Census 2000 TIGER/Line(R) files Technical Documentation Chapter 3: Attributes of Geographic Objects Line Features Line features consist of one or more complete chains that share common attributes such as feature identifiers, address ranges, and census feature class descriptions. Feature Identifiers The feature identification fields contain either a general type label or a specific proper name assigned to a complete chain that identifies the feature. Each complete chain that is a part of a named feature, such as US Highway 1, has the same feature identifier. The TIGER/Line files use several related data fields to provide a structured description of the feature identifier: Feature Direction Prefix (e.g., N Adams Ave) Feature Name (e.g., US Highway 1, Jefferson St) Feature Type (Roosevelt Blvd, Mangosteen River) Feature Direction Suffix (e.g., Providence St NE) Most named street/highway features have a feature type. Numerous exceptions exist; for example, Broadway consists of a feature name with no type specified. Do not confuse feature types that form proper names with the census feature classification scheme. In the Census TIGER(R) data base, feature names are assigned to line features independently of the census feature class codes (CFCCs) of the line features. For example, major airports usually have an express highway leading to the terminal area. This highway does not have an interstate highway name such as I-95, but may have the CFCC of an interstate highway (A11) because it has the same characteristics as an interstate highway (limited access with separated, multiple lanes). The feature identifiers of line features that are roads may include either a direction prefix or suffix. Some may have both a direction prefix and suffix. The feature name fields for line features that are roads may contain both a name and a feature type. For all hydrography and non-road features, the feature type will follow the feature name in the feature name field. In some instances, the feature type is commonly considered part of the name and is combined with the feature name in the TIGER/Line files to avoid confusion; for example, US Hwy 1. The Census TIGER System identifies US Hwy as a feature type used as a prefix to the name and 1 as the feature name. The feature types, such as US Highway, State Highway, and Interstate that normally precede the name appear in the name field. Generic feature identifiers have a name listed in the names field, but do not have a feature type or direction. Some examples of generic names include ramp, power line, and reservoir. Generic feature identifiers are selectively added to features that do not have proper names. In most cases, complete chains without proper names have no feature identifier. The TIGER/Line files do not support a data level above the complete chain that allows the construction of higher level objects (features). Complete chains with the same name may represent separate features; for example, a county may contain several Main Streets located in different geographic entities (e.g., towns or cities) scattered throughout the county. The ability to group chains together to include the entire length of a street feature, such as US Route 66, depends on the uniqueness of the identifiers and the consistency of the feature identifiers along the length of the feature. The U.S. Census Bureau makes no guarantee that the complete chains have uniform names or contain all of the known feature identifiers. The U.S. Census Bureau has taken steps to improve the consistency of feature identifiers and to add feature identifiers to fill in gaps along street features. The U.S. Census Bureau also has eliminated some alternate spellings in favor of the spelling confirmed by the ZIP+4 file of the U.S. Postal Service. The census feature class codes (CFCCs) may vary for chains with the same feature identifier. For example, the most frequent CFCC for a state highway is A21, but the complete chains marking the location of State Highway 32 may have a CFCC of A11, A21, or A31 (see the Census Feature Class Codes section in this chapter). The TIGER/Line file structure allows up to 4,996 feature identifiers for a complete chain. The primary feature identifier appears in Record Type 1. For street features, the primary feature identifier is usually the name most commonly associated with the address range. Up to five alternate feature identifiers are cross-referenced in each Type 4 record, and a single complete chain can have up to 999 Type 4 records. Alternate feature identifiers include highway designation numbers for named streets, former names, and alternate spellings where source material provided conflicting data. Where the complete chain represents a limited access highway, the highway type and route designator, such as I-95, should ideally become the primary name, and the local designation, such as Cross County Expressway or Capital Beltway, should become the alternate name. However, this is not always true in the TIGER/Line files. The primary and alternate feature identifiers can be independent of each other. There is no assurance that the same combination of primary and alternate feature identifiers will appear together in a sequence of complete chains. There also is no assurance that a feature identifier will consistently appear as the primary identifier; it might be recorded as an alternate feature identifier for some complete chains and a primary feature identifier for others. During TIGER improvement operations, the U.S. Census Bureau has taken steps to make the Interstate highway route designator the primary feature identifier for Interstate highways, and the common street name used in mail delivery the primary name on all other roads. The order of identifiers follows this hierarchy: Interstate highway, common name, US highway, state highway, county highway, with town and township road at the bottom of the list. Record Type 5 contains a record for each feature identifier used as either a primary or an alternate name. The TIGER/Line files link the alternate names in Record Type 5 to Record Type 1 through the use of the alternate feature identification code index that forms Record Type 4. See the Feature Identifier Record Linkage section in this chapter. Feature Identifier Record Locations Record Type Field Name Description 1 FEDIRP Feature Direction, Prefix 1 FENAME Feature Name 1 FETYPE Feature Type 1 FEDIRS Feature Direction, Suffix 5 FEDIRP Feature Direction, Prefix 5 FENAME Feature Name 5 FETYPE Feature Type 5 FEDIRS Feature Direction, Suffix Feature Identifier Codes Direction (Prefix and Suffix) Direction consists of a 2-character abbreviation, left-justified in the data fields, and is used for road features only. Abbreviation Explanation (blank) No Direction N North, Norte S South, Sur E East, Este W West, Oeste NE Northeast, Norte Este, Nordeste NW Northwest, Norte Oeste, Noroeste SE Southeast, Sur Este, Sudeste SW Southwest, Sur Oeste, Sudoeste NO Norte Oeste, Northwest SO Sur Oeste, Southwest O Oeste, West EX Extended, Extension Feature Names Feature names consist of a 30-character text string with words separated by blanks. Feature names contain upper- and lower-case characters. The feature name is truncated if it is over 30 characters long. The U.S. Census Bureau is no longer using codes to represent the diacritical marks. Beginning with the 1999 TIGER/Line files, the U.S. Census Bureau is using the ISO 8859-1 character set, commonly referred to as Latin-1, to identify characters with diacritical marks. ISO 8859-1 is not ASCII or "extended ASCII," but rather ASCII compatible in that the first 127 character codes of ISO 8859-1 are identical to ASCII. ISO 8859-1 uses the space left vacant by ASCII in the 8-bit range to represent additional characters. The following 16 characters from the ISO 8859-1 may appear in the Census 2000 TIGER/Line files: Character Name ISO (dec, hex) A A-Acute Accent 193,c1 a a-Acute Accent 225,e1 E E-Acute Accent 201,c9 e e-Acute Accent 233,e9 I I-Acute Accent 205,cd i i-Acute Accent 237,ed N N-Tilde 209,d1 n n-Tilde 241,f1 O O-Acute Accent 211,d3 o o-Acute Accent 243,f3 U U-Acute Accent 218,da u u-Acute Accent 250,fa U U-Diaresis 220,dc u u-Diaresis 252,fc A A Ring 197,c5 a a Ring 229,e5 In the 1998 and earlier TIGER/Line files the U.S. Census Bureau used the following codes to represent diacritical marks: ] Preceding character has an acute accent [ Preceding character has a dieresis # Preceding character has a tilde The feature name field may contain abbreviations to represent some feature types. See Appendix D-Standard Abbreviations. Feature Types The feature type field for road features consists of a 4-character text string. For all hydrography and non-road features, the feature type will follow the feature name in the feature name field. The abbreviations in Appendix D-Standard Abbreviations may appear in the feature type field or the feature name field. Corporate Corridors and Corporate Offset Boundaries A corporate corridor is a narrow, linear part of an incorporated place (or in a few instances, another legal entity). The corporate corridor includes the street and/or right-of-way, or a portion of the street and/or right-of-way within the incorporated place. It excludes from the incorporated place those structures such as houses, apartments, or businesses that front along the street or road. A corporate limit offset boundary exists where the incorporated place lies on one side of the street and may include all or part of the street or right-of- way, but excludes from the incorporated place, the structures located along that side of the street.. To facilitate address coding, the Census TIGER data base contains duplicate street name and address ranges on complete chains with a CFCC of F11 (nonvisible offset boundary of a legal entity) or F12 (nonvisible corridor boundary of a legal entity). The duplicate street names for the F11 and F12 features are on Record Type 5; the duplicate address ranges are on Record Type 6. Record Type 1 will not contain feature identifiers for complete chains with CFCCs of F11 or F12. Feature Identifier Record Linkage Record Type 4 provides the link required to find any alternate feature identifiers belonging to a complete chain. Record Type 4 cross-references each TLID with an Alternate Feature ID code (FEAT) assigned to each record in Record Type 5. Record Type 5 contains all feature identifiers including those that are used only as primary identifiers. However, only the FEATs for complete chains that have alternate feature identifiers appear in Record Type 4. Complete chains that have no alternate feature identifier will have no Type 4 record. To find the alternate feature identifiers for a complete chain, begin by determining the TLID for the complete chain. Then search for this TLID in Record Type 4. If the complete chain has any alternate feature identifiers, Record Type 4 should provide at least one record. Once found, the Record Type 4 entries will each contain from one to five FEAT numbers. The FEAT fields are blank when no further alternative identifiers exist. The first FEAT field (FEAT1) should always have a valid FEAT number. Finally, find the records in the Record Type 5 file that match the FEAT codes from Record Type 4. The TIGER/Line file provides a record sequence number to identify multiple Type 4 records that might exist for one TLID. Even though Record Type 5 contains all feature identifiers, Record Type 4 contains only references for alternate feature identifiers. Data users cannot link all of the names in Record Type 5 to all of the associated complete chains in Record Type 1 by using Record Type 4. Feature Identification Numbers Record Locations Record Type Field Name Description 1 TLID TIGER/Line ID, Permanent Record Number 4 TLID TIGER/Line ID, Permanent Record Number 4 RTSQ Record Sequence Number 4 FEAT1 Line Additional Name Identification Number, First 4 FEAT2 Line Additional Name Identification Number, Second 4 FEAT3 Line Additional Name Identification Number, Third 4 FEAT4 Line Additional Name Identification Number, Fourth 4 FEAT5 Line Additional Name Identification Number, Fifth 5 FEAT Line Name Identification Number 9 FEAT Line Name Identification Number Feature Identification Code The FEAT and sequenced FEAT data fields contain an 8-digit integer number (without leading zeros). A FEAT is assigned sequentially, beginning with 1, to each feature identifier in Record Type 5. The FEAT is not a permanent identification number. TLID is the record identifier for the complete chain. See Chapter 2 for a full discussion of TLIDs. RTSQ is a 3-digit integer that uniquely identifies multiple Type 4 records with the same TLID. RTSQ equals 1 for the first occurrence of a TLID in Record Type 4 and can reach a maximum of 999 for subsequent occurrences. Address Ranges and ZIP Codes The TIGER/Line files contain address ranges, not individual addresses. The term address range refers to the first possible structure number and the last possible structure number along a complete chain side relative to the direction in which the complete chain is coded. The address ranges in the TIGER/Line files are potential ranges that include the full range of possible structure numbers even though the actual structures might not exist. The address numbers used to create the address ranges are commonly known as house number-street name style addresses. A house number-street name style address minimally consists of a structure number, street name, and a 5- digit ZIP Code; for example, 213 Main St 90210. In the TIGER/Line files, the ZIP Codes usually appear only on those complete chains that have address ranges identified. However, they may appear on some road features without the address ranges. An address range also may have the full 9-digit ZIP Code that includes the USPS's 4-digit ZIP+4 Add-On code. The U.S. Census Bureau has added the Postal Add-On code to the Census TIGER data base using an automated match to the USPS's ZIP+4 file. The codes in the TIGER/Line files are the street-level codes the USPS has assigned to address ranges. The USPS may assign more specific codes to companies and buildings, and to apartments, floors, or suites within buildings. Some address coding software that uses the USPS's ZIP+4 file may provide the more specific codes. However, the TIGER/Line files contain only the more general codes. Usually the ZIP+4 Add-On code is not required to uniquely identify an address range. There are a few situations where a street name and address range legitimately appear more than once in the same 5-digit ZIP Code. Usually the USPS distinguishes these duplicates by using different postal station names. However, the Postal Add-On code will uniquely identify these cases. Puerto Rico is a special case because many addresses were uniquely assigned within an urbanizacion (a community or development) and could duplicate another address in a different urbanizacion with the same 5- digit ZIP Code. To resolve this problem, the USPS added an additional line to the address to identify the urbanizacion. The 9-digit ZIP Code also may serve to uniquely identify these address ranges. We do not yet have all of these 9-digit ZIP Codes in the Census TIGER data base. Address Ranges Complete chains in the TIGER/Line files have one end point labeled as the start node and the other end point labeled as the end node. The start and end nodes also are referred to as from and to. The start node always corresponds to the beginning of the complete chain identified by the start node coordinates FRLAT and FRLONG. The order of the addresses follows the sequence of the nodes on the complete chain; the nodes may not be related to the low to high orientation of the address range. The start address may be higher or lower than the end address for a complete chain. Structure numbers usually, but not always, systematically increase or decrease while moving along a street in a set direction from one complete chain to the next. Record Type 1 contains the initial address ranges for the left and the right sides of a complete chain. A complete chain side may have multiple address ranges. The TIGER/Line files use Record Type 6 to store any additional ranges as required. The Type 1 record will hold the ranges with the largest sequence of numbers. However, Record Type 6 may hold a significant number of additional ranges. Data users must use Record Type 6 to obtain the entire picture of the possible address ranges along a complete chain. In Record Types 1 and 6, both the left- and the right-side address ranges have a start and an end address range field that can contain a maximum of 11-alphanumeric characters. The address range fields are right-justified. Each address range in the TIGER/Line files has only one parity. Only odd- numbered addresses are contained within an address range with odd start and end structure numbers. Likewise, only even-numbered addresses belong to an address range with even start and end structure numbers. The value zero is not used as a valid address range end value. Generally, the left and the right sides of a complete chain have opposite parities. If both odd and even addresses exist on the same side of a complete chain, the TIGER/Line files provide both an even and an odd parity range for that side of the complete chain. One of the ranges appears in Record Type 1, while the other range appears separately in Record Type 6. Some basic characteristics of address ranges are as follows: The TIGER/Line files generally contain only those house number-street name style address ranges used for mail delivery. They do not show rural route and post office box addresses. They may contain structure numbers assigned in select areas for use by local emergency services, but not for mail delivery. The TIGER/Line files do include address ranges and ZIP Codes in some small places where the USPS provides only post office box service, not street delivery. These address ranges represent the structure numbers collected during the 2000 census field operations, supplemented with addresses provided through local participant programs. Where these address ranges exist, they may be used to geocode a structure to the census block. These structure- number addresses may have ZIP Codes associated only with post office box addresses. The ZIP Codes represent the post office boxes. The address ranges in these areas do not have Postal Add-On codes since the USPS does not use them for street delivery. Gaps may exist between multiple ranges for a single complete chain. A gap may be significant, since any numbers missing from one complete chain may actually appear on another complete chain in the case of address anomalies such as out-of-parity or out-of-sequence addresses. Beginning with the 1999 TIGER/Line files, the U.S. Census Bureau will not include any single address-address ranges in the TIGER/Line files including out-of-parity and out-of sequence addresses. That is, when there is a single address that is "out of place" geographically (for example, across the street from all other odd addresses or three blocks away from all other 1200-series addresses), the U.S. Census Bureau will exclude that single address from any address range. The U.S. Census Bureau created many new address ranges using addresses from the Census 2000 official census address list. Suppression of single address- address ranges is to protect the confidentiality of individual addresses collected through Census 2000 census field operations as specified by Title 13 of the U.S. Code. In a few rare cases, address ranges can include numbers with alphabetic characters. These characters help uniquely identify addresses within acounty. For instance, certain unincorporated areas of Genesee County, Michigan add a letter G prefix to the address number. The characters are consistently placed within the address range field; for example, the letter G maintains a consistent column placement in the range G1 to G99. Address ranges exist only for street features, and in some cases, corporate corridor and corporate offset boundary features. Address ranges (consisting of a unique combination of structure number, ZIP Code, feature name, feature type, and directional) should not overlap; addresses should belong to only one range. The U.S. Census Bureau edits the address ranges to locate possible overlaps, but cannot guarantee that all possible overlap situations have been identified. Address ranges in the TIGER/Line files are usually associated with both the primary and alternate feature identifiers. Caution: Address range overlaps may occur if primary address ranges are linked to alternate feature identifiers that identify route numbers. Some address systems use a hyphen to separate avenue numbers, private road designators, and grid cell numbers from the structure numbers; for example, 10-01 Reynolds St uses a hyphen to separate the avenue number from the structure number. Imputed Address Ranges Imputed address ranges occur during the process of updating the Census TIGER data base when a new complete chain intersects an existing complete chain with address ranges. The intersection splits the existing complete chain and produces two new complete chains connected by a new node located at the intersection point. The update program divides the old address ranges among the two new complete chains and imputes the address range ends at the new node. The impute process allocates either all or part of each original address range to each of the new complete chains in proportion to their lengths. For each side of the original complete chain, the process considers all address ranges appearing on each side and determines the overall low and high address. The process assumes the addresses are evenly distributed over the length of the complete chain, and applies the proportion of complete chain lengths to the overall address ranges to calculate a split point address for each side. Address ranges that fall entirely above or below the split point address are moved intact to one of the new complete chains. The process divides any address ranges that contain the split point address and allocates each part to one of the new complete chains. The new address range ends created from the split are imputed values and have an impute flag. Some intermediate address range ends also may carry the impute flag. These address range ends fall between the overall high and low address for complete chain sides that have more than one address range. The impute flags on these range ends often mark splits created by adding different nine-digit ZIP Codes to parts of the original address range. These impute flags are not significant and should be disregarded. The impute flags identify address ranges that have been through the impute process. Each record in the TIGER/Line files contains four separate 1-character impute flag fields, one for each address range end. ZIP Codes The ZIP Code is an attribute of the address ranges. The TIGER/Line files have a five-character ZIP Code field containing a numeric code with leading zeros. Both the left- and right-side address ranges share the ZIP Code that appears in the same Type 1 or Type 6 record. Each address range belonging to a complete chain can have a different ZIP Code. Where ZIP Code boundaries follow a street, the complete chain may have different left- and right-side ZIP Codes, or different ZIP Codes along its length. Because the Census TIGER data base identifies only one ZIP Code for each address range record, address ranges with different ZIP Codes must appear in separate records. The address range(s) with one ZIP Code will appear in Record Type 1, and the address range(s) with the other ZIP Code(s) will appear in Record Type 6. For example, one complete chain making up Duke Street is a ZIP Code boundary; the left-side range 1-99 has a ZIP Code of 12345, and the right-side range 2-98 has a ZIP Code of 54321. The range 1-99 with a ZIP Code of 12345 will appear in Record Type 1, and the right-side range fields will be blank. The range 2-98 with a ZIP Code of 54321 will appear in Record Type 6, and the left-side range fields will be blank. If the complete chain had additional address ranges with a ZIP Code of either 12345 or 54321, these additional address ranges would appear withone of the existing ranges or as additional Type 6 records. For example, a right- side range of 150-198 with a ZIP Code of 12345 could appear on the Type 1 record with the left-side range of 1-99. However, a right-side range of 150- 198 with a ZIP Code of 54321 could not appear on the Type 6 record with the range 2-98. Instead, the range would have to appear in a second Type 6 record. Since the ZIP Codes in the TIGER/Line file relate to mail delivery along addressed streets, they are not true area features. It is possible that a polygon may contain addresses associated with more than one delivery ZIP Code. Postal Add-On Code The TIGER/Line files have a 4-character Postal ZIP+4 Add-On code which is located on Record Type Z. Record Type Z may link to a left- or right-side address range in Record Type 1 or in Record Type 6. By using the TLID fields, data users can match the Postal +4 Add-On codes on Record Type Z to an address range in either Record Type 1 or Record Type 6. If the Record Sequence Number (RTSQ) field on Record Type 6 contains a 0, the Postal +4 Add-On codes apply to the address ranges in Record Type 1. If the RTSQ field contains a number greater than 0, the Postal +4 Add-On codes apply to the address ranges in the Record Type 6 that have the identical RTSQ value. The first two characters of the Postal +4 Add-On code indicate the USPS sector code; the last two characters represent the USPS segment code. As stated earlier, the U.S. Census Bureau used an automated match process to assign the Add-On codes to the address ranges in the Census TIGER data base. The match utilized only the street type records from the ZIPMOVE and ZIP+4 files. These records identify a single Add-On code for a range of addresses. The ZIP+4 file also contains company and high-rise building records that supply specific codes to companies, buildings, and floors or suites within buildings. The U.S. Census Bureau did not match these codes to the Census TIGER data base because it was not practical to add all of the building features to the Census TIGER data base. Also, it was not feasible to split the address ranges for individual building-level codes. The match process attempted to relate the 5-digit ZIP Code, street name identifier, and address ranges for each feature in the Census TIGER data base to the corresponding street type record in the USPS ZIPMOVE file, which identifies all 5-digit ZIP Code changes for the previous five years. If an address range (or portion thereof) in the Census TIGER data base matched a range in the ZIPMOVE file, the U.S. Census Bureau then compared the range to the USPS ZIP+4 file. If the address range matched the ZIP+4 file, then the ZIP Code for that address range was updated in the Census TIGER data base. Where successful, the process added the Postal +4 Add-On codes to the address ranges in the Census TIGER data base. Beginning with the 1999 TIGER/Line files there will be multiple Postal ZIP+4 Add-On codes associated with a single address range. The reason for this is that the U.S. Census Bureau no longer is including any single address-address ranges in the TIGER/Line files. Suppression of single address-address ranges is to protect the confidentiality of individual addresses as specified by Title 13 of the U.S. Code. To avoid creating single address-address ranges the U.S. Census Bureau no longer will split address ranges where a Postal +4 Add- On code covers only part of the address range. Rather, the TIGER/Line files will include multiple Postal +4 Add-On codes for an address range. The Postal +4 Add-On codes may appear on more than one complete chain. This results because the potential address ranges used by the U.S. Census Bureau differ from those used by the USPS, and because the U.S. Census Bureau recognizes complete chain breaks and intersections not recognized by the USPS. Address Information and Key Geographic Locations (KGLs) KGLs represent a special class of address information. They provide a geocoding tool like address ranges, but also identify a spatial object similar to a landmark. The U.S. Census Bureau uses KGLs to identify named buildings where the use of the feature name enhances the ability to geocode addresses. These cases include airports, shopping centers, schools, condominiums, hotels, and apartment complexes. In the TIGER/Line files, each KGL usually has a CFCC and KGL feature name. The street feature identifier associated with the KGL is obtained by linking the FEAT field to Record Type 5 which contains the list of all street name identifiers. The KGLs are linked by the CENID and POLYID to the GT-polygons. To locate the KGL, link the CENID and POLYID on Record Type 9 to the CENID and POLYID identifiers on Record Types A or S. Even though the KGLs appear to identify specific structures, the KGL descriptions do not include location coordinates or address information. Address Information Methodology Pre-Census 2000 Address Ranges Before the 1990 census, the Census TIGER data base contained address ranges only for the area covered by 1980 Geographic Base File/Dual Independent Map Encoding (GBF/DIME) files and a few file extension areas prepared in conjunction with 1980 census activities. These ranges were used to geocode a list of addresses to geographic areas for use in the 1990 questionnaire mail-out. For the 1990 census, the U.S. Census Bureau purchased the list of addresses from commercial vendors for the geographic areas where the Census TIGER data base included address ranges. To verify the accuracy of the addresses, the U.S. Census Bureau began with an initial assignment of residential addresses to the 1990 census tracts and blocks. Clerical review of the results of the assignment process provided additional address range updates. In the early 1990s, the U.S. Census Bureau expanded its address range coverage to include the entire United States by creating new ranges based on the Address Control File (ACF) used in the 1990 decennial census. The ACF was a master list of addresses geocoded to the census block level. For each block, the individual structure addresses were grouped by feature identifier and sorted into numerical order to extract an actual range. To maintain confidentiality of individual addresses, the U.S. Census Bureau converted the actual range to a potential range. This was accomplished by expanding the actual range to complete a hundred range, splitting the difference between coverage gaps, and in some cases disguising the range by the random addition or subtraction of addresses. In addition to merging the addresses in the Census TIGER data base and the ACF, the U.S. Census Bureau edited address ranges for overlaps or other inconsistencies. Orientation edits attempted to standardize the low to high orientation of address ranges along a chain of street feature complete chains with the same feature identifier. Parity edits attempted to place the even- and odd-parity ranges consistently on the same side of a feature chain. Complete chains with address ranges that were specifically identified as orientation or parity anomalies were automatically excluded from these edits. The U.S. Census Bureau conducted a general ZIP Code clean-up and staff added new ZIP Codes created since the 1990 census. Street names and address ranges in the Census TIGER data base were compared to those in the ZIP+4 file of the U.S. Postal Service. If a street name and address range did not have a ZIP+4 code, the code was copied from the ZIP+4 file to the Census TIGER data base. The consistency of highway names and feature identifiers also was improved. Census 2000 Address Ranges For Census 2000, the Master Address File (MAF) replaced the ACF of the 1990 census. The MAF is a list of all living quarters nationwide along with their geographic locations. The U.S. Census Bureau originally created the MAF by combining the addresses in the 1990 ACF with the U.S. Postal Service Delivery Sequence File. The MAF is maintained through partnerships with the U.S. Postal Service (USPS), with Federal, State, regional, and local agencies, and with the private sector. U.S. Census Bureau staff updated and supplemented the MAF with address information provided by census programs such as the TIGER Improvement Program (TIP) and the Local Update of Census Addresses (LUCA) in which local and tribal governments provided address updates as well as through Census 2000 field operations. As part of the TIGER Improvement Program (TIP) local governments were provided address range "clusters" from the USPS ZIP+4 file that failed to geocode to the Census TIGER data base. Using local sources and expertise, participants annotated maps derived from the TIGER data base to correct errors and add missing streets, street names, address ranges, and/or ZIP Codes. U.S. Census Bureau staff then incorporated participant updates and corrections into the TIGER data base, thus enabling the address clusters to geocode. In areas not participating in TIP, U.S. Census Bureau staff researched the clusters and made corrections. The U.S. Census Bureau periodically receives updated information from the USPS which it matches against the MAF. In situations where addresses fail to geocode to the TIGER data base, U.S. Census Bureau geographic staff research the addresses and make the necessary updates and corrections to enable the addresses to geocode. In late 1999, the U.S. Census Bureau initiated a process to compare the addresses in the MAF to existing address ranges in the Census TIGER data base and to create or modify the TIGER address ranges where necessary. This automated program matched field verified MAF address/collection block relationships to address ranges on either primary or alternative feature names in the Census TIGER data base. The program eliminated potential address ranges in the Census TIGER data base that conflicted with the address/collection block number relationships from the MAF, and built potential ranges around the new MAF-based actual address ranges. When discrepancies occurred between the MAF and Census TIGER, the MAF was deemed to be more accurate because of address information obtained through local partnership programs. This automated match shifted address range ends along complete chains, flipped address ranges from one side of a complete chain to the other to correct parity reversals, and expanded potential ranges for each complete chain. In cases where MAF-verified addresses resulted in orientation or parity reversals along a complete chain, or out-of-sequence addresses, the address ranges were accepted as verified exceptions and were not adjusted. The address match also combined the actual MAF and potential Census TIGER address ranges into the largest possible potential range(s) for each complete chain side. It retained high and low address range ends and discarded intermediate address range breaks at the end of the process. This closed coverage gaps, and provided full potential addresses ranges in Census TIGER. This was done to facilitate geocoding new or commercial addresses. No single address-address ranges appear in the 1999 or later versions of the TIGER/Line files including out-of-parity and out-of-sequence addresses. Many new address ranges were created through the automated address range match using addresses from the official Census 2000 census address list. Suppression of single address-address ranges is to protect the confidentiality of individual addresses collected through census field operations as specified by Title 13 of the U.S. Code. As a result, any single address that is "out of place" geographically (that is, across the street from all other even addresses or several blocks away from all other addresses in that address series) will not appear in any address range in the TIGER/Line files. For example, address 709 Main Street is in the middle of the even-side of the 700 block of Main Street and will be suppressed because it is a single address-address range. The following addresses ranges for the 700 block of Main Street will appear in the TIGER/Line files: 700-798 Main Street, 701-707 Main Street, and 711-799 Main Street. Based on the information provided data users cannot tell where 709 Main Street is located. Both primary and alternate feature identifiers can be used in geocoding, but great care should be used with the alternate identifiers. In the case of corporate corridors and corporate limit offset boundaries, the alternate address linked to the boundary should be used for geocoding rather than the primary range linked to the street (see the Corporate Corridors and Corporate Limit Offset Boundaries section in this chapter). Address Range Record Locations Record Type Field Name Description 1 FRADDL Start Address, Left 1 TOADDL End Address, Left 1 FRADDR Start Address, Right 1 TOADDR End Address, Right 6 FRADDL Start Address, Left 6 TOADDL End Address, Left 6 FRADDR Start Address, Right 6 TOADDR End Address, Right 9 KGLADD Key Geographic Location Address Impute Flag Record Locations Record Type Field Name Description 1 FRIADDL Start Imputed Address Flag, Left 1 TOIADDL End Imputed Address Flag, Left 1 FRIADDR Start Imputed Address Flag, Right 1 TOIADDR End Imputed Address Flag, Right 6 FRIADDL Start Imputed Address Flag, Left 6 TOIADDL End Imputed Address Flag, Left 6 FRIADDR Start Imputed Address Flag, Right 6 TOIADDR End Imputed Address Flag, Right ZIP Code Record Locations Record Type Field Name Description 1 ZIPL ZIP Code, Left 1 ZIPR ZIP Code, Right 6 ZIPL ZIP Code, Left 6 ZIPR ZIP Code, Right Z ZIP4L +4 Postal Add-On Code, Left Z ZIP4R +4 Postal Add-On Code, Right 9 KGLZIP Key Geographic Location ZIP Code 9 KGLZIP4 +4 Postal Add-On Code for KGL Address Ranges and Impute Flag Codes Address Ranges Numeric characters or a mixture of numeric and alphabetic characters (maximum of 11 characters). Address range fields are blank when no address range is available. Both the start and end address range fields are blank, or both have non-zero values. The KGLADD field on Record Type 9 is blank in this version of the TIGER/Line files. Impute Flags (1-character numeric code) blank- No address range available 0- Not imputed 1- Imputed ZIP Codes See the U.S. Postal Service (USPS) Publication 65, National Five-Digit ZIP Code and Post Office Directory for a list of valid 5-digit ZIP Codes. The Census 2000 TIGER/Line files may not contain all delivery ZIP Codes and may contain some non-delivery ZIP Codes. The distribution of ZIP Codes in the TIGER/Line files may not reflect the exact USPS ZIP Code service area. Limitations Users of the address ranges in the TIGER/Line files should check for address range overlaps, gaps, odd/even reversals, and other situations that may be incorrect. While the U.S. Census Bureau continues to edit for, and correct these situations, it is possible that some still exist. Corporate Corridors and Corporate Limit Offset Boundaries A corporate corridor is a narrow, linear part of an incorporated place (or in a few instances, another legal entity). The corporate corridor includes the street and/or right-of-way, or a portion of the street and/or right-of-way within the incorporated place. It excludes from the incorporated place those structures such as houses, apartments, or businesses that front along the street or road. A corporate limit offset boundary exists where the incorporated place lies on one side of the street and may include all or part of the street and/or right-of-way, but not the structures located on that side of the street. See the Places section in Chapter 4. To facilitate the coding of addresses to the correct geographic entity, the Census TIGER data base contains duplicate street name and address ranges on complete chains with a CFCC of F11 (nonvisible offset boundary of legal entity) or F12 (nonvisible corporate corridor of legal entity). The duplicate street names for the F11 and F12 features are on Record Type 5; the duplicate address ranges are on Record Type 6. Complete chains with CFCCs of F11 or F12 will not contain the duplicate names or address ranges in Record Type 1. Record Type 1 does not indicate that the street or right- of-way lies within a corporate corridor or offset boundary. Therefore, the address ranges lie outside the corporate corridor or offset boundary and are encoded on either side of these lines. Data users planning to geocode addresses in areas with these boundary types must identify the duplicate feature identifiers and ranges from Record Types 5 and 6 (the names and address ranges for CFCC F11 and F12 features), locate the street feature with those ranges, and remove the street feature's address ranges and geographic codes from the geocoding process. Record Linkages The TIGER/Line files store address range information in two record types. Record Type 1 contains the basic complete chain attributes, including one basic address range. Record Type 6 stores the additional ranges when the complete chain has more than one range on one or both sides. The TLID field links Record Types 1 and 6. Since a complete chain can have more than one set of address ranges, multiple Type 6 records can exist with the same TLID. The TIGER/Line files distinguish these records with a record sequence number (RTSQ). All Type 6 records that have the same TLID appear sequentially in the file even though the records are not sorted by TLID. The TIGER/Line files do not contain a field indicating whether a Type 6 record exists for a specific TLID; the user must scan any existing records in Record Type 6 for a TLID match. Boundaries of Geographic Entities The TIGER/Line files store geographic codes as either a polygon or complete chain attribute. In the case of state and county level geography, and some other areas, the codes appear in both complete chain and polygon record types. Refer to Chapter 4 for descriptions of geographic areas, and to Chapter 6 for the data dictionary that describes the record type fields. Record Linkages and Boundary Extraction The codes assigned to the complete chain belong to the areas referenced by the left and the right sides of a complete chain. Only those features that have different geographic codes on the left and the right sides of a line become boundary features. Information from multiple TIGER/Line data fields is required to uniquely identify the boundary of some geographic entities. For instance, both the census block and census tract codes are required to identify a block boundary. Block 1011 in census tract 2101 could neighbor block 1011 in census tract 2998. Be sure to use both the basic number and the suffix when extracting either Census 2000 census tract or block boundaries. Data users who have combined TIGER/Line files should include the Census 2000 state or statistically equivalent entity and Census 2000 county or statistically equivalent entity codes to extract Census 2000 census tract boundaries. The extraction of boundary features from polygon attribute codes requires making a link between the polygon and the complete chain data records, then identifying the features with different left- and right-side geographic codes. For a description of the record linkage process, see the Polygon Features section in this chapter. Boundary rings consist of multiple complete chains that are sequentially linked together and connected to form a closed ring. The process of linking all of the boundary complete chains that outline the same geographic entity requires the extraction of all complete chains that have that entity's code on either the left or the right side (but not both). Linking the chains together will form a polygon; each polygon may represent one of the GT-polygons described in Record Types A, P, and S, or a collection of these GT-polygons. Caution: Some types of geographic areas must end at a county/file boundary while others can continue into adjoining counties/files. For example, MCDs stop at a county boundary, whereas incorporated places can exist in several counties (See the Record Linkages/Feature Chaining section in this chapter). Single-Side Flags and County Boundaries The Census 2000 TIGER/Line files use the January 1, 2000 counties or statistically equivalent entities as the basis for the file coverage area. County boundary features are duplicated between adjoining pairs of counties so that each file is complete. However, the complete chains that constitute the boundary features contain only the geographic entity codes and address ranges relevant to each county-based TIGER/Line file. The geographic entity codes are blanked out on the outside edge of the county, even though some of these fields must normally have a non-blank code. The TIGER/Line file identifies these complete chains with a 1-character, single-side segment flag in the SIDE1 field of Record Type 1. When combining several TIGER/Line files to form a state or regional data set, the data user will need to eliminate duplicate boundary lines. Because each one of the duplicate boundary complete chains has either the left- or right-side geographic entity codes and address ranges, the elimination process will need to combine the codes and address ranges from both lines. The same situation applies to the polygon identification codes. Record Type I contains CENIDs and POLYIDs for GT-polygons within the county. If the GT-polygon is in the adjacent county, the CENID and POLYID fields are blank. Single-Side Flag Record Location Record Type Field Name Description 1 SIDE1 Single-Side Complete Chain Code Single-Side Flag Codes 1- The complete chain is a county boundary; either the left or the right side is blank blank- The complete chain is not a county boundary; neither left nor right side is blank Census Feature Class Codes (CFCCs) A census feature class code (CFCC) is used to identify the most noticeable characteristic of a feature. The CFCC is applied only once to a chain or landmark with preference given to classifications that cover features that are visible to an observer and a part of the ground transportation network. Thus, a road that also is the boundary of a town would have a CFCC describing its road characteristics, not its boundary characteristics. The CFCC, as used in the TIGER/Line files, is a three-character code. The first character is a letter describing the feature class; the second character is a number describing the major category; and the third character is a number describing the minor category. Some street features in the Census 2000 TIGER/Line files that normally would be classified as "A" class features may now be coded with a "P' instead of the "A" to indicate that the feature is a "provisional" feature. The numeric portion of the CFCC still classifies the street as if an "A" were preceding it. Provisional features are those streets that were added from reference sources or other programs in preparation for Census 2000, but were not field verified by census staff during field operations or through the use of aerial photography or imagery. As these features are verified in future operations the provisional flag will be removed for subsequent TIGER/Line file releases. Features that still have the provisional flag at the time the U.S. Census Bureau assigned the Census 2000 tabulation block numbers were not held as Census 2000 tabulation block boundaries. Feature Class A, Road The U.S. Census Bureau uses the term divided to refer to a road with opposing traffic lanes separated by any size median, and separated to refer to lanes that are represented in the Census TIGER data base as two distinct complete chains. The term, rail line in center, indicates that a rail line shares the road right-of- way. The rail line may follow the center of the road or be directly next to the road; representation is dependent upon the available source used during the update. The rail line can represent a railroad, a streetcar line, or other carline. Primary Highway With Limited Access Interstate highways and some toll highways are in this category (A1) and are distinguished by the presence of interchanges. These highways are accessed by way of ramps and have multiple lanes of traffic. The opposing traffic lanes are divided by a median strip. The TIGER/Line files may depict these opposing traffic lanes as two distinct lines in which case, the road is called separated. CFCC Description A11 Primary road with limited access or interstate highway, unseparated A12 Primary road with limited access or interstate highway, unseparated, in tunnel A13 Primary road with limited access or interstate highway, unseparated,underpassing A14 Primary road with limited access or interstate highway, unseparated, with rail line in center A15 Primary road with limited access or interstate highway, separated A16 Primary road with limited access or interstate highway, separated, in tunnel A17 Primary road with limited access or interstate highway, separated, underpassing A18 Primary road with limited access or interstate highway, separated, with rail line in center Primary Road Without Limited Access This category (A2) includes nationally and regionally important highways that do not have limited access as required by category A1. It consists mainly of US highways, but may include some state highways and county highways that connect cities and larger towns. A road in this category must be hard-surface (concrete or asphalt). It has intersections with other roads, may be divided or undivided, and have multi-lane or single-lane characteristics. CFCC Description A21 Primary road without limited access, US highways, unseparated A22 Primary road without limited access, US highways, unseparated, in tunnel A23 Primary road without limited access, US highways, unseparated, underpassing A24 Primary road without limited access, US highways, unseparated, with rail line in center A25 Primary road without limited access, US highways, separated A26 Primary road without limited access, US highways, separated, in tunnel A27 Primary road without limited access, US highways, separated, underpassing A28 Primary road without limited access, US highways, separated, with rail line in center Secondary and Connecting Road This category (A3) includes mostly state highways, but may include some county highways that connect smaller towns, subdivisions, and neighborhoods. The roads in this category generally are smaller than roads in Category A2, must be hard-surface (concrete or asphalt), and are usually undivided with single-lane characteristics. These roads usually have a local name along with a route number and intersect with many other roads and driveways. CFCC Description A31 Secondary and connecting road, state highways, unseparated A32 Secondary and connecting road, state highways, unseparated, in tunnel A33 Secondary and connecting road, state highways, unseparated, underpassing A34 Secondary and connecting road, state highways, unseparated, with rail line in center A35 Secondary and connecting road, state highways, separated A36 Secondary and connecting road, state highways, separated, in tunnel A37 Secondary and connecting road, state and county highways, separated, underpassing A38 Secondary and connecting road, state and county highway, separated, with rail line in center Local, Neighborhood, and Rural Road A road in this category (A4) is used for local traffic and usually has a single lane of traffic in each direction. In an urban area, this is a neighborhood road and street that is not a thorough-fare belonging in categories A2 or A3. In a rural area, this is a short-distance road connecting the smallest towns; the road may or may not have a state or county route number. Scenic park roads, unimproved or unpaved roads, and industrial roads are included in this category. Most roads in the Nation are classified as A4 roads. CFCC Description A41 Local, neighborhood, and rural road, city street, unseparated A42 Local, neighborhood, and rural road, city street, unseparated, in tunnel A43 Local, neighborhood, and rural road, city street, unseparated, underpassing A44 Local, neighborhood, and rural road, city street, unseparated, with rail line in center A45 Local, neighborhood, and rural road, city street, separated A46 Local, neighborhood, and rural road, city street, separated, in tunnel A47 Local, neighborhood, and rural road, city street, separated, underpassing A48 Local, neighborhood, and rural road, city street, separated, with rail line in center Vehicular Trail A road in this category (A5) is usable only by four-wheel drive vehicles, is usually a one-lane dirt trail, and is found almost exclusively in very rural areas. Sometimes the road is called a fire road or logging road and may include an abandoned railroad grade where the tracks have been removed. Minor, unpaved roads usable by ordinary cars and trucks belong in category A4, not A5. CFCC Description A51 Vehicular trail, road passable only by 4WD vehicle, unseparated A52 Vehicular trail, road passable only by 4WD vehicle, unseparated, in tunnel A53 Vehicular trail, road passable only by 4WD vehicle, unseparated, underpassing Road with Special Characteristics This category (A6) includes roads, portions of a road, intersections of a road, or the ends of a road that are parts of the vehicular highway system and have separately identifiable characteristics. CFCC Description A60 Special road feature, major category used when the minor category could not be determined A61 Cul-de-sac, the closed end of a road that forms a loop or turn-around A62 Traffic circle, the portion of a road or intersection of roads forming a roundabout A63 Access ramp, the portion of a road that forms a cloverleaf or limited-access interchange A64 Service drive, the road or portion of a road that provides access to businesses, facilities, and rest areas along a limited-access highway; this frontage road may intersect other roads and be named A65 Ferry crossing, the representation of a route over water that connects roads on opposite shores; used by ships carrying automobiles or people Road as Other Thoroughfare A road in this category (A7) is not part of the vehicular highway system. It is used by bicyclists or pedestrians, and is typically inaccessible to mainstream motor traffic except for private-owner and service vehicles. This category includes foot and hiking trails located on park and forest land, as well as stairs or walkways that follow a road right-of- way and have names similar to road names. CFCC Description A70 Other thoroughfare, major category used when the minor category could not be determined A71 Walkway or trail for pedestrians, usually unnamed A72 Stairway, stepped road for pedestrians, usually unnamed A73 Alley, road for service vehicles, usually unnamed, located at the rear of buildings and property A74 Driveway or service road, usually privately owned and unnamed, used as access to residences, trailer parks, and apartment complexes, or as access to logging areas, oil rigs, ranches, farms, and park lands Feature Class B, Railroad Railroad Main Line A railroad in this category is the primary track that provides service between destinations. A main line track often carries the name of the owning and operating railroad company. CFCC Description B11 Railroad main track, not in tunnel or underpassing B12 Railroad main track, in tunnel B13 Railroad main track, underpassing Railroad Spur A railroad in this category is the track that leaves the main track, ending in an industrial park, factory, or warehouse area, or forming a siding along the main track. CFCC Description B21 Railroad spur track, not in tunnel or underpassing B22 Railroad spur track, in tunnel B23 Railroad spur track, underpassing Railroad Yard A railroad yard track has parallel tracks that form a working area for the railroad company. Train cars and engines are repaired, switched, and dispatched from a yard. CFCC Description B31 Railroad yard track, not in tunnel or underpassing B32 Railroad yard track, in tunnel B33 Railroad yard track, underpassing Railroad with Special Characteristics A railroad or portions of a railroad track that are parts of the railroad system and have separately identifiable characteristics. CFCC Description B40 Railroad ferry crossing, the representation of a route over water used by ships carrying train cars to connecting railroads on opposite shores. These are primarily located on the Great Lakes. Railroad as Other Thoroughfare A rail line that is not part of the railroad system. This category is for a specialized rail line or railway that is typically inaccessible to mainstream railroad traffic. CFCC Description B50 Other rail line; major category used alone when the minor category could not be determined B51 Carline, a track for streetcars, trolleys, and other mass transit rail systems; used when the carline is not part of the road right-of-way B52 Cog railroad, incline railway, or logging tram Feature Class C, Miscellaneous Ground Transportation Miscellaneous Ground Transportation With Category Unknown Source materials do not allow determination of the miscellaneous ground transportation category. CFCC Description C00 Miscellaneous ground transportation, not road or railroad; major and minor categories unknown Pipeline Enclosed pipe, carrying fluid or slurry, situated above ground, or in special conditions, below ground when marked by a cleared right-of- way and signage. CFCC Description C10 Pipeline; major category used alone Power Transmission Line High voltage electrical line, on towers, situated on cleared right-of-way. CFCC Description C20 Power transmission line; major category used alone Miscellaneous Ground Transportation With Special Characteristics A portion of a ground transportation system that has separately identifiable characteristics. This category is for specialized transportation, usually confined to a local area, that is separate from other ground transportation. CFCC Description C30 Other ground transportation that is not a pipeline or a power transmission line; major category used alone when minor category could not be determined C31 Aerial tramway, monorail, or ski lift Feature Class D, Landmark Landmark is the general name given to a cartographic (or locational) landmark, a land-use area, and a key geographic location (KGL). A cartographic landmark is identified for use by an enumerator while working in the field. A land-use area is identified in order to minimize enumeration efforts in uninhabited areas or areas where human access is restricted. A key geographic location is identified in order to more accurately geocode and enumerate a place of work or residence. Landmark With Category Unknown Source materials do not allow determination of the landmark category. CFCC Description D00 Landmark; major and minor categories unknown Military Installation Base, yard, or depot used by the U.S. Army, Navy, Air Force, Marines, the Coast Guard, or the National Guard. With the exception of the Coast Guard which is administered by the Department of Transportation, and the National Guard which is administered by states, these areas are administered by the U.S. Department of Defense CFCC Description D10 Military installation or reservation; major category used alone Multihousehold or Transient Quarters CFCC Description D20 Multihousehold or transient quarters; major category used alone when the minor category could not be determined D21 Apartment building or complex D22 Rooming or boarding house D23 Trailer court or mobile home park D24 Marina D25 Crew-of-vessel area D26 Housing facility for workers D27 Hotel, motel, resort, spa, hostel, YMCA, or YWCA D28 Campground D29 Shelter or mission Custodial Facility This category includes institutions that have personnel such as guards, nurses, and caretakers to preserve the welfare of those individuals resident in the facility. CFCC Description D30 Custodial facility; major category used alone when the minor category could not be determined D31 Hospital D32 Halfway house D33 Nursing home, retirement home, or home for the aged D34 County home or poor farm D35 Orphanage D36 Jail or detention center D37 Federal penitentiary, state prison, or prison farm Educational or Religious Institution CFCC Description D40 Educational or religious institution; major category used alone when the minor category could not be determined D41 Sorority or fraternity D42 Convent or monastery D43 Educational institution, including academy, school, college, and university D44 Religious institution, including church, synagogue, seminary, temple, and mosque Transportation Terminal The facility where transportation equipment is stored, the destination for travel on the transportation system, or the intermodal connection facility between transportation systems. CFCC Description D50 Transportation terminal; major category used alone when the minor category could not be determined D51 Airport or airfield D52 Train station D53 Bus terminal D54 Marine terminal D55 Seaplane anchorage Employment Center This category includes locations with high-density employment. CFCC Description D60 Employment center; major category used alone when the minor category could not be determined D61 Shopping center or major retail center D62 Industrial building or industrial park D63 Office building or office park D64 Amusement center D65 Government center D66 Other employment center Tower CFCC Description D70 Tower; major category used alone when minor category could not be determined D71 Lookout tower Open Space This category contains areas of open space with no inhabitants, or with inhabitants restricted to known sites within the area. CFCC Description D80 Open space; major category used alone when the minor category could not be determined D81 Golf course D82 Cemetery D83 National Park Service land D84 National forest or other Federal land D85 State or local park or forest Special Purpose Landmark This category includes landmarks not otherwise classified. CFCC Description D90 Special purpose landmark; major category used alone when the minor category could not be determined D91 Post office D92 Urbanizacion, an identifiable community development in Puerto Rico D93 Fire Department D94 Police Station D95 Library D96 City/Town Hall Feature Class E, Physical Feature Physical Feature With Category Unknown Source materials do not allow determination of the physical feature category. CFCC Description E00 Physical feature, tangible but not transportation or hydrographic; major and minor categories unknown Fence This category describes a fence that separates property. For example, a fence around a military reservation or prison separates the reservation from civilian land. Thus, a fence line is a property line marked by a fence. CFCC Description E10 Fence line locating a visible and permanent fence between separately identified property Topographic Feature This category refers to topographical features that may be used as boundaries or as a reference for an area. The Census TIGER data base contains topographic features used to define the limits of statistical entities in locations where no other visible feature can be identified. CFCC Description E20 Topographic feature; major category used when the minor category could not be determined E21 Ridge line, the line of highest elevation of a linear mountain E22 Mountain peak, the point of highest elevation of a mountain E23 Island, identified by name E24 Levee, an embankment, as of earth or concrete, used to prevent a river or other body of water from overflowing Feature Class F, Nonvisible Features Nonvisible features are used to delimit tabulation entities, property areas, and legal and administrative entities. The U.S. Census Bureau separately identifies nonvisible boundaries only when they do not follow a visible feature such as a road, stream, or ridge line. Nonvisible Boundary With Classification Unknown or Not Elsewhere Classified CFCC Description F00 Nonvisible boundary; major and minor categories unknown Nonvisible Legal Entity Boundary CFCC Description F10 Nonvisible jurisdictional boundary of a legal or administrative entity F11 Offset boundary of a legal entity F12 Corridor boundary of a legal entity F13 Nonvisible superseded 2000 governmental unit boundary F14 Superseded 1990 legal boundary F15 Superseded 1990 legal boundary, corrected through post census process F16 Superseded legal boundary, current at the time of the 1997 Economic Census F17 Nonvisible State Legislative District boundary F18 Nonvisible Congressional District boundary F19 Nonvisible corrected 2000 governmental unit boundary Nonvisible Features for Data Base Topology This category contains various types of nonvisible lines used to maintain the topology in the Census TIGER data base. CFCC Description F20 Nonvisible feature for data base topology; major category used when the minor category could not be determined F21 Automated feature extension to lengthen existing physical feature F22 Irregular feature extension, determined manually, to lengthen existing physical feature F23 Closure extension to complete data base topological closure between extremely close features (used to close small gaps between complete chains and create polygons to improve block labeling on cartographic products) F24 Nonvisible separation line used with offset and corridor boundaries F25 Nonvisible centerline of area enclosed by corridor boundary Point-to-Point Line CFCC Description F30 Point-to-point line, follows a line of sight and should not cross any visible feature; for example, from the end of a road to a mountain peak Property Line CFCC Description F40 Property line, nonvisible boundary of either public or private lands, e.g., a park boundary ZIP Code Tabulation Boundary CFCC Description F50 ZIP Code tabulation boundary, used in delineating ZIP Code Tabulation Areas Nonvisible Statistical Boundary CFCC Description F70 Statistical boundary; major category used when the minor category could not be determined F71 1980 statistical boundary F72 1990 statistical boundary; used to hold 1990 collection and tabulation census block boundaries not represented by existing physical features F73 Internal U.S. Census Bureau use F74 1990 statistical boundary; used to hold a 1990 tabulation census block boundary not represented by an existing physical feature Nonvisible Other Tabulation Boundary CFCC Description F80 Nonvisible other tabulation boundary; major category used when the minorcategory could not be determined F81 School district boundary F82 Internal U.S. Census Bureau use F83 Census 2000 collection block boundary; used to hold Census 2000 collection block boundaries not represented by existing physical features F84 Census 2000 statistical area boundary; used to hold Census 2000 statistical area boundaries not represented by existing physical features F85 Census 2000 tabulation block boundary; used to hold Census 2000 tabulation block boundaries not represented by existing physical features F86 Internal U.S. Census Bureau use F87 Oregon urban growth area boundary F88 Current statistical area boundary Feature Class G, U.S. Census Bureau Usage The U.S. Census Bureau uses this feature class for internal programs. Feature Class H, Hydrography Basic Hydrography This category includes shorelines of all water regardless of the classification of the water itself. CFCC Description H00 Water feature, classification unknown or not elsewhere classified H01 Shoreline of perennial water feature H02 Shoreline of intermittent water feature Naturally Flowing Water Features CFCC Description H11 Perennial stream or river H12 Intermittent stream, river, or wash H13 Braided stream or river Man-Made Channel to Transport Water These features are used for purposes such as transportation, irrigation, or navigation. CFCC Description H21 Perennial canal, ditch, or aqueduct H22 Intermittent canal, ditch, or aqueduct Inland Body of Water CFCC Description H30 Lake or pond; major category used when the minor category could not be determined H31 Perennial lake or pond H32 Intermittent lake or pond Man-Made Body of Water CFCC Description H40 Reservoir; major category used when the minor category could not be determined H41 Perennial reservoir H42 Intermittent reservoir Seaward Body of Water CFCC Description H50 Bay, estuary, gulf, sound, sea, or ocean; major category used when the minor category could not be determined H51 Bay, estuary, gulf, or sound H53 Sea or ocean Body of Water in a Man-Made Excavation CFCC Description H60 Gravel pit or quarry filled with water Nonvisible Definition Between Water Bodies The U.S. Census Bureau digitizes nonvisible definition boundaries to separate named water areas; for instance, an artificial boundary is drawn to separate a named river from the connecting bay. CFCC Description H70 Nonvisible water area definition boundary; used to separate named water areas and as the major category when the minor category could not be determined H71 USGS closure line; used as a maritime shoreline H72 Census water center line; computed to use as a median positional boundary H73 Census water boundary, international in waterways or at 10-mile limit; used as an area measurement line H74 Census water boundary separating inland from coastal or Great Lakes; used as an area measurement line H75 Census water boundary separating coastal water from territorial sea at the 3-mile limit; used as an area measurement line Special Water Feature Includes area covered by glaciers or snow fields. CFCC Description H80 Special water feature; major category used when the minor category could not be determined H81 Glacier Feature Class P, Provisional Features The U.S. Census Bureau has created a new CFCC type that may appear on street features only. Some streets that normally would be classified as "A" class features may be coded with a "P" instead of the "A" to indicate that the feature is a "provisional" feature. Provisional features are those streets that were added from reference sources or other programs in preparation for Census 2000, but were not field verified by census staff during field operations or through the use of aerial photography or imagery. As these features are verified in future operations the provisional flag will be removed for subsequent TIGER/Line releases. The numeric portion of the CFCC still classifies the street as if an "A" were preceding it. Feature Class X, Not Yet Classified Classification Unknown or Not Elsewhere Classified CFCC Description X00 Feature not yet classified All complete chains, landmarks, and key geographic locations have a code representing their census feature class. Only those GT-polygons associated with an area landmark have a CFCC. Most CFCCs in the feature classification scheme apply only to complete chains. In a few instances, the same feature code may apply to complete chains as well as to point and area landmarks. Only those features required for census operational purposes are classified and inserted into the Census TIGER data base. Therefore, not all features in a county will appear in the TIGER/Line files. Since features are classified with only a single code, a road that also is a boundary will have only the CFCC of a road even though a CFCC for a boundary exists in the classification scheme. CFCC Record Location Record Type Field Name Description 1 CFCC Code assigned to the complete chain 7 CFCC Code assigned to a point or area landmark 9 CFCC Code assigned to a key geographic location Points Describing the Complete Chain The TIGER/Line files describe the spatial/geometric position and shape of a complete chain using shape points and nodes; see the section entitled Topology in Chapter 1. Latitude and longitude coordinate fields identify the shape points and nodes. The Census TIGER data base does not support node identification numbers. Nodes Nodes are topological objects that mark the end location of each complete chain. Every chain has two nodes, a start node and an end node (using the Spatial Data Transfer Standard, or SDTS, terminology). Earlier releases of the TIGER/Line files refer to these nodes as the from node and the to node. The order of the nodes establishes the left and the right sides of the line and sets the sequencing order for the shape points. The node coordinates are stored in Record Type 1. Shape Points The U.S. Census Bureau uses the term shape points to describe the non- topological points that describe the position and shape of a chain. Shape points exist only where required; straight-line complete chains require no shape points. Shape points are associated only with one complete chain and are listed in order from start node to end node. The TIGER/Line files store shape points in Record Type 2 and link them to the nodes in Record Type 1 using the TLID. The shape points for a chain can fill several Type 2 records. Coordinates for Nodes and Shape Points Coordinates are decimal degrees expressed in Federal Information Processing Standard (FIPS) notation, where a positive latitude represents the Northern Hemisphere and a negative longitude represents the Western Hemisphere. All coordinates are expressed as a signed integer with six decimal places of precision implied (see the section, Positional Accuracy, in Chapter 5). Actual TIGER/Line File Latitude 15 Deg. S to 72 Deg. N -15000000 to +72000000 Longitude 64 Deg. W to 131 Deg. E -64000000 to -180000000 +179999999 to +131000000 For the 48 contiguous states, the District of Columbia, Alaska, Puerto Rico, and the Virgin Islands, the coordinates in the 1995 and later versions of the TIGER/Line files are in the North American Datum of 1983 (NAD83). The coordinate datum for the above areas was NAD27 in all previous versions of the files prior to 1995. For Hawaii and the Pacific Island Areas, the Census Bureau used a variety of sources for building the original digital file in the late 1980s. Neither the specific identities of each of these sources nor their datums were recorded. The information that does exist for this operation indicates that the current USGS topographic quadrangles and/or Defense Department maps were typically, though not necessarily exclusively, used as sources. These would have been based on local datums, however, the Census Bureau does not have information specifically identifying these datums. Such information was not needed for Census Bureau mapping operations when the TIGER data base was created for these areas. Coordinate Values All nodes have non-zero coordinates within the range specified in the Coordinates for Nodes and Shape Points section on the previous page. Shape point coordinates are expressed in the same manner. However, unused Record Type 2 fields are zero-filled and begin with a "+" sign. Record Locations for Nodes and Shape Point Coordinates Record Type Field Name Description 1 FRLONG Start Longitude 1 FRLAT Start Latitude 1 TOLONG End Longitude 1 TOLAT End Latitude 2 LONG1 Point 1, Longitude 2 LAT1 Point 1, Latitude 2 LONG2 Point 2, Longitude 2 LAT2 Point 2, Latitude 2 LONG3 Point 3, Longitude 2 LAT3 Point 3, Latitude . . . . . . . . . 2 LONG10 Point 10, Longitude 2 LAT10 Point 10, Latitude Record Linkages/Feature Chaining Plotting a complete chain requires using the nodes from Record Type 1 and all of the shape point records in Record Type 2 with the same TLID, if any. Plot the start node first, then search Record Type 2 for any matching records. If there is a match, the record will contain from 1 to 10 shape points. If all 10-point fields are filled with non-zero values, there may be an additional matching Type 2 record. Type 2 records are not sorted by TLID, but all records with the same TLID should appear together in sequence by the record sequence number (RTSQ). Plot the shape points from all Type 2 records and end the complete chain by plotting the end node. Street features may consist of multiple complete chains that are sequentially linked together. Linking all of the features with the same name requires the extraction of all Type 1 and Type 2 records with the same feature identifiers in Record Types 1 and 5. Boundary generation requires the extraction of all features that have different left and right geographic codes. The placement of the complete chains into a boundary-ring sequence requires a procedure to match the end of one complete chain to the beginning or end of the next complete chain. The complete chains will probably not have the same to-from or start-end orientation down the length of the street or boundary. Therefore, the procedure must reverse the order of the nodes and shape points that form some complete chains to achieve a correct and consistent sequence of nodes and shape points. Since the nodes that identify the ends of the complete chains do not have an identification number, the procedure must match the nodes based on the latitude and longitude coordinates. Combining the coordinates into a single peano key code composed of alternating latitude and longitude digits might facilitate the match. Sorting nodes using the peano key will cluster nodes that are spatially close together. Polygon Features The TIGER/Line files contain identification and geographic codes for each GT-polygon in the Census TIGER data base. These GT-polygons are the smallest areas identified in the TIGER/Line files. Geographic entities and area landmarks have specific identification codes and form more complex polygons. The TIGER/Line files link these features to GT-polygons, but do not directly identify the more complex polygons. GT-polygons are building blocks that form features. They are not features and do not have their own feature name or CFCC. However, GT-polygons may be a part of many area landmark features that have their own feature name and CFCC. GT-polygons have unique GT-polygon identification codes (CENID and POLYID), a set of geographic entity codes, and an internal point location. Refer to Chapter 2 for more information on GT-polygon identification codes and Chapter 4 for a description of the geographic entities in the TIGER/Line files. Information and record linkage keys for GT-polygons are distributed over several record types: Record Type P - provides the GT-polygon internal point location Record Type A - provides the Census 2000 geographic entity codes and areas Record Type 8 - links GT-polygons to area landmarks Record Type 9 - links GT-polygons to key geographic location features Record Type I - links GT-polygons to complete chains Record Type S - provides Census 2000 geographic entity codes and areas Updates to the Census TIGER data base include new street and boundary complete chains that create new GT-polygons. Thus, each version of the TIGER/Line files will have a single, unique set of GT-polygons, each with a corresponding Record Type A, S, and P. The CENID and POLYID identification codes link records together, but are not permanent GT- polygon identification codes. Geographic Entity Codes Geographic entity codes can be attributes of a set of polygons, a complete chain, or both. Refer to Chapter 6 for the data dictionary that describes the record type fields and to Chapter 4 for descriptions of geographic areas. Internal Points The internal point is a point location within each GT-polygon that is unique to that GT-polygon. The TIGER/Line files exclude the internal points from the node-complete chain-polygon topology; do not confuse the internal point with a centroid. In a polygon with an irregular shape, such as a doughnut or crescent shape, the true centroid could fall outside the polygon. Unlike true centroids, the internal points should always fall within the GT-polygon or on the GT-polygon boundary. Some of the GT-polygons (approximately a dozen nationwide) are so small that the internal point may be identical to a point on one of the lines bounding the GT-polygon, or identical to one of the nodes. Depending upon the precision of a particular software or hardware system, the data user may find the internal point outside the correct GT-polygon, or find that a GT- polygon may contain two internal points. Changes to the shape and location of complete chains forming polygon boundaries will change the polygon internal point coordinates even though the topology of the polygon remains the same. Such changes complicate the matching, using internal point coordinates, of polygons from different versions of the TIGER/Line files. All internal points have non-zero coordinates. Coordinates are expressed in standard FIPS PUB 70 notation. See the Coordinates for Nodes and Shape Points section in this chapter. GT-Polygon Internal Point Coordinates Record Locations Record Type Field Name Description P POLYLONG Polygon Internal Point Longitude P POLYLAT Polygon Internal Point Latitude Record Linkages The topological network of complete chains divides the surface area of geographic entities into GT-polygons. There is a one-to-one relationship between the GT-polygons constructed from Record Types 1 and 2 and those appearing in Record Type P. In constructing the GT-polygons from Record Types 1 and 2, users are cautioned to be sure their software has the necessary coordinate precision and does not snap together complete chains that are merely close. Record Type I provides a direct link from each complete chain in the TIGER/Line file to its adjoining GT-polygons. It contains both the TLID and the polygon identification codes for each side of the GT-polygon. Record Type I facilitates the transfer of polygon geographic codes to the complete chain, but also provides the link back from polygon to complete chain. In this case, finding all complete chains associated with a GT- polygon is more difficult. The procedure involves searching every Type I record to locate all instances where a CENID and POLYID appear on either the left or the right side of a complete chain. Area landmarks also must link to the GT-polygons in order to establish their geographic location. Record Type 8 provides the link from GT-polygon to area landmark. See the Area Landmark Locations section in this chapter. Landmark Features The U.S. Census Bureau includes landmarks in the Census TIGER data base for locating special features and to help enumerators during field operations. Some of the more common landmark types include airports, cemeteries, parks, and educational facilities. The U.S. Census Bureau added landmark features on an as-needed-basis and made no attempt to ensure that all instances of a particular feature were included. The absence of a landmark does not mean that the living quarters, e.g., hospitals and group quarters associated with the landmark were excluded from the Census 2000 enumeration. The address list used for the census was maintained apart from the landmark data. Landmarks with special address information are called key geographic locations (KGLs). A landmark can be either a point, line, or area type. In some cases, the Census TIGER data base permits a choice of types. For instance, an airport or airfield might appear as a point, line, or area; the approach depends on the size of the feature and the depiction of the feature in the source document. Line features such as airfields could appear as one or more complete chains; they are not identified in the landmark record types. See the Point, Line, and Area Landmark CFCCs section in this chapter to identify the possible codes that could appear as complete chains. In addition to landmark data, the TIGER/Line files contain the CFCCs and names for bodies of water including ponds, lakes, oceans, and the area covered by large streams represented as double-line drainage. See Chapter 4 for a complete description of census blocks covering land and water. Landmark and water features can overlap. The most common situation is a park or other special land-use feature that includes a lake or pond. In this case, the GT-polygon covered by the lake or pond belongs to a water landmark feature and a park landmark feature. Other kinds of landmarks can overlap as well. Area landmarks can contain point landmarks; these are not linked in the TIGER/Line files. Record Type 7 contains point and area landmarks. Most but not all water areas are identified as an area landmark whether named or not. The other landmarks may be identified only by a census feature class code and may not have a name. During the extraction of this data, the U.S. Census Bureau assigned a temporary landmark identification number (LAND) to each landmark record. Record Type 8 uses the LAND to link the area landmark records in Record Type 7 to the GT-polygons. Record Type 7 and Record Type 8 exist only when the county file contains landmark features or water features. Record Type 9 contains the key geographic locations (KGLs) in the Census TIGER data base. The KGLs are linked by the CENID and POLYID to the GT-polygons. Point, Line, and Area Landmark CFCCs All landmarks, including KGLs, have a CFCC. In the Census TIGER data base the CFCCs of the complete chains forming the polygon boundary are independent of the CFCCs assigned to the area landmark or the water feature filling the polygon. Landmark CFCC Record Locations Record Type Field Name Description 7 CFCC Code assigned to point and area landmarks 9 CFCC Code assigned to key geographic location Landmark CFCC Codes CFCC Description Point Line Area D00 Landmark feature, classification unknown, or not elsewhere classified P L A D10 Military installation P - A D20 Multihousehold and transient quarters P - A D21 Apartment building or complex P - A D22 Rooming or boarding house P - - D23 Trailer court or mobile home park P - A D24 Marina P - A D25 Crew-of-vessel area P - - D26 Housing facility for workers P - A D27 Hotel, motel, resort, spa, YMCA, or YWCA P - A D28 Campground P - A D29 Shelter or mission P - A D30 Custodial facility P - A D31 Hospital P - A D32 Halfway house P D33 Nursing home, retirement home, or home for the aged P - A D34 County home or poor farm P - A D35 Orphanage P - A D36 Jail or detention center P - A D37 Federal penitentiary, state prison, or prison farm P - A D40 Educational or religious institution P - A D41 Sorority or fraternity P - - D42 Convent or monastery P - A D43 Educational institution P - A D44 Religious institution P - A D50 Transportation terminal P L A D51 Airport or airfield P L A D52 Train station P - A D53 Bus terminal P - A D54 Marine terminal P - A D55 Seaplane anchorage P - A D60 Employment center P - A D61 Shopping center or major retail center P - A D62 Industrial building or industrial park P - A D63 Office building or office park P - A D64 Amusement center P - A D65 Government center P - A D66 Other employment center P - A D70 Tower P - - D71 Lookout tower P - - D80 Open space - - A D81 Golf course P - A D82 Cemetery P - A D83 National Park Service area P - A D84 National forest or other federal land P - A D85 State or local park or forest P - A D90 Special purpose landmark P - A D91 Post office P - A D92 Urbanizacion, an identifiable community development in Puerto Rico P - A H00 Water feature, classification unknown, or not elsewhere classified P L A H11 Perennial stream or river - L A H12 Intermittent stream, river, or wash - L A H13 Braided stream or river - L A H21 Perennial canal, ditch, or aqueduct - L A H22 Intermittent canal, ditch, or aqueduct - L A H31 Perennial lake or pond - - A H32 Intermittent lake or pond - - A H41 Perennial reservoir - - A H42 Intermittent reservoir - - A H50 Bay, estuary gulf, sound, sea, or ocean - - A H51 Bay, estuary gulf, or sound - - A H53 Sea, or ocean - - A H60 Gravel pit or quarry filled with water - - A H80 Special water feature - - A H81 Glacier - - A Landmark Feature and KGL Names The TIGER/Line files contain an optional 30-character text string used to identify the proper name of the landmark feature or water area. The text string includes upper- and lower-case characters. The feature name may carry an imbedded feature type (e.g., River, Military Reservation, Garden, Park, and Lake). The U.S. Census Bureau has not standardized or edited the feature types or names for landmarks in the Census TIGER data base in all areas. The U.S. Census Bureau does not guarantee that the landmarks or water areas are consistently identified in the TIGER/Line files. Area landmarks added to the Census TIGER data base in different update actions with the same name and CFCC will produce separate landmark records in the TIGER/Line files. The landmark records may contain variant spellings of the feature name or different CFCCs even though they refer to the same feature. These differences could result in the fragmentation of a large landmark. For instance, a water body could have the name Lake Redmand with a CFCC of H31, while another part could have the same name, but a CFCC of H30, and still a third part could have the name York County Reservoir. Because area landmarks can overlap, it is possible, although not likely, for one polygon to belong to several landmarks. Area landmarks and water area labels can have alternate names. Each feature name will appear as a separate Type 7 record, but each record will have the same LAND. Type 7 Records with the same LAND will have the same landmark or water area label. Each unique combination of primary and alternate names becomes a separate landmark record even though the primary name and the CFCCs match the adjoining landmark features. The TIGER/Line files do not show all water bodies as landmark records. Using Record Type 7 (area landmarks) and Record Type 8 (polygons linked to area landmarks) will not necessarily provide all water areas. Record Type S contains a water flag (WATER) to identify polygons associated with water bodies. Water bodies are identified with a value of 1 in the WATER field. Key geographic location names uniquely identify the landmark separately; for example, Springfield Shopping Center. Landmark Feature Record Locations Record Type Field Name Description 7 LANAME Landmark name 9 KGLNAME Key geographic location name Landmark Feature Name Codes The LANAME and KGLNAME field may include any ASCII text string. The fields can be blank where the feature is unnamed. Point Landmark Locations The TIGER/Line files identify the location of point landmarks with a single coordinate point. The presence of coordinate data in Record Type 7 distinguishes point landmarks from area landmarks that have blank coordinate fields. Coordinates Coordinates are expressed in standard FIPS PUB 70 notation. For additional information, see the Coordinates for Nodes and Shape Points section in this chapter. Point Landmark Coordinate Record Locations Record Type Field Name Description 7 LALONG Longitude 7 LALAT Latitude Coordinate Values All point landmarks have non-zero coordinates within the range specified above. The coordinate fields for area landmarks are blank-filled. Area Landmark Locations To find the location of each area landmark, link the basic landmark description in Record Type 7 to all of the elementary polygons that belong to the landmark. Record Type 8 serves as a bridge between these two record types. The TIGER/Line files provide a Type 8 record for each polygon linked to a specific landmark. Polygons belonging to multiple landmarks appear once for each landmark. The TIGER/Line files use the LAND and the polygon identification codes (CENID and POLYID) to actually make the link. See Chapter 2 for a description of the LAND, CENID, and POLYID codes and fields. Locate the polygons for an area landmark by searching Record Type 8 for all of the CENIDs and POLYIDs with the specified LAND. Record Type 8 is in LAND sort sequence. Once the polygons are linked to the area landmark, use Record Type I to locate the complete chains that form the landmark's polygon boundaries. Record Type I contains a record for all complete chains and identifies the polygons located on either side of the complete chains. The search procedure must look for all instances of Record Type I and evaluate the left- and right-side polygon identifiers for a possible match. Data users may need to eliminate complete chains that are internal to the polygon and landmark, depending on the application. KGLs To find the location of KGLs, link the description in Record Type 9 to the elementary polygon in which the KGL is found. Use the polygon identification codes (CENID and POLYID) to make the link. To link the KGL to a feature, use the FEAT field (alternate feature ID code) to link to the feature identifier in Record Type 5.