USVI Public Records

Document text

ISLAND RESOURCES FOUNDATION Caribbean Headquarters RED HOOK 80X 33. ST. THOMAS U.S. VIRGIN ISLANDS 00802 (809) 775-6225 Washington, D. C. Office 1718 P STREET. N.W .. SUITE T4 WASHINGTON. D.C. 20036 (202) 265-971 2 VIRGIN ISLANDS LAND USE SURVEY Final Project Report Prepared for: Department of Planning and Natural Resources Government of the U.S. Virgin Islands Prepared by: Island Resources Foundation St. Thomas, U.S. Virgin Islands MAY 1989 Edward L Towle, Ph.D. President Steve L Colman, M.R.P. Jean-Pierre Bacle, B.A. Clarence Cuthbertson, B.S. Ian Jones, B.S. Team Leader PROJECT STAFF JudHh A. Towle, M.P.A. Bruce Potter, B.A. Sandra Tate TABLE OF CONTENTS Page No. Chapter 1 Introduction 1 Chapter 2 The Mapping Process 2 Chapter 3 Geographic Reference Systems and Geopositioning 9 Chapter 4 Land Use Categories 18 Chapter 5 Land Use Database Design 25 Chapter 6 Seminar on GIS 39 Chapter 7 Planning, Data Management and Policy Recommendations 50 CHAPTER 1 INTRODUCTION On March 8, 1989, the Governor of the Virgin Islands signed contract number PC-P&NR - 203-89 between the Department of Planning and Natural Resources (DPNR) and Island Resources Foundation (IRF). The contract called for Island Resources Foundation to conduct a land use survey for the U.S. Virgin Is- lands. A major work product of the land use survey project is a series of maps that depict the existing land uses in the Virgin Islands. This report, to which the maps are appended, provides a description of the land use mapping process utilized by the project as well as limited analysis and recommendations ex- tracted from data generated during the project. The report also conveys the other work products of the project, which are: a description of the land use classification intensity scale system; a description of a geo-referencing system; a description of a land use database (including an expanded land use classification system); a summary of a seminar on geographic information systems (presented to DPNR on April 17,1989); and a bibliography. The land use maps are a key work product which can stand alone, but this re- port contains important supplementary information on the land use classifi- cation system and the mapping process. Further, we note in passing some very serious project constraints, such as the eight-week time frame (foreshortened from six months), incomplete information, missing DPNR baseline maps and in- ventory data, and map-scale issues all provided a difficult context for the project which is important to understand. IRF believes that the information, analysis, and maps produced by this project are a substantial step towards the development of a suitable land use information system for the Virgin Is- lands. It is impossible in an eight-week project to do more than provide the basis upon which DPNR can build an appropriate land use and natural resources man- agement database. IRF is confident that the material conveyed in this report provides DPNR with a more accurate picture of the Territory's current land use regime than it has had at any time in the past twenty-five years. In addition, the project has provided the tools to construct a sophisticated land use database system. CHAPTER 2 THE MAPPING PROCESS This chapter describes the process used by IRF to map the land uses of the Vir- gin Islands. A discussion of map scale and air photograph interpretation is provided to enable a better understanding of the map products. In addition each of the land use categories is described in detail. Many of the parameters that applied to the map work in this project were dic- tated by the terms of IRF's contract with DPNR. The most significant of these were: 1. A map scale of 1:24,000 for island-wide land use maps; 2. A map scale of 1 :12,000 for the "urbanizing" areas of St. Thomas and St. Croix; and 3. The use of 1988 black and white air photographs as the primary data source. As discussed earlier, the time constraints of the project allowed a maximum of four weeks to complete all air photo interpretation. The mapping process followed a ten step procedure as follows: 1. Devise a land use classification system; 2. Utilize 1988 air-photographs to identify each land use; 3. Outline each land use onto a mylar over lay at the appropriate scale; 4. Note on the mylar overlay those areas where there is uncertainty and field checking is required. 5. Field check as necessary; 6. Correct lines on mylar overlay to reflect field checked information; 7. Ink lines on mylar overlay; 8. Run inked mylar overlay together with mylar base map to produce a draft land map; 9. Further refine draft maps; 10 Produce final maps. Chapter 2 The Mapping Process IRF's contract with DPNR required that the island-wide land use maps be produced at a scale of 1:24,000 (one unit on the map equals 24,000 units on the ground; one inch on the map equals two thousand feet on the ground). The contract, as clarified, specified a scale of 1:12,000 for the "urbanizing" areas of St. Thomas and St. Croix. DPNR requested that the following urbanizing areas be mapped: East End and Tutu Planning Districts, St. Thomas; and Sion Farm and Southwest Planning Districts, St. Croix. IRF's contract also specified that the land use mapping be based on interpreta- tion of the 1988 series of air photographs flown for the U.S. Army Corps of En- gineers. The series were flown on various days of March, 1988, at a flight level of 4,800 feet. The scale of the photographs is as follows: at sea level, 1 inch equals 833 feet; at 200 feet above sea level, 1 inch equals 800 feet; and at 400 feet above sea level, 1 inch equals 767 feet. The focal length of the camera lens was 152.6 mm. A transparent grid was ruled to show the size of one acre on the air pho- tographs. This was used to determine the density of structures. This density information combined with locally obtained information facilitated the photo interpretation process. The U.S. Geological Survey topographical maps were used as the base maps for the island-wide mapping. These maps had been photo-revised in 1982, (based on photos from before 1978) and indicated recent developments in magenta. The base map for the urbanizing areas was the CH2M Hill, 1:12,000 se- ries. A considerable amount of field verification was required to ensure map accuracy. Field verification involved a simple but time consuming process of "windshield surveying" those areas where the land use could not be gleaned from the air- photograph. The actual land use would be mapped directly onto the air photo- graph with a grease pencil and the information transferred to a map at a later date. It is important to make several observations regarding map scale. The scale se- lected (1;24,000) is one typically used for planning purposes and is appropriate to indicate island-wide land use patterns. It is not the scale of choice for future more detailed mapping. For the purposes of this project the map scale is adequate to provide a general sense of the location, size and pattern of the ba- sic land uses in the Virgin Islands. It is not adequate to allow for detailed management, planning or engineering. When the Virgin Islands decides to pursue a geographic information system or other more detailed land use mapping program, attention must be paid to the selection of appropriate map scale. Chapter 2 The Mapping Process THE LAND USE CLASSIFICATION SYSTEM USED FOR MAPPING There are four main considerations that went into the development of the land use classification system and intensity scale. These are: 1. The scope of the project (limits of available time and budget); 2. The quality, quantity and availability of land and water use data; 3. The purpose for which the classification system and map products will be used; and, 4. Compatibility with prior land use mapping efforts in the Virgin Is- lands. Project time constraints require that the land use system always be a balanced compromise between the ideal and the practical. For instance, Teytaud, 1981, developed a classification system which has seven level-1 (general) land use types, 36 level-2 (detailed) types and some 100 level-3 (very precise) types. Other land use inventory systems that were examined include Zube's 1968 land use survey (12 categories), a 1962 land use classification manual (O'Harrow, 1962), and the Virgin Islands Planning Office's 1977 Land Use and Housing Element (12 categories). In addition, the Virgin Islands Zoning Law (Title 29, Chapter 21, V.I.Code) and the Virgin Islands Coastal Zone Management Act and Coastal Land and Water Use Plan (Title 12, Chapter 21, V.I.) were consulted. As noted, it is possible to devise a very fine land use classification system that identifies up to several hundred land uses. For instance agricultural land can be mapped simply as "agricultural land" or can be broken down into use types such as crop production, silviculture, animal pasture, and aqua-culture or hy- droponic farms. Each of these can, in turn, be further characterized; for in- stance, crop land can be identified as bananas, pineapple, or sugar cane, if the mapper has sufficient time and adequate data. An underlying objective of this land use survey is to provide a sufficient level of detail within the time and budget limits of the project. The land use maps that are the final product of this project are expected to be used principally as a planning tool. The level of detail is adequate for planning purposes. The maps displaying this classification system indicate patterns of development, ar- eas that can be expected to develop in the future, and areas that are in need of protection or intensive management. The primary data source for conducting the land use inventory is the set of air- photographs of the Virgin Islands flown by the Army Corps of Engineers in March, 1988. There are numerous secondary sources including field surveys, 1989 air- photos, and current land use data from the Virgin Islands Government. Our research has indicated that there has not been a survey of the major land uses of the Virgin Islands since the 1977, Land Use and Housing Element. The Land Use and Housing Element was funded by the U.S. Department of Housing and Urban Development and conducted by the Virgin Islands Planning Office in 1976 and 1977. An original intent of this project was to replicate some important aspects of the Land Use and Housing Element, such as the land use classifica- tions and database. Unfortunately, as a result of the limited scope and time Chapter 2 The Mapping Process frame of this land use survey project, it is not practical to replicate the 1977 land use classifications. In addition, the land use categories are not defined in sufficient detail to be replicated in this project. Since the data generated by the 1977 land use study is based on inadequately defined criteria, it is not possible to generate meaningful statistical land use change information, based on the 1977 data. The land use classification system developed for this project is intended to be replicable. The land use categories that have been selected for inclusion are the most significant land uses in the Virgin Islands and are of the type that are typically used for planning purposes. Further, the categories roughly cor- relate to the existing classification of the Virgin Islands Zoning Law (low-den- sity residential, agricultural, commercial, etc.). The classification system also will lend itself to further refinement and possible use as a base for a computerized geographic information system (GIS) [See Chapter 4J. A final consideration that entered into the selection of the land use categories was the extent to which each use could be identified from the 1988, black and white air-photographs and how such information would be transferred onto maps. A preliminary land use classification system was developed on March 15, 1989, and reviewed by the staff of the Department of Planning and Natural Resources. The system was tested and revised the following week. The system described in this report has been tested and found to be workable for the presentation of the major land and water uses of the Virgin Islands. The system also provides a level of detail that is appropriate for the two map scales that are being used in this project, i.e., 1:24,000 and 1:12,000. Following is a description of the land use classification system, by land use type, as used in the Virgin Islands Land Use Inventory Project. The intensity scale for each use, as appropriate, is also provided. The classification system is intended to reflect the actual, not potential, uses of land and water. Thus, undeveloped land that is zoned for medium-density residential development will be mapped as undeveloped land. In those cases where a mix of land uses are noted and determined not to be separately identifiable, the predominate land use has been mapped. For example, if a single family home is identified within a farm, the area will be mapped as agricultural land. The minimum mapping unit for this project ranges from one acre for residential uses to 5 acres for re- tail/commercial uses. LAND USE CLASSIFICATION SYSTEM 1. AGRICULTURAL: All land dedicated to agriculture falls into this category. This is a broad category, intended to capture the variety of agricul- tural use in the Virgin Islands. Dairy farming, cattle grazing, crop production, tree farming, and agricultural experiment and field sta- tions are examples of this classification. 2. RESORT/HOTEL: In an economy where tourism is one of the largest sectors, such as the Virgin Islands, it makes sense to identify resort and ho- tel uses. All resorts and hotels that have been listed by the Virgin Islands Department of Economic Development have been mapped. Hotels and resorts are facilities that: are used primarily by visitors to the islands; rent on a daily or weekly basis; often include ancillary uses Chapter 2 The Mapping Process such as swimming pools and restaurants; and usually are owned or man- aged by a single corporation. We have distinguished hotels and resorts from non-resort condominiums. While increasingly, condominiums are being used for the same purpose as hotels, they are closer to being a residential land use and gener- ally are mapped as a medium- or high-density land use. In those cases where the distinction is blurred, such as hotels where rooms are owned by individuals, a judgment has been made based on the primary use of the facility. 3. RETAIL/COMMERCIAL: This category is intended to include a wide variety of uses and activities., It is typical to find these uses clustered to- gether in small shopping centers or as strip development along the ma- jor roads. Examples of uses included in this classification are gas stations, convenience stores, fast food establishments, small and large shopping centers, restaurants, auto dealerships-auto-servicing centers, and office buildings. 4. INDUSTRIAL/MANUFACTURING: Industrial uses of land in the Virgin Islands range from watch assembly and other light manufacturing to oil refin- ing, electrical generation and water desalination. Other uses in- cluded in this category are mining/quarrying, auto-graveyards, ware- houses, and non-marine storage and distribution centers. Intensity of use will be indicated by the actual size (as mapped) of the indus- trial/manufacturing facility. 5. PUBLIC FACILITIES/ PUBLIC INSTITUTIONS: This classification recognizes there are a broad range of uses that are either publicly ,owned or pub- licly used, but which do not fit neatly into a single land use cate- gory. Examples are airports, schools, fire stations, churches, sewage treatment plants, sanitary land fills, and certain public buildings such as museums, libraries, post offices, universities, and prisons. Intensity of the uses in this category are indicated by their actual land coverage as mapped. To the extent possible the major public fa- cilities are identified by name, e.g., Fort Christian. Other facili- ties in this category such as schools and churches are identified by a symbol on the maps. 6. RESIDENTIAL: A significant portion of the land of the Virgin Islands is devoted to residential uses. As one would expect, the residential districts in the Zoning Law, when taken together, comprise the largest zoning category. Over sixty per cent of the land area of St.Croix, over eighty per cent of St.Thomas, and over ninety per cent of the non-federal land on St.John is zoned for residential uses. The Zoning Law (Title 29, Chapter 3, Virgin Islands Code) has set the pattern for residential development since its passage in 1973. Because residential uses of the land of the Virgin Islands are of such importance, they have been divided into three subcategories: low, medium, and high density. a. LOW DENSITY RESIDENTIAL: This category reflects the fact that low- density residential development is one of the most significant (in Chapter 2 The Mapping Process terms of land coverage) uses in the Virgin Islands. The pattern of development and density has been guided by the zoning law which de- fines low density as one or two structures per acre. The R-1 low-den- sity residential zoning district allows one structure (which may con- tain two dwelling units) per half-acre. For the purposes of this in- ventory the low-density residential category is defined as having one or two residential structures per acre. In addition, such structures should have no more than two dwelling units and be no more than two stories high. It should be noted that this category is intended to capture, as a mapping unit, those areas that are actually utilized for low-density residential development - not those areas that are zoned for but not yet developed as low-density residential. b. MEDIUM DENSITY RESIDENTIAL: This is an intermediate class of develop- ment that does not correspond directly to the zoning law·s classifica- tions. It is based, however, on actual land use development patterns in the Virgin Islands. In reality it is easier to identify those ar- eas that are clearly being used for low-density or high-density resi- dential uses. This category has been created to capture those areas that, in the Virgin Islands· context, fall somewhere between low and high density residential development. Thus a range of 3-6 residential structures (one or two dwelling units per structure) per acre is con- sidered medium density residential development. Development in areas zoned R-2 (low-density residential) is medium density in this survey because generally one finds four structures per acre in this zone. Multi-family dwellings that have 6-10 dwelling units per structure also fall within this category provided that there are a maximum of two structures per acre). Such multi-family structures generally are IIl ow rise ll (2-3 stories). C. HIGH DENSITY RESIDENTIAL: This category includes IIhigh rise ll (6-8 stories) residential structures as well as various apartment buildings and condominiums that meet the threshold requirements of the category. Therefore any residential development that is either over 6 structures per acre or over 20 dwelling units per acre is high density. This category includes most public housing, various apartment buildings and certain condominiums. As noted earlier, the differences between condominiums and hotels, from a use perspective, is becoming blurred. However, it is possible to identify those condominiums where the ma- jority of use is by the month or year and thus can be considered IIres idential.II Major high density residential uses are identified by name either on the map or in a key code. 7. URBAN: These are areas of the Virgin Islands that are characterized by highly mixed land use activities, very dense development, and a high concentration of population. These areas are, in fact, very difficult to map adequately by simply using air photographs. Therefore, for the purposes of this survey, urban areas will be identified as a single land use category. A determination has been made that the urban boundaries shall be as follows: Charlotte Amalie, census tract bound- aries; Christiansted, census tract boundaries; and Frederiksted, cen- sus tract boundaries. Chapter 2 The Mapping Process 8. WATERFRONT/MARINE: As small remote islands, waterfront and marine uses are of vital importance to the Virgin Islands. This category includes those uses that depend on a location on the shoreline or in the coastal waters. Examples of uses within this category include cruise ship docks, cargo and container ports, oil terminals, marinas, boat- yards, and mooring areas. Intensity of the various uses is indicated by their relative size as mapped. To the extent possible, waterfront and marine uses are identified by name and function. 9. PARKS/ DEDICATED OPEN-SPACE/ RECREATION: This category depicts those uses of land and water that are dedicated to recreational activities or land conservation, or in some instances a combination of the two. Uses in this category include recreational areas such as basketball courts, tennis courts, and baseball fields, beaches, and land and wa- ter within the Territorial and National Park system. Other land and water that is dedicated to open space such as land owned by the Nature Conservancy and other non-profit organizations is also indicated. 10. UNDEVELOPED: After mapping all of the other land uses in the classi- fication system, any remaining land is classified "undeveloped." Since this is a survey of land uses, not land characteristics, the un- developed land will not be labeled as IIwoodland" or "dry scrub vege- tation" or "steep slopes." In addition, the identification of land as "undeveloped" will provide an indication of the remaining amount of land that potentially can be developed. CHAPTER 3 GEOGRAPHIC REFERENCE SYSTEMS AND GEOPOSITIONING INTRODUCTION In a discussion of information management needs for the Virgin Islands Biosphere Re- serve, Potter, Green and Goodwin (1988) stressed the crucial role of a standard geo- graphic reference system. Their major points, which apply equally well to the needs of DPNR, are briefly summarized in this introductory section. The second section discusses mapped grid systems used in the Virgin Islands, and the third section pre- sents recommendations for developing a standardized geographic reference system. Ap- pendix 1 provides some concrete examples of the use of map grids in geopositioning and gives methods of converting between different coordinate systems. A geographic reference is defined as a definite, unambiguous point, line, or enclosed two- or three-dimensional space which is related to some specified spatial coordinate system. A review of prior studies on land use, biology, and environmental monitoring carried out for the V.I. Biosphere Reserve found that much of the data had no reli- able geographic references. There were two main factors which contributed to this problem: (a) A lack of standardization and consistency in notation made it difficult to correlate the results of different studies. Geographic references ranged from geo- graphic coordinates in the "degrees, minutes and seconds II of latitude and longitude, to imprecise "metes and bounds" legal property descriptions, to naming of general ar- eas based on the use of the old Danish system of estates and quarters. (b) Even when a standardized measure such as latitude and longitude was used, in many cases the coordinates given were either incorrect or too imprecise to permit relocation of the exact same spot by future workers. These same kinds of deficiencies are probably just as common in DPNRls data, since the Department does not currently employ a standardized system of geographic refer- encing. However; because the Virgin Islands are characterized by small size, criti- cal resource constraints, and intensive human use, resource managers and planners have a correspondingly greater need for precise and unambiguous geopositioning information. Potter, Green and Goodwin stated that: The promulgation and enforcement of a geographic reference standard for the informa- tion resources ... is the single most significant step that can be taken to improve the long-term effectiveness of [any] information management strategy. Among other advantages, the consistent, universal application of a single geographic reference system will greatly reduce the costs of implementation and data entry for any subse- quent Geographic Information System. Chapter 3 Geographic Reference System single geographic reference system will greatly reduce the costs of imple- mentation and data entry for any subsequent Geographic Information System. MAPPED GRID SYSTEMS IN THE VIRGIN ISLANDS Three different coordinate systems appear on the maps and charts commonly used in the Virgin Islands: the Puerto Rican Coordinate System, Virgin Islands Extension ("PR/VI Coordinate System"); the Universal Transverse Mercator (UTM) System; and the latitude and longitude system. Some of these maps display their coordinate system as a rectangular grid of lines as an aid to accurate geopositioning. Nautical Charts Nautical charts displaying a latitude and longitude grid are available, at a variety of scales, for all coastal and marine areas in the Territory. Nautical charts do not normally show the other coordinate systems. USGS "Quads" and "V.I.P.O. 1963" There are two primary map sources in the Virgin Islands which provide grid systems for terrestrial areas. These are the U. S. Geological Surveyls (USGS) standard 7.5-minute topographic quadrangle maps (latest edition 1982, but based on data from before 1978; scale 1:24,000 or 1" = 2,000 1 ) and the V.I". Planning Officels (V.I.P.O.) large-scale topographic maps produced by Aero Service Corporation (1963; scale 1:2,400 or 1" = 200 1 ). Many other maps in common use by the local planning and permitting agencies are based on enlargements or reductions of these maps with various details added. The 1963 V.I.P.O. maps display a 1,000-foot grid based on the PR/VI Coordinate System. Each sheet of this map covers an area corresponding to one minute of latitude and one minute of longitude. At the latitude of the Virgin Islands, this means that each V.I.P.O. grid cell has an area of ap- proximately 22.96 acres or 9.18 hectares. Coordinates of latitude and longitude are shown at the corners of each map sheet. The 1982 USGS 7.5-minute quadrangles display the Universal Transverse Mercator (UTM) 1,000-meter grid, with each cell covering an area equal to 100 hectares or 247 acres. Cells can be subdivided into 100 subcells; each would then cover 1 hectare or 2.471 acres. The USGS maps also provide tick marks along their margins for latitude/longitude (every 2 minutes 30 seconds of arc) and the PR/VI Coordinate System (every 10,000 feet), thereby facili- tating conversions among all three coordinate systems. It should be noted that while the latitude/longitude grid and the PR/VI Coordinate grid are aligned to true north, the UTM grid has a declination, so that it appears slightly skewed in relation to these. This can be seen by inspection of the USGS map; the amount of the UTM grid declination at the center of each quadrangle map is shown (in both degrees and minutes and in mils) in a diagram at the bottom of each sheet. When measuring angles rela- tive to true north using the USGS map, it is important to measure from the sides of the map, which correspond to meridians of longitude, rather than from the UTM grid lines. Chapter 3 Geographic Reference System Water Resource Maps (CH2M Hill) The 1:12,000 or 1" = 1,0001 scale series of "Water Resources Maps" (1976) and "Sediment Reduction Program Maps" (1979) produced for DPNR by BC&EjCH2M Hill have no actual grid, but carry tick marks for the PRjVI Coordinate System, spaced 10,000 feet apart and 20,000 feet apart, respectively. These maps also show tick marks for latitude and longitude along their margins. WAPA The Virgin Islands Water and Power Authority (VIWAPA) has a computer system running AutoCad software on which it maintains digitized 1" = 200 1 scale planimetric maps, showing buildings and infrastructure interpreted from 1986 aerial photography. The margins of these maps carry tick marks which are labeled "an approximation of the Puerto Rican Coordinate System." There is a Sheet Index Map for each island which shows sheet numbers for the 1" = 200 1 maps and which also displays a grid; however, comparison with the USGS map shows that this grid is actually the UTM 1,000-meter grid, not the PRjVI Coordinate System. Virgin Islands Real Estate Atlas . There is one other "grid system" in the Virgin Islands which is important, although it is not in itself intended to provide accurate geopositioning. This is the grid system employed on the Virgin Islands Real Estate Atlas, Map and Ownership Volume published by Real Estate Data, Inc. The grid, with numbered cells, is displayed on a set of small "Index Maps" (no scale shown) found in the front of the Real Estate Atlases. The lines of this grid correspond to the edges of each sheet of the 1963 V.I.P.O. topographic map, and the numbers in the grid1s cells correspond to the sheet numbers of the V.I.P.O. map. Each map sheet also contains a grid reference to sectors within that map. Parcels are numbered as a sub-feature of the map-and-grid numbers. The importance of the V.I.P.O. Index Map grid is mainly that it allows one to correlate the proper sheet of the V.I.P.O. maps with the V.I. Tax Assessor1s Plot Maps. When a particular location can be identified on the V.I.P.O. map, it is then possible to find the correct Tax Assessor1s Plot Map in the Real Estate Atlas by utilizing the Index Map grid along with a system of subgrids. This leads via a plot code to land ownership information in the records of the Tax Assessor1s Office. RECOMMENDA TI ONS DPNR should adopt a standard geographic reference system according to the following specifications (based on Potter, Green and Goodwin, 1988): (1) The Virgin Islands Extension of the Puerto Rican grid ("PRjVI Coordinate System"), expressed in feet (i.e., 10,000IS), should be adopted as the geographic reference standard for DPNR and all other resource management and planning agencies in the Virgin Islands. Most land survey work for planning, land use mapping, project feasibility studies, permitting and development is Chapter 3 Geographic Reference System based on either the Puerto Rican Coordinate System/ Virgin Islands Extension or the Universal Transverse Mercator (UTM) system rather than latitude and longitude. All terrestrial mapping work should henceforth use the PR/VI coordinate system as the reference standard for geopositioning. (2) The geographic references for coastal and marine mapping may also employ the system of latitude and longitude coordinates but each map should show at least three geographic reference coordinates for the PR/VI Coordinate System. Latitude and longitude is an important secondary coordinate system for the Virgin Islands because of the area's general maritime orientation, the critical importance of its marine and coastal resources and the many re- cent surveys which establish easily verified "ground truth" points along the coastlines. Precise location of virtually all marine resources or sampling sites in the Territory can be established by triangulation and interpolation from known coastal and offshore features and US Geological Survey benchmarks. In addition, the latitude/longitude coordinate system is the only one which is displayed in common among all the important maps and charts used in the Territory. (3) It should be DPNR's policy to require and enforce the inclusion of PR/VI Coordinate System marks on all maps which the Department accepts for environmental assessment reports, site plans, biological monitoring studies, etc. One way to enforce this is to require all map submissions to DPNR to use map bases purchased from the Department, at the appropriate scale. In addition, on each map at least one pair of geographic references should be labeled with latitude and longitude and UTM coordinates, to facilitate conversions to the other coordinate systems. (4) DPNR should maintain geographically referenced data in some type of computerized database (not necessarily a full GIS). Geographic references should be entered exclusively in terms of the PR/VI Coordinate System. Conversion of coordinates to other coordinate systems can be done automatically by a small, stand-alone computer program; software to accomplish such conversions is readily available. DPNR may permit applicants to provide geographic references in the other two coordinate systems, but for purposes of quality control references should be checked by DPNR's computer before storage in the database. Existing geographic references for important previous studies and mapped information in DPNR's files should be recalibrated to the PR/VI coordinate system as a necessary long-term revision in the Department's information re- sources. (5) DPNR should maintain a large-scale display map (1" = 1,000') with detailed overlay grids calibrated to the PR/VI Coordinate System, and with tick marks for latitude and longitude and the UTM system. This should be used to assist staff and permit applicants to estimate geographic reference coordinates. (6) Reference coordinates for mapped features should follow accepted mapping standards. In general, lines should be identified by the midpoint on the line, and polygons should be identified by a centroid, within the polygon. Chapter 3 Geographic Reference System (7) When reporting positions that have been determined by standard surveying methods, the precision of the geographic references should be keyed to the size of the unit being reported, with smaller units requiring more precise references. When making imprecise position estimates (e.g., measurements made with a ruler on a map), the reported precision of the geographic refer- ences should be consistent with the errors inherent in the scale of the map and the method of measurement. (8) There are existing Federal cartographic standards for map accuracy which should be adopted by DPNR for all mapping projects, e.g., the National Bureau of Standards (US Dept. of Commerce) FIPS Publication #70, or the standards emerging from the Federal Interagency Coordinating Committee for Digital Cartography (FICCDC). APPENDIX TO CHAPTER 3: EXAMPLES OF GEOPOSITIONING AND CONVERTING BETWEEN COORDINATE SYSTEMS PROBLEM A: Using the most recent USGS map of St. Croix, find the WIVI radio tower at Fort Louise Augusta, northeast of Christiansted, and designate it as a geographic reference point, giving coordinates in the UTM, PR/VI and lati- tude/longitude systems. METHOD: Looking at the top margin of the map, notice that the vertical UTM grid lines are labeled 3 10 000 E, 3 11, etc., and that they increase in 1,000-meter increments from west to east. Locate the nearest vertical UTM grid line to the west of the radio tower, this is the one labeled 3 20. The grid line 1,000 meters to the east of this is labeled 3 21, and the tower lies between these. Using a magnifying glass and the finest (60 divi- sions/inch) scale on a triangular architect's rule, count the number of divisions between these two vertical grid lines; you should get 99. Then measure from the 3 20 line east to the radio tower; you should get 39 divi- sions. Now you have to set up a proportion to calculate the number of meters east of the 3 20 grid line that the tower is located. Since the 99 divisions on the ruler represent 1000 meters on the map, the number of meters that the tower lies east of the grid line is given by 39/99 x 1000 meters, or about 394 meters. 20,000 meters plus 394 meters equals 20,394 meters; therefore the vertical grid coordinate of the tower is 3 20 394 E. Now look at the right side of the map and find the nearest horizontal grid line to the south of the station; this is the 19 63 N (unlabeled in this instance). The tower lies between this line and the next one to the north, the 19 64 line. Measure from the 19 63 line to the radio tower; you should get about 98 units. By setting up a similar proportion, 98/99 x 1000 meters, we can calculate that this represents about 990 meters. 63,000 meters plus 990 meters equals 63,990 meters, hence the horizontal grid coordinate is 19 63 990 N. Now look at the left-hand bottom margin and right-hand top margin of the map, and find the tick marks on the outside of the margins labeled 1 070 000 FEET and 1 110 000 FEET, respectively. There are also ticks on the outside of the Appendix to Chapter 3 Geographic Reference System top right-hand margin and bottom left-hand margin, respectively labeled 90 000 FEET and 50 000 FEET. These ticks represent the coordinates of the Puerto Rican Coordinate System, Puerto Rico, St. Croix zone (or Puerto Rican Coordinate System, Virgin Islands extension for St. Thomas and St. John). The coordinates increase from west to east and south to north in increments of 10,000 feet. Using the 60-scale rule, count the number of divisions between the 70,000 feet and the 80,000 feet tick marks on the right-hand map margin; you should get 301 units. Using a long ruler, draw in a line between the 70,000 feet marks on each side of the map. Now measure perpendicularly from this line north to the radio tower; you should get 148 units. Setting up a proportion, 148/301 x 10,000 feet equals 4917 feet. Then 70,000 feet plus 4,917 feet equals 74, 917 feet, which is the horizontal grid coordinate. Count the number of divisions between the 1 100 000 FEET tick and the 1 100 000 FEET tick on the upper margin; you should get 299. Then draw in a line between the 1 100 000 feet ticks at the top and bottom margins. Measure from the 1 100 000 feet line east to the radio tower; you should get 155 units. Then 155/299 x 10,000 feet equals 5184 feet, and 1,100,000 feet plus 5184 feet equals 1,105,184 feet, which is the vertical grid coordinate. Now look at the top and bottom right-hand corners of the same map, respectively labeled 17 degrees 48 minutes and 17 degrees 40 minutes 30 seconds in the latitude/longitude system. These coordinates represent a difference of 7.5 minutes of north latitude between the top and bottom edges of the map. Tick marks are shown every 2130" of latitude, on the inside of the right-hand margin. Thus the next latitude mark up from the bottom of the map is the 43 1 tick just inland from the southern coastline, and the one after that is the 45 130" tick just inland from the northern coastline. Similarly, there is a 7.5 minute difference in west longitude between the right and left map margins, respectively labeled 64 degrees 40 minutes 30 seconds and 64 degrees 48 minutes. Using the 60 divisions/inch scale, count the number of divisions between the 17 degrees 43 minutes and the 17 degrees 45 minutes 30 seconds latitude ticks; you should get 455. Now using a long ruler, draw in the latitude line across the map between the 43" ticks on each side. Measure from this 43 1 line to the radio tower; you should get 435. Because there are 60 seconds in every minute of latitude, 2130" equals 150 seconds. Since the 455 ruler divisions represent 150 seconds of latitude on the map, the tower lies 435/455 x 150 or about 143 seconds north of the 43 1 line. Converting back to minutes and seconds, this gives 2123". The latitude of the tower is therefore 17 degrees 45 minutes 23 seconds N, which can be expressed in minutes and decimals as 17 degrees 45.4 minutes N. Appendix to Chapter 3 Geographic Reference System Now count the number of divisions between the right-hand margin of the map (64 degrees 40 minutes 30 seconds) and the 64 degrees 43 minutes tick; you should get 434. Measure from the right-hand margin to the radio tower, you should get 197. Since 434 divisions represent the 150 seconds of longitude, the tower lies 197/434 x 150 or about 68 seconds west of the 40 130 11 line. Converting back to minutes and seconds, this gives 1108 11 . The longitude of the tower is 64 degrees 41 minutes 38 seconds W, or 64 degrees 41.6 minutes W in minutes and decimals. SOLUTION: Since the convention is to name the horizontal coordinate first, the geographic reference of the tower is: 19 63 990 N, 3 20 394 E in the UTM system; 74 917 FEET, 1 105 184 FEET in the PR/VI Coordinate System; and 17 degrees 45.4 minutes N, 64 degrees 41.6 minutes W in the latitude/longitude system. Because we are measuring to the nearest 1/60th of an inch, and 1 inch equals 2000 feet (606.1 meters, or 19.7 seconds of latitude) on the map, this means that the error in our measurements is plus or minus about 33 feet, or ten meters, or 0.3 seconds of latitude. PROBLEM B: Using the same USGS map as above, designate the point where UTM grid line 3 21 intersects the shoreline of Beauregard Bay to the east of the WIVI radio tower as a geographic reference. Then transfer this geographic reference point to the V.I.P.O. 111 = 200 1 scale map, which has no UTM grid, and give its coordinates in the latitude/longitude system. METHOD: Since both maps have the latitude/longitude and PR/VI coordinate systems, the geographic reference could be transferred by computing its coordinates in either of these systems on the USGS map using the methods of Problem A, and then repeating the process on the V.I.P.O. map. However, there are other methods which depend on the fact that there is a nearby point whose coordinates are known and which is common to both maps, namely the WIVI radio tower. We can use this common geographic reference to fix the position of other points on the V.I.P.O. map. Since the scale of the two maps is different, we must first compute a scale factor which we can use to convert measurements from one map to the other. In this case we know that one inch on the USGS map equals 2,000 feet on the ground, while one inch on the V.I.P.O. map equals 200 feet. Therefore the scale of the V.I.P.O. map is ten times as large as that of the USGS map, and the scale factor is 10. If we didnlt already know the map scales, or if we wanted to check on their accuracy, we could compute the scale factor by tak- ing measurements between two identical points on each map and dividing the larger measurement by the smaller. Appendix to Chapter 3 Geographic Reference System Draw a line on the USGS map from the WIVI tower to the intersection point of UTM grid line 3 21 with the shoreline of Beauregard Bay. Use the 60 scale on an architect's rule to accurately measure the length of the line between the radio tower and the point on the shoreline; you should get 59 divisions. Convert the length of the line between the radio tower and the shoreline point (on the USGS map) to the scale of the V.I.P.O. map by multiplying by the scale factor; therefore we have a length of 59 x 10 = 590 divisions on the 60 scale rule. Find the radio tower on the V.I.P.O. map and set the zero mark of the rule on it. Then, holding the zero mark on the tower, mark the point at which the 590th division intersects the shoreline to the east of the tower. This is the location of UTM grid line 3 21, our geographic reference. SOLUTION: Once the point is located, we can easily figure out its latitude/longitude coordinates by the methods of Problem A. They are approximately: 17 degrees 45.4 minutes N, 64 degrees 41.3 minutes W. [Note that in this last example the geographic reference point was easy to transfer because it was chosen as the intersection of the shoreline and the UTM grid line. It would not be so simple if the geographic reference point were, say, a house which was shown on the USGS map but not on the other map. Then it would be necessary to establish bearings to the house from two or three other points common to the two maps. The bearings could be found on the USGS map by using a protractor to measure the angles between true north (indicated by the longitude meridian at the margin of the map) and the lines drawn from the common points. Then lines with the identical bearings can be laid off from the same points on the other map. If this is done accurately the lines should intersect at a point or form a small triangle which repre- sents the location of the house. Yachtsmen may also be familiar with a third method of solving this problem, using a three arm protractor. [Another, less accurate, method can be used if only a single point is common to both maps. Draw a line between the point and the house on the USGS map, then measure the bearing of this line as well as the distance between the common point and the house. Layoff the same bearing line on the V.I.P.O. map and then measure off the appropriate scaled-up distance along the line to determine the location of the house.] 17 CHAPTER 4 LAND USE CATEGORIES The land use classification scheme developed in this section has been designed to provide the Department with a more elaborate classification matrix than can be supported by the simple use of aerial photo interpretation and occasional ground truth checking. The detailed land use classifications presented in this chapter would be derived from a variety of records available to DPNR, either as part of the permit application process or as other publicly available or readily observable behavior. This classification scheme, or any reasonable modification of it, can be easily implemented in a land use database system, such as that proposed in Chapter 5, below. Decision Rules Early in the process of defining detailed land use classifications and intensities, a series of rules were developed to help guide the discussions. These rules were presented for the Department's approval, along with the original proposed land use classes. They are: 1. Use existing Ten Classes. That is, the detailed classifications should be a refinement and a detailing of the mapped classifications, rather than an altogether different set of criteria. 2. Sub-categories must be mutually exclusive and exhaustive. By definition, each set of sub-categories must add up to 100% of each major category. 3. Prefer Zoning-based sub-categories for decision-making relevance. Stated as a preference, rather than an imperative, this rule emphasizes the desire to focus the classification system on relevant decisions. 4. Prefer categories based on existing government record systems, such as business licenses or hotel certificates. The Land Use Classification system cannot undertake to reform all of the data systems of Government, nor can it propose to unilaterally impose new standards (such as new hotel/guest house categories) on the well-es- tablished practices of other agencies. Chapter 4 Land Use Categories 5. Prefer photogrammetrically verifiable characteristics. In addition, indicators of intensity should be factors which are generally easily observable, or available from public records (such as permit or tax abatement applications). 6. Within each Class, identify the relative intensity of use of each sub-category. The intention of this exercise is to provide a rough yardstick or red flag mechanism to alert planners and permit givers to proposed uses which might have especially adverse effects. 7. For management purposes, prefer to keep total number of categories to less than 100. As currently defined, this list contains 109 detailed land use classifications. CLASS Sub-category 1. Agriculture lA Pasture lB Garden Crops lC Tree Crops iD Livestock iE Aquaculture iF Animal Husbandry iD Other Agriculture Intensity LO---MED--HI x x X X X X X Comment: Intensity based predominantly on negative en- vironmental impacts, especially ' water and chemical usage. Positive environmental effects can be experienced from well-managed pasture and aquacul ture projects. 2. Resort/Hotel 2A Hotels 2B Smaller Hotels 2C Condos and Apartments 2D Guest House and Other X x X x Cooment: categories based on resort and hotel registrations with the Department of Economic Development and Aqr icul ture. other includes villas, cottages and campsites. Intensity based on energy use, employment, and gros9 revenues per square foot of structure (availabl e from IOC). Generation of sewage, solld waste, and vehicular traffic are additional negative factors contributinq to high intensity use. [Murphy's Law Factor: On-si te power generation, fresh water production, sewage tr'eatment and solid waste disposal technologies should be evaluated in the context of a high propensity for mechanical systems to break.] 3. Retail/Commercial 3A Retail Sales and Restaurants 3B Offices 3C Movie Theaters and Drive-ins 3D Gas Stations 3E Mixed Use Shopping/Commercial Parks 3F Other Commercial X X X X X x Corrment: Intensity based on economic turnover, energy use especially for air conditioning -- and pollution poten- tial. Most significant polluting activities include generation of sewage, solid waste, and vehicular traffic. Chapter 4 CLASS Sub-category Intensity LO---MED--HI 4. Industrial/Manufacturing 5. 4A Light Manufacturing 4Aa Watch Manufacturing 4Ab Rum Refining 4Ac Textile Processing 4Ad Other Light Manufacturing x x x x 4B Heavy Manufacturing 4Ba Oil Refining x X 4Bb Other Basic Industries 4Bc Other Heavy Manufacturing 4C Wholesale and Warehousing 4D Mixed Use Industrial Parks 4E Other Industrial Use x X X X Comment: Intensity based on capital and energy intensity of the enterprise, and pollution potential, especially for shorelines and irreplaceable resources such as groundwater. For example, if an oil refinery uses ten times more energy in the oil refining process than all other energy consumers in the territory combined, it should be recognized as a high intensity land use. Then add in the costs of a spill from one of the tankers which bring oil from Valdez. Public Facilities 5A 5B 5C 5D 5E 5F 5G 5H 51 5J Schools 5Aa Private Elementary & 2ndary X 5Ab Public Elementary & 2ndary X 5Ac Private Post-Secondary School X 5Ad Public Post-Secondary school X 5Ae Other School X Airports X Post Offices X Churches X Fire & PolicD Stations X Military or Auxiliary Facility X Museums & Libraries X Landfills X Other Federal Public Facility X Other Non-Federal Public Facilities X Comment: Intensity based on extent of land alteration, pollution potential, stimulus to vehicular traffic and economic impact of use. Chapter 4 CLASS Sub-category Intensity LO---MED--HI 6. Residential 7. 6A Low Density Residential 6Aa 6Ab Low Density Single Family Low Density Two-Family x X Comment: Based on one or two dwelling units per structure, and less than three structures per acre. 6B Medium Density 6C 6Ba 6Bb 6Bc Medium Density Single Family Medium Density Two Family Medium Density Multi-Family X X X Comment: Assumes three to six single family structures (containing one or two dwelling units) per acrej and multi-family structures containing six to ten dwelling units per structure and not more than'twq" structures per acre. High Density 6Ca High Density Single Family X 6Cb High Density Two Family X 6Cc High Density Multi-Family X Cornment: Assumes more than six single or three duplex family structures per acrej or more than ,io, dwelling units~ in multi-family structures. Comment: Residential intensities are directly proportional to the number of inhabitants supported per land area unit. Urban 7A Urban Resort 7Aa Urban Resort Hotel 7Ab Urban Resort Small Hotel 7Ac Urban Resort Condo & Apts. 7Ad Urban Resort Other 7B Urban Retail/Commercial 7Ba Urban Retail and Restaurants 7Bb Urban Office Structures 7Bc Urban Mixed Commercial Use 7Bd Urban Gas Stations & Parking X X Lots X x X X X X Chapter 4 CLASS Sub-category Intensity LO---MED--HI 7C Urban Industrial/Manufacturing 7Ca 7Cb 7Cc Urban Light Industry Urban Wholesale & Warehouse Urban Other Industry 7D Urban Public Facilities 7Da 7Db 7Dc 7Dd 7De 7Df 7Dg 7Dh 7Di 7Dj Urban Schools Urban Airports Urban Post Offices Urban Churches X Urban Fire & Police Stations X Urban Military Facility Urban Museums & Libraries X Urban Landfills Urban Other Federal Facilities Urban Other Non-Federal Facilities 7E Urban Residential 7Ea 7Eb 7Ec Urban Low Density Residential Urban Medium Density Res. Urban High Density Residential 7F Urban Waterfront/Marine 7Fa 7Fb Urban Waterfront Industrial Urban Waterfront Pleasure X X 7G Urban Parks, Open Space, & Recreation X 7H Urban Undeveloped X X X X X X X X X X X X X X 7I Other Urban Use X S. Waterfront/Marine SA Waterfront BAa SAb BAc BAd BAe BAf BAg Cruiseship Piers & Docks Cargo Piers & Docks Marinas Marine Craft Sales/Service Waterfront Recreation Non-Commercial Waterfront Use Other Waterfront Uses X X X X X X X Con1nent: Intensity based on scale of shoreline alteration for the structure, and the potential for chronic or catastrophic pollution incidents. Chapter 4 CLASS Sub-category 8B Marine Uses Mooring Areas (permitted) Anchoring Areas (not permitted) Channels (buoyed and not) Non-Commercial Transportation Intensity LO---MED--HI X X X X 8Ba 8Bb 8Bc 8Bd 8Be 8Bf Fishing and Trapping X Other Marine Uses X 9. Parks/Open Space/Recreation 9A Active Recreation Areas 9Aa Baseball Diamonds 9Ab Tennis Courts 9Ac Football/Soccer/Rugby Fields 9Ad Cricket Pitches 9Ae Other Racket Courts 9Af Track and Field Play Areas 9Ag Multipurpose Playing Fields 9Ah Other Active Recreation Areas 9B Stadiums & Racetracks 9C National Parks 9C Territorial Parks 9D Recreational Beaches 9E Protected Natural Areas 9F Other Parks and Open Spaces X X X X X X X X X X X X X X Comment: Intensity based on environmental impact of the use (e. g., tennis courts are rrore defacing of the landscape than cricket pitches) and its ancillary facilities, such as parking lots, high intensity lighting, etc .. 10. Undeveloped lOA Mangroves, Salt Ponds & Wetlands lOB Barren Land & Rocky Shorelines lOC Scrub and Grasslands lOD Other Undeveloped Land Chapter 4 X X X X Page ··24 CHAPTER 5 LAND USE DATABASE DESIGN This discussion of a database design for land use records for the Virgin Is- lands Department of Planning and Natural Resources includes four elements: (1) a brief survey of existing record systems; (2) a proposal for a land use database; (3) file and record descriptions for that proposed system in dBASE III Plus; and (4) a new copy of a dBASE ill Plus work-alike (dBXL, by WordTech Systems), ass~ting that DPNR can run the proposed system on any com- puter without infringing on copyright and use licenses. [When DPNR begins to develop regular reporting systems based on the land use database, we suggest the Department acquire one of the stand-alone report writers, rather than relying on the cumbersome report writing functions of dBASE and its clones. Island Resources Foundation has been using R&R Re- lational Report Writer with success in recent months, but there are a number of similar systems on the market at prices around $150.J Existing Land Use Record Systems There are three major sources of parcel-based land use information in the Virgin Islands. The first is the Department of Planning and Natural Re- sources, which has land use information at both the area, and the parcel level. At the area level, the Department has a land use map (not referenced to parcels) which illustrates general patterns of land use in 1978. This project is producing an update of the earlier map -- at somewhat greater standards of accuracy, but not referenced to parcels -- for la~d uses seen in 1988-89. DPNR Parcel Record Systems Offices generating and tracking land use information in the Department of Planning and Natural Resources are widely dispersed in function and geo- graphic location. Multiple, non-compatible, and non-integrated record sys- tems are maintained independently by DPNR agencies with such diverse mandates as: the Virgin Islands ' Cadastre Office; permitting functions for land development and coastal zone alter- ation; comprehensive territory-wide planning functions; and, the issuance and policing of all fishing and hunting licenses. Ch apter 5 Land Use Database Except for the Cadastre Office, DPNR offices make no attempt to establish or maintain a comprehensive (i.e., involving all parcels) database of all Virgin Island land units or owners. In general, each office tracks only those units which are licensed or which apply for permits from that office. In addition, it is unclear, for example, whether a property which receives a special land use permit from the Coastal Zone Commission is perpetually subject to monitoring, and whether the record of that permit and its attendant use con- ditions are actively maintained by DPNR. [This section could be expanded, depending on information from permit appli- cations and enforcement forms and records used by the Department. But this information was not available.] Recorder of Deeds A second major source of definitive land use information is the Recorder of Deeds in the Office of the Lieutenant Governor. Records in the Office of the Recorder of Deeds are manual -- most record en- tries are handwritten. The unique property numbering system used by the Recorder of Deeds is based on the historic Estate and Quarter system of the Virgin Islands. This system is referred to as the "legal description" of the property. The cross reference with the unique "map and grid" parcel number used by the Tax Assessor1s Office seems to be maintained by the Tax Assessor -- not the Recorder of Deeds. Tax Assessor1s Office The Office of the Tax Assessor is also part of the Office of the Lieutenant Governor. The St. Thomas Office of the Tax Assessor has recently been moved to the Nisky Center office complex, and will be close to many of the plan- ning, permitting and executive offices of the Department of Planning and Nat- ural Resources. The Tax Assessor maintains a computerized record for every parcel (and condo- minium) in the Virgin Islands. The record system is well established and must be considered among the most reliable in the Virgin Islands, since it is used to generate the annual real estate tax bills. Any serious attempt by the DPNR to build a comprehensive parcel and land use record system should start from, and develop update procedures with, the Tax Assessor1s system. The current Tax Assessor system is operated as a unit record file, processed in batch mode on the central computer system of the Department of Finance. Modifying and transferring this system to an on-line, networked database ac- cessible to both the Tax Assessor1s Office and DPNR should generate substan- tial benefits for both agencies. The map and grid system which is used by the Tax Assessor to assign unique parcel numbers is based on large scale maps (apparently larger than 1:1000) p~ovided to the Tax Assessor (s~ecifically, in St. Thomas, Mr. Martin) by the Cadastre Office of DPNR. The exact scale and accuracy standards of these maps are not known. Chapter 5 Land Use Database Other Agencies Other important sources of information about land use which need to be con- sidered and eventually incorporated into a comprehensive DPNR land use database include: Detailed farm plans maintained by the delegate agency of the USDA Soil Con- servation Service. Business licenses issued by the Department of Licensing and Consumer Affairs. Tax incentives awarded by the Industrial Development Commission and other programs of the Department of Economic Development and Agriculture. National Park use and management plans for park land on St. John, Buck Island and Hassel Island, as well as administrative properties in St. Thomas and St. Croix. Power and Water networks serviced by WAPA. Various responsibilities and programs of the Department of Public Works, in- cluding highways and planned highways, earth change permits, water well permits, and septic tank and leach bed permits. Hazardous materials use based on reports filed with the Fire Service or VITEMA. Housing projects and infrastructure maintained by the Virgin Islands Housing Authority. Facilities and activities of both the marine and air divisions of the Virgin Islands Port Authority. Hydrology, including flood zones, erosion zones, and aquifer recharge zones, from the Water Resources Office of the Caribbean Research Insti- tute. Monumentation -- that is, the construction, recording and maintenance of spe- cific geodetic reference points -- by the USGS, the National Oceanic and Atmospheric Administration, the Federal Aviation Au- thority, the Army Corps of Engineers, and the Department of the Navy. Other information relevant to land use, but not necessarily the specific responsibility of anyone Virgin Islands government office includes: geol- ogy; soil classifications; seismic conditions; and topography. Database Design A detailed needs analysis of the Department of Planning and Natural Resources was not conducted for this database design. Furthermore, no effort was made to seek detailed verification of data and other resources potentially avail- able to the Department (e.g., data in the Tax Assessor1s files was not Chapter 5 Land Use Database analyzed). This design is based on accepted land use database designs used in a variety of local government settings, extensively modified to reflect specific Virgin Islands opportunities, constraints and resources. As a generic model this proposed system is comprehensive to a fault. That is, it might require converting and updating many variables which are not critical to the decision making needs of the Department. As submitted, this system design could be made operational within two or three days. (The only task remaining is that of assigning specific codes to variables in the database -- e.g., developing a series of reasonable three digit codes for Virgin Islands soil types.) This approach is not recom- mended. Instead, we advise conducting an in-house DPNR needs analysis, hew- ing closely to the specification of information critical to the actual deci- sion making processes of the Department. When this self-examination is com- plete, the design submitted with this report should be studied rigorously to assure the need for each data item proposed. This two-stage adoption of the proposed database design should permit the Department to eliminate many vari- ables from the submitted design, saving considerable money in both initial system construction and routine database maintenance. We are confident the current staff and leadership of the Department are well qualified to carry out such a study. Land Record User Matrix As a preliminary step in the process of gauging user needs -- both inside and outside the Department -- we offer the tables presented on the two pages fol- lowing. First is a blank matrix of potential Virgin Islands land record users and major information sources with geographic references. The second table is a very preliminary attempt to fill in such a matrix. Given the very quick process of development, and the absence of any substan- tial feedback or reality testing, this second table should not be assumed to be authoritative -- for now, it simply serves as an interesting, if somewhat speculative, illustration of the kinds of information which a user needs study should begin to consider. It will be noted that the table groups DPNR users and generators of land record information in the upper left corner of the matrix. The filled-in ex- ample demonstrates two important facts, related to the future of planning DPNR land record and geographic information systems: 1.) DPNR accounts for a large portion of the total users and generators of map and land record information. This implies that even without the active financial Pftrticipation of other agencies, DPNR should be able to develop persuasive cost-effectiveness measures for many land record automation proce- dures. 2.) There are a large number of important users of land record information generated by DPNR. This implies that political and financial support (e.g., by user fees) should not be too difficult to develop. LAND RECORD USERS MATRIX [Blank Form for Data Collection] /015 DPNr. Aetivi +'':1 Ctrs <Qt1B t'$) tHee fRel) Enfo RIJVC Insp Cad,} CapP CQ"" rAPI L£f'l Must' LSGA EnvP loth~ 6ov.t-~'t Flget'\ci~z : IJtili tie~ lnd (I I Land Use R~cord: ----- 4~11 49124913 4'324 4934 4935 ~942 4'351 4952 4953 4963 4966 43131 : L tGv Fir~ 'lTMP Dec L3:CA BIP. Pole Ed Hlth W~ Park EcDv \.IAPA 'JlPA VfLC R.;>al [1 .. 1.' '.II liP 5K ; AgEH -,- J Pan:~l ID .and Location · · • · · · · • · • • • Pt"opp.rt!J (lun~rsnip • " I' Lat~d U$!? · · • • • • Zoning • • • • kiotural J!~source$ • • • -1 l1iner al ke-solln:~s • • Land SIJtVE'~S • • · • · • • • . Coastal Zone htp.act • · • • Other Enuiro~tal hlpact · · • J Histori~ Pre~rv~tio~ 1 • • • • - ! Flood Plains I • • • • BuildiYlg Code Vi ol.ati ems · · • · • I ~11 I Ylfon.ation · · • • • • • • Topogr~&eology • · • • · • • • .J Air Qu.ali bJ l1ordtoring • • • • Llater Quality Monitoring · · • · • • • • • • 6roun~-!teor I'kmitoring • • • • · · • • Recrl?atiof'l · · • • ArchP.oll)9\I • • • • · -1 Other P.irc:el CharacUri.stics • · • • • • Earth Chartge Pendts · · • " • " Forest Cover · " . • " · · -- "I Uater Tral'lSport · · · " • j Koadz · · · · • • · · · 0 · • • • Se~age Pl~ts and l1aitlS · · • · · • • IUt~ ~ins · · · · • • • · • · · " • • Ph~ ~ Cable TV Lines • 0 • • • · Ghutz, Cul~rts ~ Bridges · • • · · • . • Public Faciliti~z .- 0 • · · • • • • " j Air ~nd Other rr.am:port · • • . .. Boat R€',~s.tr.ti on · · • · · • · • Fish Catch Data • · • " • • · · • Fish Li '-!"rISes · · · · • · · · 1 Hunting Data · · • · • · • • • · • Flood Plains • · · • " l De~d$, LiE>ns and Ease~l?nts • 0 • • ft.isesSMl?nt Dah · · " • • " Soilz · · • · • • Fant Pl.:mning · · " · • PropertJJ Sales: • · · · • • • Buildin9s &; Il'lprovl?l'telits • • · · Epi deKe"l clgy • · • • J . 0 • Voting Ois:tricts --I J • · J • School Di:.tricts · · · · l1olM1erlt Location» Cel'lslJs Data • j Cril'les · • • Fires I · · t Al'lbulan.:e Calls .-. · · l 6 = Generator of Data U = User of Data B = Both ? = Que-stiotliTo Be Oet~ndned 1 J -I I j I 1 l .J J 1 I j J I J VI LAND RECORD USERS Preliminary Assessment -- April l 1989 OO~R Iktivi ty Ctrs (OHP t'5) E:-II?C n-=-v Em", ~:l?vC In.p Cad~ CapP Cm"P FllPl LRPl t1useo lSCA EnvF' I Otheot- I,~rn~~tlt Ag~ie-~: Late! Uzt hc:crd: ------ "-'11 4912 4913 ~924 4934 4935 4~42 4'?51 49:2 4'35;, 4%3 49&t 'r~81 :Lt~ Fir<:- I.'rNA Hee: L3:CA BI~ PI,lc P~'!'l 10 -3nd Locatiorl Pnpl?f"~.J wnc:r#.ri p L4Id Use Zorirl9 N~r~l ~~$our,-~s l1ite!-31 ~e<sourc~.; LiKf Sf.IrVSolJS CCtital Z~ I~tpact Otl~r En~iron"~ntal I~pact Hiltoric Preserl)~tion Flxod Phi ns Building (;Qde lJichtions W.ll Info!"'Hation T~graph~Geoloqy Ail' w.ali bJ ~.onitoring U~r ~ua1it~ ""!'Ii tori fig Gnrumd~at~ Moni+.ori~g Rweation Anh~logy ~er Parcel Characbr1s:ticJ Earth Chanqt? PM' .. i ts Forest Co~'P.r Um Tra!'lSport ROfds SelagE' Plants (~ Mal. ns j./~r PhP'Jts ~ rhi ns Phml? ~ Gable TV Li!'les: Gmt.;, Culverts &. Bridge! PuUic Facilities Ail- and O~r franzDOrt BOft Rp.gi rtr ati on Fith Cater. Data Filh Liceonses Huttiflg Data FIxod Phi ns: DE!ds, Li S'I'I$ .atld Ease?'!€'nts A~s$~E1)t Dah So.il.s Fit!'! Pl anni ng Pl"'lpt?t'ty ;.ales Buildit'!gs i I~provO?"'l?nts Ep.i:il?\'1eol (u7'J Vcdng Oistricts Sdool Di rtri cts NOII.M?t\t Locati. arts emus Oata . Cr.l1~ Fi~z At$.J1.ar!c~ Call s U B U u U B U u u u u U B B IJ u u u u u u u u u u IJ u U B u u U B B 8 B G B U U ? B U U U U 8 B U B. U B. 8. B U 8 U U U U U u . . u u u u u • • u u U IJ IJ IJ B B U '/ U B B B IJ B f.I • • IJ IJ IJ U B B B B B ? B u G u • • u • ~ " . ? r • • II e = Generator of Data IJ = User of Data B = Both u U B IJ B i U ? IJ • • • • l' ?' ? ? U u ? u l' ? l' • • G B U U B ? G ? u U 8 U? • B B U 1" = '-!uesti (mIT 0 Be D~t~rKi neod u u u u U B lJ U U B U U B U IJ u u U IJ u • • U IJ IJ U U U B B U u u u u u u u u u u u B IJ IJ u IJ U !J • u U U U U U B B U U B u u u u u u u u u u u u u u u u u u U B U U u u u u u U B • u u u u u B U U U U U u u u u u u u l' " U 6 u e 6 IJ IJ I) u r, IJ IJ • • • u U !J u B u . . u u B IJ B B B B B B B B 9 ? ? B B '! U ? 1.1 U B B B B u • • u u u • B G n u B ? u B B G B B B u u • • IJ ? u u u u u u u B B • • • • u u ij U !J B '( u u u u u u U ? B U U l' B U U U u u u t U u u U • u u • u u B u • • IJ U B IJ IJ ? " B U ? IJ ? U IJ IJ IJ U • u u B . " B U • • .. " ? B u u u u • • u u u IJ 8 Ed U . . U . .. u B B B u Hlth • LI IJ U IJ U U II u u B B B JJ B l!tUitil!!'s and 0 DPLI ParI< E:Ov !.lAPA VIPA IJrLC ?~dl u B U ? B U B B B e U B B e ,.. ! B B B B B B B B U B U B B U IJ U U • • • • I) IJ IJ IJ IJ IJ u B B IJ IJ 1) B IJ IJ · . p, IJ • B U 1I ~ .' E: E: B U u • u IJ U B u • • B B 1I B' B' U U II U 8' B E: U U U 1I 1I B S B B u u B IJ U B U u u . .. IJ U .. 1I u u u B U U U B B U U U B B • • • IJ • • • B u u B B U u u u u u u • • u u u '-' u 1I 1I U u u u u ij u B U G U u u u Df?1.' VINf' 1.1 IJ B B IJ U B G B B 8 U U IJ B U U B B e u IJ B B IJ B B B B U B B tJ B B U U G U U B U u U B . . IJ U U 5C;; A-;EH B B U B • • B B . . B B U U u u · . u U U B U u B · . u U u u .. .. u u u U B U B U U !J U IJ . . Chapter 5 Land Use Database Parcel-based Database Design The proposed database for land uses in the Virgin Islands is designed to be a parcel-based system, which should be closely tied to the records of the Tax Assessor's Office to keep pace with the high rate of property transactions experienced in the Virgin Islands. Integration with fast-changing Tax Asses- sor records provides the advantage of a more up-to-date database. Full operation of a land use database by the Department is estimated to take at least a year to implement. The advantage of refining such a system is that this effort replicates (in miniature)) most of the steps the Department will have to follow to implement a geographic information system (GIS). In fact, a careful land use database design will solve many of the design and operational problems which must be addressed by the GIS. The land use database will anticipate, not duplicate, many of the most important tasks of the GIS. Marine Parcels The parcel-based land use database described here will face special problems in the marine sphere, as do other land record systems. In brief, all Virgin Island land record systems have been designed and implemented on land. To the best of our knowledge: (1) There is no readily accepted definition of the seaward boundary of the Tax Assessor's map-and-grid coordinate system for property identification; and (2) There is no accepted nomenclature as an alternative to the "Estate and Quarter" legal descriptions of terrestrial properties. This is a difficult problem, which can only be solved with the involvement of several Virgin Islands and federal government offices and through study of alternative models which have been developed in other coastal communities. (It might be worth examining how the Danish crown was able to award property rights under Long Bay in St. Thomas Harbor to the West India Company.) In spite of its complexity, increasing pressure on land use and increasing con- flicts over marine resource uses make it important to solve the definitional issues as soon as possible. Architecture, Hardware and Database Language Detailed design considerations will determine if the land use database system should be implemented as two (or three) separate systems, operated au- tonomously on each island, or if it is possible to keep just one integrated database for all islands. Basic to this decision will be an examination of the update and revision process for the Tax Assessor's records. If the de- centralized architecture is chosen, it will still be necessary to have an in- tegrat~d database for territory-wide analyses. Chapter 5 Land Use Database Assuming a maximum of 50,000 land parcels, the proposed database is well within the processing capability of current state-of-the-art microcomputers. One such physical architecture would include a medium-speed 286 (i.e., AT compatible) processor, the standard 640K of main memory, and a 120 megabyte hard disk. Because the "Land Use Survey" Contract specifies that. this database should be in dBASE III Plus, that has been the focus of this design. Notwithstanding the contract terms, the choice of a programmable, relational database tech- nology is very appropriate for the Department's needs. IRF would recommend a dBASE III-based (or dBASE IV) system even if it were not required. With a full 50,000 parcel database, the Departmental users of the land use database will find processing speeds to be unacceptably slow for interpreted versions of the dBASE III language. Fortunately, several compiled versions of the language are available, including Quicksilver, Clipper, and Foxbase, as well as the soon-to-be-released compiler for dBASE IV. If the Virgin Is- lands Government proceeds with its plans to establish an Information Center, in which centralized batch file systems could be re-organized as on-line databases, DPNR should look to a database language with good SQL (Structured Query Language) interface. (dBXL has one of the best of such systems now -- this is new technology which should be mature in another year or two.) As with any major systems application, the DPNR land use database should be tried-out experimentally as a prototype. The island of St. John provides one possible example, but there may be other ways for the Department to check out the basic design and operation of the system. This initial prototype imple- mentation can probably use an interpreted dialect (cheaper, easier to modify) of dBASE. Major Database Fields Data from the Tax Assessor: Parcel 10 Using the Island-Map-Grid-Parcel Number identification system as- signed by the Tax Assessor has several advantages, not the least of which is that aggregation at the Map or Grid level for some vari- ables could quickly provide gross indices of many characteristics. Even in the absence of a graphic interface for a true GIS, analyses which can highlight general areas of the territory with special characteristics can be very useful for planning. Legal Description This is a two field description, based on the traditional house number, Estate and Quarter property descriptions still used by the Recorder of Deeds. [See the section on "Marine Parcels," above, for a discussion of the special problems of relating marine spaces to both traditional legal descriptions and Parcel IDs.] Chapter 5 Land Use Database Owner Name and Address These codes can be borrowed from the Tax Assessor records, but they will require significant editing to be useful. For example, last name and first name of owners are in one field, as are city, state and zip code. Land and Structure Area and Related Information Use Type Tax Assessor records can be used to provide land area of the parcel directly. Area of structures on the land can be calculated from the records, but it will require careful examination and analysis to determine if these are sufficiently accurate to warrant inclu- sion in the land use database. A special problem faced by the land use database is the question of how to count (and what to record) multiple buildings on the same parcel of land. At the least, the database should record the num- ber of structures on the parcel. Additional factors of this nature which are provided on the Tax As- sessor record system include: the "class" of each floor; the "ceiling height;" the "number of units;" the shape of the struc- ture; the year of construction; and numerous indicators of building infrastructure, including salt or fresh water toilet systems, cis- terns, swimming pools, fences, and so on. The Tax Assessorls record has a field for "Use Type," which should probably be captured. According to coding notations, however, the only characteristics recorded are, "single family housing, multi- family housing, and industrial." As with most other variables in the Tax Assessorls file, the actual coding employed for this field should be checked. In addition, the land use database will include a three digit field to record the parcel IS predominant land use, according to the de- tailed land use classification schema proposed in this project. Land and Improvement Value These are basic items which should be included, if privacy laws permit. It should be noted that these are raw estimated values of land and improvements, not property assessments. The record does not appear to indicate the date of either the estimate or the cur- rent assessment. Assessed Value As permitted, the assessed value of parcels and improvements should be recorded to facilitate later analysis and modeling of the real estate tax implications of alternative land use policies. Other Tax Related Information Important additional land use information in the tax record is the existence of "Exemptions" for farming, homesteads, and the elderly as well as the new sewer fees being levied on property owners Chapter 5 Land Use Database receiving such service. Although not directly relevant to land use, "Veterans l Exemptions" should be coded, if only to capture a complete picture of the operation of the property tax system. A Hi story Fil e As currently constituted, the Tax Assessor's file is only a state- ment of current conditions. Updates overwrite and destroy previous information. For the analytical and planning needs of DPNR, it will be important to build a history file, associating each parcel with changes in status and the dates of those changes. At this time, it is impossible to predict the pattern of changes in land parcel records. Therefore, the simplest way to build a his- tory file is to duplicate the prior record and write it to the his- tory file, every time a record is changed. After accumulating a few months (a year, if disk space permits) of such status changes, the history file can be analyzed to determine the best means of capturing those changes in the Tax Assessor's records that are most relevant to DPNR needs. This is the apparent limit of the information which can be gleaned from the basic source records in the Tax Assessor's automated files. In addition to the data described above, there is one other piece of information which must be added to the available tax assessor information for each individual parcel record. That is the geographic coordinates, as described briefly below, and extensively in Chapter 3, above. Geographic Coordinates These should be based on the grid system and parcel identification technique which the Department decides to adopt. This report rec- ommends the use of the Virgin Island extension of the Puerto Rican grid for all coordinates. The database should keep coordinates in only one system, and provide the translation to the other(s), as needed for specific system outputs. Land Use Data from Other Sources The following pieces of information (data record fields or related files) need to be extracted from other sources. It is legitimate to question whether such data is worth the cost of collection and maintenance, if it has to be manually extracted and associated with each of the 40,000 property parcels in the territory. Three alternatives to this costly and time-consum- ing process are possible: Hold off on collecting the information until it becomes more cheaply available as part of a larger mapping or GIS process. Eliminate the information as being too costly for the intended uses of the database. (See the recommendation for a detailed DPNR needs assessment in the first part of this chapter.) Set up an interim database design which would relate many of these vari- ables only to larger geographic units such as the map or grid coor- dinates of the Tax Assessor's maps. This would reduce the cost of Chapter 5 Land Use Database collecting and tabulating such data, at a significant reduction in the usefulness of such data in areas of high density or complex multiple use. It is assumed that for the interim period, this im- precision would not be a crippling defect in the development of this phase of the database. This is the alternative which we have followed for the design of this system. It is also worth noting that the information in the following classes of data is generally more stable than the information which is available from the Tax Assessor's records. This implies that a greater conversion cost for these variables might be acceptable, because the longer term costs of maintenance would be less. Land Use As mapped in this survey. The proposed detailed land use cate- gories would include a three digit code, referring to seventy-some classifications. Zoning and Coastal Zone Designations These should be provided in two separate variables to avoid unnec- essary complexity in coding. Design Issues This should be at least two different variables. One would be a three digit code indicating such key conditions as being in an His- toric Preservation District, an archaeological or historic site on the parcel, a significant cultural or natural landmark, visual quality factors, and so on. In addition, there should be an open- ended 50-character text variable to note major issues. Land Use Plan Categories A three digit variable reserved for the day when the territory has a land use plan. Soil Classifications A three digit code identifying major Virgin Islands soil types. Slope and Aspect Hydrology As the simplest topographical variables affecting land use, slope and aspect (direction of downslope) should be recorded at the par- cel level. As a true GIS is implemented, however, topography at a considerable level of detail should be included. This will permit the crucial viewshed and line-of-sight analyses which should guide permitting and land use policy making. One three digit variable can be used to highlight flood risk, sen- sitive erosion-prone watercourses and shorelines, and aquifer recharge zones. Seismic Constraints At the present time, the best indicator that can be achieved is a three digit variable indicating one of eight categories of risk for a parcel, based on the following factors: over or adjacent to Chapter 5 Land Use Database known faults; areas of potential liquifaction; tsunami potential; inundation (subsidence risk) area; seismic zones of intensities one (low) to four (highest). Geological Constraints Conclusion This category may cover land use factors which have been suf- ficiently handled by other variables. Major values would include: landslide susceptibility; subsidence or sinkhole potential. For further information on this design, see the dBASE database description which follows, and which has also been submitted on a floppy disk. The proposed database consists of two related files: one (DPNRTAX.DBF) con- sisting primarily of data at the parcel level, extracted from the georefer- encing system and the records of the Office of the Tax Assessor; and one (DPNRLUSE.DBF) consisting of more general attribute information related to land use, registered at the map-grid number level. These two files are re- lated through the map-grid index number. In addition, the parcel-based file is indexed on the parcel identification number assigned by the Tax Assessor. Not shown in the following lists is the history file, which, as described above, should duplicate DPNRTAX.DBF, the parcel-based file, with a field for the date of the transaction. The actual method by which this record is gen- erated will depend on the Department's overall systems design for updating these records. As with any major database design, these proposed files should be considered preliminary and tentative. In the specific case of this study, lack of ac- cess to DPNR record systems for the Department's land records has been a spe- cial problem. While we know the detailed composition of the Tax Assessor's computer files, we don't know the form or filing procedures for DPNR's own land record systems. In particular, there may be ways to get good indicators of major issues, such as energy use, building height, water storage and use, sewage systems, and so on. A major advantage of designing a database manage- ment system such as dBASE III is that record systems and files can be revised with relative ease. Chapter 5 Land Use Database The structure for the database file DPNRTAX.DBF is as follows: Field Field Name Type 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 PARCEL NO MAP GRIDNO LEGAL DES1 LEGAL-DES2 LNAME- FNAME MNAME ORG NAME ADD1 ADD2 CITY STATE COUNTRY ZIP LAND AREA BLDG-AREA TA USE TYP LAND VALUE IMPRV VALU FARM EXEMP HOME-EXEMP ELDEREXEMP VETS EXEMP SEWER FEE ASSESVALUE DATE BUILT NUM SLOGS CONST MATS GEO COORDX GEO-COORDY Character Character Character Character Character Character Character Character Character Character Character Character Character Character Numeric Numeric Character Numeric Numeric Numeric Numeric Numeric Numeric Numeric Numeric Date Numeric Character Numeric Numeric Length Comment 12 10 24 20 20 20 12 24 24 24 24 20 15 9 12 12 1 12 12 12 12 12 12 12 12 8 3 1 12 12 Tax Assessorls Parcel 10 Calculate from Parcel 10 Tax Asses or Rec. of Deeds Tax Asses or Rec. of Deeds Adapt from Tax Assessor II II II II II II II II II II II Cadastre or Tax Assessor Cad., Tax Assessor or DPW Tax Assessor Rec #5 Tax Assessor Rec #2 Tax Assessor Rec #2 Tax Assessor Rec #2 Tax Assessor Rec #2 Tax Assessor Rec #2 Tax Assessor Rec #2 Tax Assessor Rec #2 Calculated Tax Assessor Rec #6 Tax Assessor Rec #6 Tax Assessor Rec #6 DPNR GeoReference System DPNR GeoReference System The total record is 416 characters. Two index files are specified, based on the PARCEL NO (DPTAXPAR.NDX) and the MAP_GRIDNO (DPTAXMAP.NDX). Chapter 5 Land Use Database The structure of the DPNRLUSE.DBF file is as follows: Field Field Name Type Length Comment --------- ------ 1 MAP GRIDNO Character 10 from Real Estate Map 2 LAND USE Character 3 DPNR 3 ZONING Character 6 DPNR 4 CZM CAT Character 6 DPNR Permits 5 DESIGN Character 3 DPNR Tech Review 6 DESGN CMNT Character 50 DPNR Tech Review 7 LUSE PLAN Character 3 DPNR Comp Planning 8 SOIL-CLASS Character 3 SCS or Dept of Ag 9 SLOPE Character 1 DPNR Tech Review 10 ASPECT Character 1 DPNR Tech Review 11 HYDROLOGY Character 3 CRI Water Resource Center 12 SEISMIC Character 3 VITEMA 13 GEOLOGY Character 6 The total length of the file record is 99 characters. The index file is DPLUSMAP.NDX, based on the MAP GRIDNO field. CHAPTER 6 SEMINAR ON GIS The following notes are adapted from lecture notes distributed at a seminar on current topics in Geographic Information Systems, held in the Conference Room of the Department of Planning and Natural Resources on April 17, 1989. Summary Information on: Geographic Information Systems Land Information Systems Multi-Purpose Cadastre Multi-Purpose Land Information Systems Automated Mapping/Facilities Mapping Systems (all of the above are roughly the same) A suggested definition: Geographic Information System: A Technology for combining Graphic (map) Data and Tabular (database) Information in a topologically related system permitting analysis and display of both graphic and tabular information. Chapter 6 Geographic Information Systems WHY NOW? GIS/LIS projects are not new. Even state-wide programs date back to the 1960's. 1. Power of cheap "super-micros" and workstations. 2. Durability and flexibility of digital mapping systems 3. Wide use of computerized databases. 4. Demands for increased accuracy and productivity (including increased le- gal liability). 5. Improved high resolution display and hardcopy devices. Chapter 6 Geographic Information Systems "GRAPHIC DATA:" MAPS separated into Themes (Layers) (e.g.): Quarter and Estate Boundaries Water Land Use Parcels Roads Geodetic Reference * * * * * * * * * * * * "TABULAR DATA:" From files and databases currently maintained by PNR and other Government Agencies. Need Relational DB Structure Breakthru technology: "truly distributed" relational database technology Chapter 6 Geographic Information Systems Overlay Example for GIS Analysis A. Parcels Lt. Governor B. Zoning DPNR C. Floodplains -- VITEMA D. Wetlands -- DPNR E. Land Cover -- ED & Ag F. Soils -- SCS G. Reference USGS or DPNR H. OVERLAY Chapter 6 Geographic Information Systems GIS GRAPHIC SYSTEMS: Grid-Based (Raster) PRO: Easy data entry Very strong analysis Used for Natural Resource Management Efficient data storage CON: Coarse resolution EXAMPLES: MAP, etc. EPPL7 MOSS GRASS ERDAS (satellite display) SPAN by GEOVISION Chapter 6 Geographic Information Systems GIS GRAPHIC SYSTEMS: Vector-based (arc) Systems: PRO: High Resolution Accuracy CON: Complex links to tabular data Proliferation of slivers CPU-intensive processing EXAMPLES: ARC/INFO by ESRI Synercom Intergraph GFIS by IBM New Technology (possible breakthrough): "FEATURE ORIENTED" SYSTEMS System 9 by Wild TIGRIS by Intergraph Chapter 6 Geographic Information Systems Recommended System The Rules Do not unnecessarily commit to a single system or vendor Choose elements with broad acceptance for the intended use (e.g., Synercom for mapping, not necessarily for GIS) Choose modular, proven upgradeable systems (many false claims of compatibility) Both for hardware platforms, and for expanded software functionality. The VI needs new base map. Get maximum affordable resolution (with minimum number of features) Conduct extended proto-typing * * * * * * * * * * * * THE SYSTEM 1. AutoCad or Intergraph mapping system(s) 2. ARC/INFO or pcARC/INFO GIS, integrated with ERDAS raster-based display technology Chapter 6 Geographic Information Systems PREDICTABLE PROBLEMS 1. Inter-departmental cooperation 2. Inter-divisional coordination 3. Sustaining a five-year project 4. Retaining qualified staff 5. Managing multiple contracts for geodesy, photogrammetry, data conver- sion, etc., etc. 6. Maintaining databases in the future. * * * * * * * * * * * * PARTIAL SOLUTIONS Active gubernatorial support High level project leadership in PNR Contract most non-operational expenses to reliable "prime" consultant/ con- tractor Develop cost-sharing systems with private sector. Chapter 6 Geographic Information Systems URISA AAG RESOURCES Urban and Regional Information Systems Association 319 C Street SE Washington, DC 20003 "Best Buy" among the associations. Association of American Geographers Arlington, VA AM/FM International ASPRS ACSM Automated Mapping/Facilities Mapping (Low-scale mapping and engineering drawing -- the province of depart- ments of public works.) Chicago, IL American Society for Photogrammetry and Remote Sensing Arlington, VA American Congress for Surveying and Mapping Arlington, VA Note: ASPRS and ASCM have two videotapes on GIS technology that might be a good introduction for PNR staff. GIS/LIS '89 Fourth annual November meeting, sponsored by five groups above, devoted exclusively to GIS issues. In each of the past three years, annual participation has doubled. Three thousand in San Antonio last Decem- ber. 1989 Meeting will be in Orlando. Institute for Environmental Studies University of Wisconsin Publishes the series: "Wisconsin Land Information Reports"at $7.50 per volume, these are a "Best Buy"for introductory information on GIS. North Carolina has also had a recent, well documented state LIS pro- gram. Chapter 6 Geographic Information Systems Annotated List of References Provided: Chrisman, Nicholas R. "Effective Digitizing: Advances in Software and Hard- ware,"A Primer on Multipurpose Land Information Systems. Wisconsin Land Information Report 4, Institute for Environmental Studies, pp. 83-91. Madison, 1988. Describes why data entry is so costly, and the relatively minor increases in productivity which result from the latest technologies. Good sober antidote to salesmen. Croswell, Peter L. "Definition of Applications as a Basis for GIS Planning and System Procurement," in URISA Proceedings for 1988 Annual Meeting, Mapping the Future, Volume II, page 13-20. Washington, 1989. Excellent short summary of GIS planning process. Dangermond, Jack. "The Land Record Transaction," paper presented and pri- vately distributed at AutoCarto 9, Baltimore, March, 1989. Wisdom about processing land records in a GIS, from the President of the Environmental Systems Research Institute, publishers of the most successful GIS in the world. Dueker, Kenneth J. "Urban Applications of Geographic Information Systems Technology: A Grouping into Three Levels of Resolution," in URISA Proceedings for 1988 Annual Meeting, Mapping the Future, Volume II, page 104. Washington, 1989. Short, tabular presentation of scale and cost factors for various sorts of GIS, ranging from three week to ten year projects. Hanigan Group. "GIS Recognized as Valuable Tool for Decision Makers," in THE GIS FORUM. Premier Issue, page 4. San Antonio, 1989. Good survey of the breadth of GIS-type applications, with the major vendors in each market. Hanigan Group. "Metrocom, A $25 Million Parable," in THE GIS FORUM. Premier Issue, page 1. San Antonio, 1989. Houston's costly experience in learning that GIS requires basic changes in management, more than reorganization of dp shop. Thompson, Robert W. "Conversion Issues," paper presented and distributed in mimeograph at AutoCarto 9, Baltimore, March, 1989. Operational pointers on how to plan and manage the GIS data conversion process from one of America's most experienced practitioners. Chapter 6 Geographic Information Systems Thorpe, John. Privately circulated lecture notes, "Comparative GIS base map cost factors for various features and scale." GIS/LIS III. San Antonio, December 1989. Tilley, Scott and Stephen L. Sperry. "Raster and Vector Integration," in Computer Graphics World, August, 1988. Description of the technology used to integrate ERDAS and ARC/INFO GIS display systems, permitting the apparent combination of satellite imagery and more conventional map and GIS products. Leading edge technology. CHAPTER 7 PLANNING, DATA MANAGEMENT AND POLICY RECOMMENDATIONS In concluding this project,the research and mapping team reached a kind of consensus on three general observations: (1) the pace and scope of development on St. Thomas, St. Croix and St. John have increased significantly in the last decade, and both the visible and more subtle induced changes in the landscape warrant more detailed assessment and mapping; (2) the quality ,quantity, accessibility and flexibility of the Virgin Islands' land use database in its various manifestations is simply not adequate to meet the complex planning and management needs of today; and (3) more anticipatory planning, more aggressive growth management, and more efficient natural resource system protection appear necessary to reverse the increasingly visible gradual decay of environmental quality in the Virgin Islands. This chapter will elaborate on these themes and offer some recommendations re- garding identified areas of concern. IRF's eight week contract with DPNR called for not only a land use mapping initiative demarcating distinctive kinds of land use but also called for the calculation of the proportion of each of the major land uses in the Virgin Islands. IRF has measured areas covered by each class and then calculated the proportionate share of the total for each land use class or category that was used in preparing the land use maps. The results of these calculations are as follows: Chapter 7 LAND USE CLASSES AGRICULTURE RESORT/HOTEL RETAIL/COMMERCIAL INDUSTRIAL/MANUFACTURING PUBLIC FACILITIES LOW DENSITY RESIDENTIAL MEDIUM DENSITY RESIDENTIAL HIGH DENSITY RESIDENTIAL ALL RESIDENTIAL URBAN WATERFRONT/MARINE PARKS/RECREATION/OPEN SPACE UNDEVELOPED Recommendations Page 'Jil St.JOHN ST. THOMAS ST.CROIX 0-1% 0-1 % 12% 1% 1% 0-1% 0-1% 1% 0-1% 0-1% 0-1% 5% 0~1% 4% 5% 2% 12% 10% 1% 13% 3% 0% 1% 0-1% 3% 26% 14% 0% 2% 0-1% 0-1% 0-1% 0-1% 68% 1% 3% 26% 61% 59% Note 1: Totals may not add up to 100% due to rounding. Note 2: Percentages are calculated as number of acres of each use divided by total land area of each island; example: there are 2,313 acres of land in low density use in St. Thomas, there are 18,817 acres of land in St. Thomas (excluding offshore cays); 2,313 divided by 18,817 equals .1229 or 12 %. This statistical array needs to be used in conjunction with the land use maps that accompany this report. The purpose of the land use information developed by this project is to present a picture of the extent and type of land use in the Virgin Islands. As such, the maps and other information collected for this report and the accompanying graphics are one of many planning tools necessary to produce a land and water use plan. Data collection, however, should not be confused with planning. Planning is a method of developing policies, laws and other actions that manage growth and development decisions that are to be made in the near and long term future. Data collection on the other hand is an assessment of current conditions from which various conclusions can be reached and planning policies developed. Chapter 7 Recommendations Page '52 It is beyond the scope of this project, however, to make judgments regarding the land use data presented on the maps and in this report. For instance, whether there is too much or too little land on St. Thomas devoted to residential use is a policy decision that can only be made by the Government of the Virgin Islands. On the other hand, certain patterns can be discerned that indicate areas where development has tended to cluster or where there is a distinct lack of recreation areas. If one were to look only at the percentages of land use types, one would see, for example, that only about one per cent of the land area of St. Thomas is devoted to hotel-resort use. This figure provides a distorted picture of the actual situation. If one were to look at the land use map, it would be clear that the preferred location of a resort-hotel is on a beachfront site. It would also be noticed that most resort-hotels are, in fact, located on a beachfront site. It would also be noticed that beachfront sites are few in number in St. Thomas, and of these few sites the vast majority currently have situated on them at least one resort-hotel. Both the statistical evidence and the maps clearly show that the amount of land dedicated to parks, recreation and open space in St. Thomas and St. Croix is minimal. On St. John the presence of the Virgin Islands National Park skews the data. It is clear from viewing the land use pattern on St. John that the non-park areas east and southeast of Cruz Bay are experiencing intense growth pressure. Although the presentation of the proportional data of the various land uses of the Virgin Islands provides a picture of the existing situation, it would be useful and would have real planning significance to map the same land use categories with air photographs from selected prior year(s) such as 1975 or 1980. Further analysis is required to calculate the proportion of each land use to the amount of land available and suitable to accommodate that use. This task requires an analysis of both zoning categories and land capability characteristics such as slope, soils, views, and water access. PLANNING INFORMATION A geographic information system would be of great use as a tool to manage the complex interactions between land uses and land characteristics. A GIS can be used to help manage future land use based on slope, soils, road access, etc. The major advantage of a GIS is the ability to perform simultaneous analyses of multiple factors, a process that is too complex to do manually, especially when large land areas are involved. A GIS -- especially one including databases extending beyond land use -- can also help DPNR to focus on areas where environmental problems tend to cluster (e.g., first tier of the coastal zone). In addition, there is a great need in the Virgin Islands to dis-aggregate and conduct detailed analyses on those land use areas which comprise a large percentage of an island's total land area such as "low density resi~ential" and the "undeveloped" class. Again, a GIS can be of major assistance in this task. The implications of observations such as these: the apparent lack of adequate parks and recreational facilities; the few remaining beaches that do not have a hotel situated on them; the tendency of land outside of the National Park Chapter 7 Recommendations boundaries in St. John to become extremely valuable and subject to intense growth pressure; the apparent loss of agricultural land on St. Thomas and St. John; the concentration of population in certain nodes such as Sion Farm, Red Hook, and Tutu; transportation problems engendered by the distance between population and employment centers and health care facilities; the distinct lack of development in certain portions of the islands (Coral Bay, St. John; West End, St. Thomas; East End, St. Croix); and the general "suburbanization" of development around the four town centers ( Charlotte Amalie, Christiansted, Frede~iksted, and Cruz Bay) all require further study of the pattern and dynamics of change and more detailed planning. It is clear that the Virgin Islands are experiencing what could be called a pattern of "Florida style" development. Such development tends to override the island's West Indian identity. It seldom recognizes the special development constraints that smaller, dry, tropical islands face. Techniques such as adaptation to steep topography, protecting views, providing adequate infrastructure and protecting unique features and regional identity need to be introduced into the planning and management process. A recent Conservation Foundation newsletter story (Buchta, 1987) has drawn our attention to many locations in the United States that are in danger of becoming "accidental cities". According to the Conservation Foundation, a wave of poorly managed growth is transforming America's suburbs and small towns, changing some suburbs into "accidental cities" and small towns into suburbs. This transformation is bringing to these communities many of the drawbacks of big cities, such as traffic congestion, with few of the positive qualities that can make cities exciting and enjoyable places. While citizens, officials and planners ponder what to do, some of the ingredients critical to making a community more livable -- open space, rivers and other natural features, scenic and productive agricultural lands, historic buildings -- are deteriorating or disappearing rapidly. The problems characterized by the Conservation Foundation as "accidental cities" are common to many locations that are subject to rapid growth. The Virgin Islands are no exception to the planning problems facing such areas. The Virgin Islands, however, perhaps have more at stake than most areas This is because the Virgin Islands are a small closed, bound system from a planning perspective. Unlike some areas where there is almost unlimited land for expansion and the accommodation of a wide variety of uses, the Virgin Islands must accommodate all uses within a limited land area. For example, many communities the size of St.Thomas could not easily accommodate a major airport, a large landfill and a hospital. St. Thomas and St. Croix, of course, must find space for these and many other potentially incompatible uses. As has been well documented, the planning problems of small islands require a high level of attention. The quality of the environment as well as many components of the quality of life for island residents can only be optimized by careful planning and informed natural resource management, sensitive to the ecological base with its various free services which undergird our community structure. As noted it is beyond the scope of this project to suggest Chapter 7 Recommendations PageA4 planning solutions and management recommendations. The results of the land use survey clearly indicate, however, that there are acute and long-range planning issues that need to be addressed. The Virgin Islands are in danger of becoming an "accidental city" if a meaningful land and water use plan is not developed, adopted and implemented in the near future. RECOMMENDATIONS A second element of this project was the development of the framework for a land use database for the Virgin Islands. Chapter 5 of this report contains the recommended database design. In the course of developing the database certain observations were made relating to land use data in the Virgin Islands. In the first place, it appears that the Department of Planning and Natural Resources is in urgent need of improving its data collection and information management system. We note that the Department does not have an orderly method of collecting, storing or retrieving land use and natural resources data. DPNR could not, for instance, provide IRF with information on the location of approved subdivisions and approved major development projects. Also unavailable was a running inventory and map of historic and archaeological sites, either marine or terrestrial. In addition information related to permit granting such as number and type of permits, value of permitted construction, and number of permits issued by year seems to be unavailable. It also appears from preliminary evidence that the various DPNR permit application forms were designed individually and do not provide a unified data collection program. An immediate improvement in data collection would be achieved should DPNR design a unified permit application, or at least a standardized cover sheet, based on the database proposed in this report. A separate but related issue is that of data storage, archiving, security, and retrieval. It appears that within DPNR there is no systematic labelling, coding, cross-referencing or cataloging of important land use data such as maps, reports, air photographs, and permit files. An expanded database can be designed which can assist with the maintenance, analysis and use of permit files. A simple coding and catalog program for maps, reports and other land use information can be easily structured. Better still, establish a library so documents can be signed out to individuals and a custody trail established. One of the advantages of a geographic information system for the Virgin Islands would be the ability to revise and produce all maps in usable and retrievable form. In addition, a GIS would enable the up-dating of land use information on a regular, even daily, basis. As recommended in this report, the Virgin Islands Government needs to move in the direction of a more systematic and standardized method of collecting land use information. As a start, systematic and standardized geo-positioning information should be required elements of all maps and reports. It is also recommended that the data use requirements of the various Virgin Islands Government agencies (DPNR, WAPA, Public Works, Port Authority Tax Assessor, Police/Fire Departments, etc.) be assessed as it is likely that a coordinated and shared data collection system can be devised. Such a system, perhaps with shared computer facilities, could be cost effective and produce major benefits to the Virgin Islands. Chapter 7 Recommendations A few final observations are in order. The first is that it has not been the intention of this report to single out the Virgin Islands as an example of a location in need of improved data management. In fact most state and local governments in the United States are in the same position. What IRF has tried to do is be specific and report findings and conclusions from the land use survey project as objectively as possible. More importantly, rather than simply highlight problems, we have tried to highlight options that would lead to realistic solutions. As noted, certain Ill ow tech" steps can be taken by DPNR to make quick improvements in land use data management. These include the development of a simple code and label system for all maps within the Department and a redesign of permit applications to ensure that geo-positioning information and other data which may be feed into the database is collected. Additionally, we have noted often in this report the importance of geographic information systems. We have provided DPNR with recent information on the application of a GIS as well as on the positive and negative aspects of such systems. The Virgin Islands Government is perhaps fortunate not to be locked into the hardware and software of anyone geographic information system. It is better to evaluate the Territory's needs and budget in a orderly fashion before committing to one GIS or another. The good news is that the technology exists to provide the Virgin Islands with a GIS that will meet the challenges of the 1990 1s. BIBLIOGRAPHY Baker, R. and N. Pachta, 1983. General management plan - development concept plan - environmental assessment. National Park Service, U.S. Department of the Interior, Denver, CO. Buchta, Ted, 1987. communities?" "Will we live in accidental cities or successful Conservation Foundation letter no. 6. CH2M Hill, 1980. U.S. Virgin Islands: Base map series. 1" = 1,000 1 • Gainesville, FL. Cook, C., et al., 1962. Land use classification manual. Detroit Metro. Area Reg. Plan. Comm., Detroit, MI. Hess, A., 1984. Land capability analysis technologies. Draft working paper. Off. Tech. Assess., U.S. Congress, Washington, DC. Potter, B., K. Green, and M. Goodwin, 1988. Management of natural resource information for the Virgin Islands National Park and Biosphere Reserve. Special biosphere reserve report. Island Resources Foundation, St. Thomas, V.I. Teytaud, A., 1981. A Virgin Islands land use/land cover classifica- tion system (based on the U.S. Geological Survey and State of Florida classification systems). Department of Conservation and Cultural Affairs, St. Thomas, V.I. Div. of CZM Tech., Circ. No. 1 • Towle, et al., 1986. Land use management plan for Southeast Peninsula, St. Kitts, West Indies. Island Resources Foundation, St. Thomas, V.I. U.S. Geological Survey, 1982. East Point, St. Croix. DMA 1621 II NE- Series E836, Reston, VA. U.S. Geological Survey, 1982. Christiansted, St. Croix. DMA 1621 II NW-Series E836, Reston, VA. U.S. Geological Survey, 1982. Frederiksted, St. Croix. DMA 1621 III NE-Series E836, Reston, VA. U.S. Geological Survey, 1982. Western St. John. DMA 1622 I SW-Series E836, Reston, VA. U.S. Geological Survey, 1982. Eastern St. John. DMA 1622 I SE-Series E836, Reston, VA. U.S. Geological Survey, 1982. Western St. Thomas. DMA 1522 I SE- Series E836, Reston, VA. Bibliography U.S. Geological Survey, 1982. Central St. Thomas. DMA 1622 IV SW- Series E836, Reston, VA. U.S. Geological Survey, 1982. Eastern St. Thomas. DMA 1622 IV SE- Series E836, Reston, VA. 8.S. Virgin Islands Government, 1973. Virgin Islands zoning law. Office of the Lieutenant Governor, St. Thomas, V.I. U.S. Virgin Islands Government, 1977. coastal zone management program. St. Thomas, V.I. Preliminary Virgin Islands Virgin Islands Planning Office, U.S. Virgin Islands Government, 1977. Land use and housing elements, U.S. Virgin Islands. V.I. Planning Off., Off. Gov., St. Thomas, V. I. U.S. Virgin Islands Government, 1980. Which way the Virgin Islands? Goals and objectives for the comprehensive land use plan. V.I. Planning Off., Off. Gov., St. Thomas, V.I. U.S. Virgin Islands Government, 1988. The Virgin Islands executive budget: fiscal year 1989. Off. Gov., St. Thomas, V.I. Virgin Islands Water and Power Authority, 1987. St. John: grid map. Sheet Index. Yaro, R., et al., 1988. Dealing with change in the Connecticut River Valley: a design manual for conservation and development. Lincoln Institute of Land Policy and the Environmental Law Foundation, Amherst, MA. Zube, E., et al., 1968. The islands: selected resources of the United States Virgin Islands and their relationship to recreation tourism and open space. Prep. for U.S. Dept. Interior, Washington, D.C.