STEP 1: DETERMINE HISTORICAL GROWTH TRENDS

USING TAX DATA AND THE PLSS

In order to determine urban growth trends in the area, we used the Florida Department of Revenue's Property Tax Data Records. These tax records contain parcel information, including ownership, property values, land use, year built, and location. For our study, we used 1992 tax data. Tax records for 1996 were available, but lacked data for a large number of parcels (over 15% of the total parcels). Therefore, we used 1992 data, which we considered sufficient for generating historical trends.

Within the tax data tables are three fields that were particularly useful for our study: DORCODE, YRBUILT, and TRS. The DORCODE field identifies the Department of Revenue's land use classification for each parcel. It was used to select all parcels with a classification of 'residential', 'commercial', or 'industrial'. The YRBUILT field indicates the year that development occurred on each parcel. Finally, the TRS field refers to the township, range, and section, or the location of each parcel within the Public Land Survey System (PLSS). This field was especially important because it provided a means by which the tax data could be joined to a digital map of the PLSS.

Digital PLSS maps were obtained from the Florida Geographic Data Library (FGDL). Tax tables were joined to digital PLSS maps on the TRS field. Joining the tax tables to the PLSS enabled us to generate a spatial representation of growth trends by decade.

GENERATING GROWTH TRENDS BY DECADE

A query-based approach was used to determine growth trends. That is, queries were run on the tax record database to extract pertinent information. An example of such a query might be, "How many parcels in each section are classified as 'residential' and experienced development between the years 1930 and 1940?" Multiple queries were run on the tax record database to obtain information needed to develop historical trends.

First, parcels with a land use classification of 'residential', 'commercial', or 'industrial' were selected from the database. The selected parcels were then grouped by the decades in which they experienced development. This was accomplished by selecting parcels with a YRBUILT value of 1900 through 1910, 1910 through 1920, 1920 through 1930, up to 1980 through 1990. The number of residential parcels, the number of commercial parcels, and the number of industrial parcels per section were summed for each decade.

We evaluated each decade of growth by looking at the change in number of parcels per section and by visually displaying the sections that had experienced growth. By evaluating each decade of growth, we saw that a significant increase in development began in the 1960s. There seemed to be a growth "boom" beginning in this decade. We then decided to concentrate on the three decades since the 1960s to determine the historical growth patterns, as we felt that these decades would more accurately represent the current growth trend in the area. Since the tax records had data up to 1992, we chose the following 10-year intervals to do further analysis on: 1963-1972, 1973-1982, and 1983-1992. The same queries that were performed for each decade were run again for these three decades. That is, residential, commercial, and industrial parcels were selected and grouped by the decade in which they experienced development; and resulting data were summarized for each section.

To model growth, we relied on the basic assumption that future growth will follow historic growth. In other words, areas that have experienced growth will most likely continue to grow. After tracking where urban growth had occurred historically, we wanted to predict where future growth is likely to occur based on proximity to the areas that have historically experienced growth. In order to accomplish this, we performed a distance analysis for each of the three decades of residential, commercial, and industrial growth.

A distance analysis is a way of classifying areas based on proximity to a feature. The classification method places greater importance on areas close to the features and less importance on areas far from the features, or visa versa. For example, businesses want to locate near highly visible and busy traffic intersections. Areas close to intersections are more appealing than areas far away. In this situation, areas close to intersections would be assigned a value of 9. Gradually, moving away from the intersections in all directions, decreasing values are assigned to areas. Areas farthest from the intersections are assigned values of 1. The result of a distance analysis is called a distance analysis surface and is essentially a map with zones ranging from 1 to 9.

For each land use type (residential, commercial, industrial) the distance analysis surfaces for the three decades were combined to obtain a historical growth trend for that land use type. Each decade of growth was given a weight, with the most recent decade being assigned the greatest weight.

Back to Growth Group Main Page   Next Step