Original Data
The environmental analysis phase of Vision: Frontera began with an evaluation of existing data sources and maps. We tried to collect as many maps, GIS files and aerial photos as possible. Institutions contacted included the U.S. Geological Survey (USGS), Texas Natural Resource Information System - Borderlands Information Center (TNRIS), Texas Natural Resource Conservation Commission (TNRCC), Texas Department of Transportation (TxDOT), University of Texas Libraries, International Boundary Waters Commission (IBWC), the Fideicomiso para el Desarrollo Norte del Estado de Nuevo Leon (FIDENOR), and others as well.

We started with Mexican natural resource thematic maps from the Dirección General de Estudios del Territorio Nacional (produced between 1978 and 1982). In general, these maps served as a reference source for primary data analysis.  The names of the map series are as follows:

We worked in teams to analyze the maps and determine their extent and limitations, and then developed strategies for field verification and additional data collection.  

We also ordered an overlapping, stereoscopic set of black and white aerial photos [April 1994, 1:24,000 scale] from TxDOT, which were enlarged and then scanned at a useful high resolution (600 dpi). Their on-land resolution is about 0.4 meters per pixel, which allows very close levels of examination. 

Another source of aerial imagery was the USGS which produces Digital Ortho Quarter Quadrangles (DOQQs).  These images are so named  because they have been orthographically rectified and they each cover the area of one quarter of a 7.5 minute USGS quadrangle map. The main advantages of the DOQQs are as follows: they were flown in the summer of 1997 --  more recently than the black and white aerials; they are color infrared images (they are useful for determining vegetation types and densities); and they have been "rectified,"  a complicated process that eliminates spatial distortions common on aerial photos, caused by elevation differences, in-flight variations, and curvature of the camera lens.  These images have an on-land resolution of approximately 1 meter. 

Digital Data Capture Method
In order to prepare data for accurate mapping, we defined a set of fixed, absolutely georeferenced "control points" that could function as anchors for all data that would be added during the project. For this purpose, we defined several distinctly recognizable ground features like road intersections, borderlines, houses, etc. that were recognizable on the 1978 Mexican maps, aerial imagery, and all other maps. We digitized these points with their absolute longitude and latitude locations using a CalComp digitizing table, with the "Digitizer" extension of ESRI's ArcView software.

The control points were used to align scanned aerial photographs with the help of the ArcView "Image Analyst" extension (in a process called "rubber sheeting") so that these images would be georeferenced. Using the georeferenced images, we could then digitize from the computer screen using the "heads up digitizing" method, instead of using the digitizing tablet. [Click here for heads-up digitizing instructions.] This procedure allowed us to produce digital versions of the 1978 maps showing vegetation and land use, soils, geology, topography, hydrologic features and manmade features.  Using ArcView GIS software, then, each thematic map could be displayed as a theme and used as an overlay to create multi-themed maps.  This digital thematic information was displayed on a background of the aerial imagery, which had been scanned and georeferenced.

Data Collection and Mapping Methods
In order to verify data from the 1978 maps and collect additional data, we made two field visits to Rancho San Eduardo.  The first trip (February 25-27, 2000) was devoted to exploration of the study area and preliminary data collection, with attention focused on how best to collect specific data.  The second trip (March 10-13, 2000) provided an opportunity to do extensive data collection and field-based analysis.  In the field, printouts of the digital maps, photocopies of the original Mexican maps and photocopies of  aerial photographs were used for orientation and recording of collected data.  Class members worked in teams to collect data and verify items noted on the 1978 maps.  In addition to general observation and sampling strategies, fieldwork was supplemented with the use of a Trimble GeoExplorer 3 global positioning system (GPS) unit.  This top-of-the-line equipment can produce digital maps with an accuracy of one to five horizontal meters and two to ten vertical meters.  On the second field trip, a more advanced Trimble-XRS GPS was used, with a sub-meter horizontal accuracy and vertical accuracy of one to two meters.  Many topographical  and line features (highways, ranch roads, perimeters of water tanks) were traversed to collect GPS data.  Point features such as buildings, soil and vegetation sample locations, and water features were also collected using the GPS unit.

Data and Maps Produced
Once fieldwork was completed, the teams began analyzing field data, comparing it to the original maps.  After importing the GPS data into ArcView, however, it was noted that field GPS data didn't line up accurately with the georeferenced aerial imagery.  This problematic disparity was diagnosed as a map projection problem, with the GPS data having been collected using a different projection system (Universal Transverse Mercator zone 14, 1927 North American Datum) than the DOQQs (UTM zone 14, NAD 83).  Skillful analysis led to a reprojection method using the ESRI extension "Projection Utility" for ArcView. [Click here for reprojection instructions.]  This proved to be a successful remedy, as reprojected files matched perfectly with the DOQQs. In order to reach a level of alignment approaching the U.S. National Mapping Standards of accuracy, the aerial imagery as well as the digitized features were also re-aligned to match the DOQQs. The result is an on-land deviation of no more than 10 meters, which is about double the quality required in the standards for map accuracy at the scale of 1:24,000.

With all digital files properly re-projected, the teams began to review changes in the study area from 1978 to 2000, based on differences between the early maps and the current fieldwork.  Each team recorded updated characteristics for the study area, codifying the current environment in map format using ArcView GIS methods.  This process is described in much more detail in the Data Collection and Analysis Methodology section.  You can use the links below to jump straight to its various subsections.

[Topography] [Geology and Geomorphology] [Soils] [Land Cover and Vegetation]
[ Roads] [Water Resources] [Cultural Features] [Land Uses and Practices] [Home]