuance of a new floodplain boundary are more strict for maps produced from detailed studies standard (see below). than for maps produced from approximate studies. The standard is aimed at ensuring that the flood maps match FEMA’S MAP MODERNIZATION PROGRAM the topographic data used, although adherence to the standard does not itself validate the topographic data. The nation has floodplains along approximately The validation guidelines for flood data and engineering 3.5 million miles of rivers and coasts (FEMA, 2006a). analyses are a set of rules which define whether a flood Prior to 2003, only 1 million miles had been mapped, study done in the past is adequate for current use or often at a lower quality than meets NFIP needs, and whether physical, hydrologic, or methodological changes most flood maps and related products were outdated since the time of the original study are sufficiently great and available only in paper form. FEMA’s Map Mod- to warrant an updated study (FEMA, 2007b). The ernization Program was established to collect new flood intention of these changes was to improve the percentage data in unmapped areas, to update or validate existing of studies meeting these criteria while relaxing the origi- flood data, and to create digital flood maps. The federal nal program goal of complete digital flood map coverage government invested about $1 billion in this 2003-2008 of the nation. Doing so is consistent with stakeholders’ mapping effort, and considerable matching funds were comments on the midcourse adjustment that “The goal provided by FEMA’s state government and local com- of digitization of the nation’s flood maps . . . should not munity partners. This investment in more accurate outweigh the goal of achieving accuracy on the newly maps was intended to benefit communities that use updated maps” (FEMA, 2008c, p. 22). A map of the data q ­ uality standards achieved for U.S. counties by March Presentation to the committee by Patrick Sacbibit, FEMA, on 2008 is shown in Figure 2.5. November 8, 2007. 

20 MAPPING THE ZONE Status of National Flood Hazard Data Quality Standards as of March 31, 2008 C A N A D A uperio e S r ak L La ke n Hu La ke Michiga o r on tari e On Lak E rie ke La A t l a n t i c O c e a n P a c i f i c O c e a n Alaska RUSSIA M E X I C O CANADA Hawaii LEGEND Meets or Exceeds National Flood Kauai Hazard Data Quality Thresholds Niihau Molokai Oahu Approaches National Flood Maui G u l f o f Hazard Data Quality Thresholds Lanai Pacific Pacific Digital Product Issued M e x i c o Ocean Ocean Hawaii Modernized Map Not Yet Issued 0 500 1,000 Miles 0 100 200 Miles Study Not Planned 0 125 250 500 750 1,000 Miles Projection: North America Albers Equal-Area Conic FIGURE 2.5  Data quality standards achieved by individual counties as of March 31, 2008. Green counties (21 percent of the population) meet or exceed the floodplain boundary standard and the engineering analysis standard. Yellow counties (47 percent of the population) meet either the floodplain boundary standard or the engineering analysis standard or part of either standard but below thresholds. In red counties (1 percent of the population), the 2-5.eps been updated digitally and a digital product has been Figure maps have issued. Compliance with data quality standards was not required for such digital conversions, although a limited FEMA audit suggests that some portions of these counties meet the standards. In beige counties (26 percent of the population), modernized maps have not yet been issued because the first phase of map production (scoping) has not been completed or quality data do not exist. No study is planned in white counties (5 percent of the population). SOURCE: Paul Rooney, FEMA. The adjusted goal is to have 65 percent of the MAP MAINTENANCE U.S. continental land area and 92 percent of the U.S. population covered by digital flood maps (Table 2.2; A map records the conditions that existed when FEMA, 2006a). For 30 percent of the mapped stream the data for its compilation were gathered. By the and coastal miles covering 40 percent of the population, time the data are gathered and analyzed and the map the maps should meet the engineering analysis stan- is published, it may already be outdated. Corporate dard. For 75 percent of the mapped stream and coastal b ­ oundaries and other non-flood-related features can miles covering 80 percent of the population, the maps change, affecting regulation of floodplain development. should meet the floodplain boundary standard. These Ground elevations in the floodplain can change—for figures illustrate the challenges of increasing flood map example, when fill is placed in the floodplain to accuracy: even if the goals articulated in the midcourse raise building sites or when a new flood control adjustment are achieved, 70 percent of the mapped p ­ roject introduces levees, reservoirs, or stream chan- stream miles will not have validated engineering nel ­modifications—affecting the spread of floodwater. a ­ nalyses supporting the flood map, and 25 percent will Small projects, such as clearing channels or building not meet the floodplain boundary standard. In addi- retention basins in new sub­divisions, commonly do not tion, this standard ensures that the maps match existing have a measurable effect on the base flood and thus do topographic data within defined error tolerances, but it not warrant a map change on their own. Cumulative does not ensure the accuracy of the topographic data. effects of small projects, however, may be significant. 

FLOOD MAPPING AND FLOOD INSURANCE 21 TABLE 2.2  Adjusted Targets for FEMA’s Map Modernization Program Performance Measure Original Target (%) Adjusted Target (%) Percentage of continental U.S. land area covered by digital flood maps 100 65 Percentage of U.S. population covered by digital flood maps 100 92 Percentage of mapped stream and coastal miles with new, updated, or validated engineering analysis   22 30 Percentage of population covered by maps with new, updated, or validated engineering analysis   15 40 Percentage of mapped stream and coastal miles that meet the 2005 floodplain boundary standard   57 75 Percentage of population covered by maps that meet the 2005 floodplain boundary standard   32 80 SOURCE: FEMA (2006a). Finally, better topographic data, models, or statistical Amendments and revisions generally result in the data on hazard events may become available, potentially issuance of a Letter of Map Change (LOMC), and improving the depiction of the flood hazard. revisions may also result in a physical map revision. L ­ etters of Map Change originated when the produc- FEMA has four approaches to changing flood tion of FIRMs was an expensive photographic-based maps: process, and it was less expensive to issue a letter than to publish a new version of an affected map panel. 1. Restudy, in which a new Flood Insurance Study Applications for LOMCs are approved if computer is carried out to establish new flood profiles, data tables, models and ground surveys technically demonstrate and flood boundaries when development has substan- that the ground surface (and the lowest floor elevation, tially changed stormwater runoff conditions or when depending on the type of LOMC) is a tenth of a foot growth is occurring in a floodprone area that lacks base above the established BFE, even though current map- flood elevations. Restudies can be completely new work ping methodologies are not that accurate. Approved or new analysis of existing data using different models, LOMCs are used with the associated FIRMs for flood- and they result in addition of or adjustment to the BFEs, plain regulation and insurance purposes. addition of the 0.2 percent annual chance floodplain, Despite ongoing changes in the floodplain, FEMA and/or changes in the horizontal extent of the SFHA. flood maps are not updated on a regular ­ schedule. 2. Limited map maintenance projects, which are Requests for changes are made irregularly and ­physical restudies that are limited in size and cost. They are map revisions are infrequent due to funding con- frequently used to increase detail in approximate studies straints. Priorities must be set, and FEMA developed in unnumbered A zones. the Mapping Needs Assessment Process and the 3. Revisions, which are made after a flood map is Map Needs Update Support System (MNUSS) to published to reflect changes in the horizontal or vertical document and rank map update needs nationally. extent of the floodplain. Revisions may add or adjust However, even high-priority updates (e.g., areas with the BFE; add, remove, expand, or contract the mapped known unmapped flood hazards, communities that are floodplain; and/or add or remove a defined floodway. undergoing rapid growth or that can contribute to the 4. Amendments, which are made to correct map update) may not be made. Moreover, the time lag mapping inaccuracies, including non-flood-related between approving and publishing LOMCs and physi- map ­ elements (e.g., north arrows, graphic scale) and cal map revisions lengthened when FEMA directed i ­nadvertent inclusion of higher areas in the mapped funds from map maintenance to digital conversion of floodplain. Inadvertent inclusions are commonly found paper maps during the Map Modernization Program. through more accurate or detailed topographic study; As a result, some parcels and structures may not be when they are too small to depict graphically, they are regulated or insured properly, even though the change only correctable in Letters of Map Amendment. in risk is known. 

22 MAPPING THE ZONE FLOOD MAP INFORMATION TECHNOLOGY ization Program. Documenting how each mile was studied—including what input data, mapping, and In the early days of the NFIP, data were published modeling methods were used, the date of mapping, the and revised in the form of paper maps, Flood Insurance contractor, and the starting and ending points of each Study reports, and Letters of Map Change—a costly, study segment—would help users better understand inefficient, and time-consuming process. Initial steps the reliability and accuracy of the data. Many of these toward a less paper intensive process led to the creation metadata are not currently included in Flood Insur- of FEMA’s Map Service Center website in the late ance Study reports, particularly to this level of detail. 1990s and the development of new mapping products. However, metadata can easily be linked with digital Through this website, users can extract images from flood map information, enabling users to examine a full-sized paper map to create FIRMettes (e.g., data age, gathering, and analysis techniques to decide Figure 2.1) that are legally equivalent to the original whether the flood data are suitable for the intended use. paper product. The recent availability of LOMCs and This is especially important, given that FEMA flood Flood Insurance Study reports online has made data data are increasingly being used for land use planning, even more accessible. Yet although more products are emergency response, and risk assessment, in addition to available and distribution has improved, digital updat- the insurance and regulatory purposes for which they ing processes have lagged. were collected. FEMA created the Mapping Information Plat- form (MIP) on a secure website to allow its mapping Recommendation. FEMA should require that every partners (e.g., communities, engineers, surveyors, flood flood study be accompanied by detailed metadata control districts, Cooperating Technical Partners) to identifying how each stream and coastline reach submit data for review and share work responsibilities. was studied and what methods were used to identify With this system, map information (e.g., flood study the magnitude and extent of the flood hazard and to data, LOMCs) is being shared, rather than the maps produce the map. themselves. This system of information sharing shows what might be possible for map updates, which are often FLOOD DATA AND A NATIONAL slow to be integrated with other map information. HYDROLOGIC INFORMATION SYSTEM Recommendation. FEMA should ensure that new The FEMA Map Modernization Program is by flood information, revisions, and Letters of Map far the largest investment that the nation has made in Change are incorporated into the digital Flood Insur- hydrologic information in recent years. It is also the ance Rate Maps as soon as they become effective. largest effort that the nation has ever made to digitally describe the morphology of its streams and rivers. This The digital environment could also facilitate com- investment could have many benefits beyond flood munication of metadata—information about how mapping. The flood models could be used for flood flood data were generated. A variety of study methods management and planning studies or for building real- are often used along a stream reach or coastline. For time flood inundation mapping systems. The digital example, different segments of the same stream flow- terrain and stream channel information could be used ing through two adjacent communities may have been for water quality studies of contaminant transport in studied using different techniques and in different streams. FEMA is one of several federal agencies gen- years. This distinction was commonly lost when the erating spatial hydrologic information and it is reason- information was consolidated in the Map Modern- able to ask how the data and models compiled during the Map Modernization Program could be made part FEMA distributes more than 1 million paper maps each year, of a National Hydrologic Information System. and the average cost of producing maps for a typical county is Each of FEMA’s flood studies covers a geographic $250,000 to $500,000. Presentation to the committee by Paul region, often a county. Within that region, each stream Rooney, FEMA, on August 20, 2007. reach is considered a separate entity with its own flood See <https://hazards.fema.gov/femaportal/wps/portal>. 

FLOOD MAPPING AND FLOOD INSURANCE 23 discharge estimate, stream cross sections, and BFE. far along and transverse to the stream centerline the The floodplain boundaries of individual reaches are water flows. Unless both sets of coordinates are stored merged to delineate the Special Flood Hazard Area on in the archived map and model information, it will be a map panel. The digital information describing a single difficult or impossible at a later date to place a hydraulic flood study is stored in hundreds or even thousands of model cross section at the correct map location along files, which must be compiled for each county mapped the stream. in the nation. A key purpose of FEMA’s MIP is to One limitation of FEMA studies is that they are store these files so that they will be available for later done county by county and there is no requirement retrieval. Two types of files are involved: the files that that the underlying streamlines match across county comprise the flood map (DFIRMs) and files of raw boundaries. This difficulty can be overcome if FEMA field data analyzed in engineering studies to define the streamline data are matched with those of the U.S. BFE (Data Capture Standard database; FEMA, 2003, Geological Survey (USGS) National Hydrography Appendix L). Dataset (NHD). The NHD is a seamless, digital rep- Walker and Maidment (2006) examined the design resentation of streams and water bodies at map scales of a geodatabase model to store flood map informa- of 1:24,000 and 1:100,000 in the continental United tion. They showed that the most critical parts of the States. Walker and Maidment (2006) showed that for data capture standards are the stream centerlines and Fayette County, Texas, the 1:24,000 NHD streamlines cross sections used in the flood hydraulics model. If cover all the streams mapped in the Map Moderniza- accurate geographic information system (GIS) files of tion Program, and that each FEMA-mapped stream these are maintained along with the flood hydraulics segment could be located in a corresponding position model, the model could be georeferenced and used on the NHD. Thus, the flood study data collected in subsequent applications. This involves preserving by FEMA could be linked to and become a part of data defining the connection between two coordinate the nation’s larger repository of hydrologic informa- systems: the ­Cartesian (x, y, z) coordinate system used tion, enabling it to be used for much more than flood to record the meandering of the channel through the mapping. landscape and the (s, n, z) coordinate system used in the river hydraulics model, in which s represents sta- Recommendation. FEMA should reference all stream tioning distance along the river and n represents the and coastal studies within its Mapping Informa- distance across a particular cross section in the river. tion Platform to the USGS National Hydrography In effect, the hydraulic model “straightens” the chan- Dataset. nel by ignoring the bends and considering only how Presentation to the committee by Sally McConkey, Association of State Floodplain Managers, on November 8, 2007. See <http://nhd.usgs.gov/>.