Abstract

A digital dataset for the Floridan aquifer system in Florida and in parts of Georgia, Alabama, and South Carolina was developed from selected reports published as part of the Regional Aquifer-System Analysis (RASA) Program of the U.S. Geological Survey (USGS) in the 1980s. These reports contain maps and data depicting the extent and elevation of both time-stratigraphic and hydrogeologic units of which the aquifer system is composed, as well as data on hydrology, meteorology, and aquifer properties. The three primary reports used for this dataset compilation were USGS Professional Paper 1403-B (Miller, 1986), Professional Paper 1403-C (Bush and Johnston, 1988), and USGS Open-File Report 88-86 (Miller, 1988). Paper maps from Professional Papers 1403-B and 1403-C were scanned and georeferenced to the North American Datum of 1927 (NAD27) using the Lambert Conformal Conic projection (standard parallels 33 and 45 degrees, central longitude -96 degrees, central latitude 39 degrees). Once georeferenced, tracing of pertinent line features contained in each image (for example, contours and faults) was facilitated by specialized software using algorithms that automated much of the process. Resulting digital line features were then processed using standard geographic information system (GIS) software to remove artifacts from the digitization process and to verify and update attribute tables. The digitization process for polygonal features (for example, outcrop areas and unit extents) was completed by hand using GIS software.

Raster datasets depicting structural surfaces of time-stratigraphic and hydrogeologic units were created from digitized contour polyline features. These line features were converted to equidistant point features that were used to interpolate a continuous raster surface. Geologic features that disrupt the structural surfaces, such as faults, were built into the interpolation process in order to maintain the correct relation between rocks on either side of the fault. Resulting raster datasets were extracted using polygons to confine their extents to those of the parent polyline features. Raster datasets were created independently of one another, and logical relations between datasets may not be maintained everywhere. Irregularities may exist especially in areas near edges or where the units are relatively thin. For example, Upper Floridan aquifer elevations should always be greater than Lower Floridan aquifer elevations, but in a few areas Lower Floridan aquifer elevations are greater. Because the digital datasets are intended to be pure reproductions of the paper maps from which they were developed, no attempt was made to quantify or correct these irregularities.

Open-File Report 88-86 (Miller, 1988) contains geologic data interpreted from well logs and drillers reports that were used to create the structural surface maps in Professional Paper 1403-B (Miller, 1986). A point feature class attributed with elevations and depths of geologic and hydrogeologic units was compiled from these data and compared to the interpolated structural raster surfaces to check accuracy. Average elevation differences between datasets ranged from 0 to 23 feet, with standard deviations ranging from 30 to 122 feet. Large differences (greater than 1,200 feet) between datasets were observed in some areas and are most likely explained by typographical errors in the RASA database of Miller (1988).

Because of slight irregularities in each paper map, not all shared edges in the digital dataset align with every map. An effort was made to maintain topological relations between polygons and polylines rather than absolute spatial position. For example, early Eocene outcrop boundaries should logically share edges with Paleocene outcrop boundaries. When comparing early Eocene outcrop polygons with the scanned image of the Paleocene outcrop areas, there is a slight offset between the shared edges, though the two digitized polygon datasets are topologically correct.

Version 1.0

Posted April 2011