The Precambrian record of southern Colorado: an integrated geologic database in ArcGIS
Department of Geological Sciences, UT Austin
GIS & Water Resources Project Report
Table of Contents
Overview and Project Goals | Data Sources and Progress | Output and Analysis | Summary and Continuing Work | References
OVERVIEW and PROJECT GOALS
There has been a tremendous amount of geologic investigation in Precambrian rocks exposed both across the state of Colorado and throughout the Rocky Mountains in general. These rocks are still sources of great interest to the geologic community because they represent the earliest existence of our continent. They hold information about how large continental masses are assembled and stabilized, the internal structure of the earth's lithosphere, and other deep crustal processes that cannot be directly studied in more modern and recent orogenic systems like the Himalayas and the Tibetan Plateau. In preparing geologic maps, field geologists rely on traditional hand drafting or, more recently, use computer graphics software (i.e., Illustrator, Canvas, Corel) to produce two-dimensional maps. These maps often contain exquisite detail and can be quite elegant, but they can obscure the various data and observations that underlie or accompany the mapped features. These data are typically published, but culling through the literature can be a laborious task. Furthermore, finding the most up-to-date information can be even harder. My main objective for this project was to use ArcGIS as a tool to bridge geologic mapping with all available analytical data (both published and unpublished) to produce an up-to-date, data-rich, and flexible map document for Precambrian rocks in southern Colorado. Such a map product would be a tremendous resource to all geologists studying these kinds of rocks in this region, and ideally it might serve as a model for future work integrating geologic mapping with other types of geologic data.
This particular project focuses on a part of southern Colorado contained within the Pueblo 1x2 degree topographic sheet and the northern half of the Trinidad 1x2 degree sheet. The geographic features contained within this area include the northern Sangre de Cristo Mountains, Wet Mountains, Arkansas River Gorge, and Salida area. While the Precambrian encompasses a tremendous span of geologic time from the beginnings of the earth ~4.7 billion years ago (Ga, or Giga annum) to the explosion of life ~550 million years ago (Ma, or Mega annum), exposures in southern Colorado are limited to a part of the Precambrian called the Proterozoic (~2500 Ma - 550 Ma). For my dissertation research and this project, I am most interested in identifying and constraining tectonic events (magmatism, deformation, and metamorphism) that occurred in the Paleoproterozoic (ca. 1700 Ma) and in the Mesoproterozoic (ca. 1400 Ma) across the Rocky Mountains. The project will contain three main elements. First, it will contain a geologic basemap for the area that shows the various Precambrian rock units exposed and contains some information about those exposures (unit name, rock type, etc.). Second, the map will contain elevation data, most likely in the form of contours of various intervals. Third, the map document will contain a series of databases that contain information like radiometric ages, isotopic data, and the like for various locations across the area. Once complete, the map will not only contain the traditional geology and topography that is present in traditional geologic maps, but a wealth of analytical data will also be displayed spatially and allow for point-and-click identification of geologic features and associated data. If all works well, this map document will represent a large step towards bridging the gap between traditional geologic mapping and the tremendous body of published and in-progress data and observations that underlie the mapping. I do not foresee the map ever being "finished" because our knowledge base continues to grow and our ideas evolve. However, I think one of the real benefits of a project like this will be its flexibility in that the map should be relatively easy to update as new data emerge and interpretations change.
DATA SOURCES and PROGRESS
The main sources of data for this project are 1) the geologic map of Colorado in ARC/INFO format (U.S. Geological Survey Open-File Report OF-92-507); 2) digital elevation data for a selected region of southern Colorado downloaded from the National Elevation Dataset; and 3) database files and web pages constructed with information culled from published geologic literature and from other unpublished research.
The geologic map of Colorado was a challenge to work with because ArcGIS 8.2 was unable to read the Arc/INFO macros included with the downloaded data. These macros are intended to rebuild the map including scales, index, and other accessory features from root data files through a series of Arc/INFO commands. Instead, I had to build a new geodatabase and add in the downloaded files as feature classes containing a large variety of data (basically 1 feature class containing all of the map data). Once I added all the data into the map, I began querying the feature classes and exporting different data types into individual feature classes. This was a tedious exercise, but it allowed me to create a map document with more precise layers that could be turned on or off depending on one's map needs. The most important feature class I extracted was the one containing only Precambrian rock units exposed across the state. I was also able to color the Precambrian units with a simple scheme that is consistent with published maps (e.g., Tweto, 1979) and allows quick delineation of Archean, Paleoproterozoic, Mesoproterozoic, and Neoproterozoic components. Click on the thumbnail below to see the Precambrian geology of Colorado.
Map of Precambrian exposures across Colorado
Once I established the basemap, I built a number of databases with data and information gathered from the published literature (e.g., Bickford et al., 1989a; Bickford et al., 1989b; Sabin, 1994) and some that I have collected and generated myself as part of my dissertation research. The most critical data are radiometric ages, and ages from the U-Pb system are particularly important because of high closure temperatures for Pb retention (~900 degrees C) and relatively slow rates of radioactive decay of parent U to daughter Pb. These data tell us precisely when certain rocks were formed and/or metamorphosed, and there is a wealth of high quality data across the area. Thus, the main table includes individual sample numbers (essential as a unique identifier when relating tables in ArcGIS), rock name and type, age with errors, radiogenic system, any descriptive information available, and the appropriate reference. Other tables include Sm-Nd and Rb-Sr isotopic data along with sample identification, etc. All of these tables also include latitude and longitude for the samples taken either from published maps or from field measurements. These geographic coordinates allowed me to import the data tables into the map document and to project the sample points properly. Click on the thumbnail below to see sample tables. In the web pages constructed for different areas across the study region, I included field photographs of the different rock types, larger scale photographs of distinctive outcrop features, and plots summarizing the various age data and isotopic data. To view a sample web page, click here. I will discuss linking the web pages with the map document in the following section.
Sample geochronology and isotopic data tables for southern Colorado
I initially planned to add topography to the geologic basemap through commercial digital topographic map software put out by iGage Mapping Corporation. Standard U.S. Geological Survey 7.5' quadrangles were ideal because these are the standard basemaps for most geological mapping, and they include a variety of practical information beyond detailed topography like forest and wilderness boundaries, towns, hydrography, etc. However, these bitmap images, once seamed to cover my area of interest, took up cumbersome, and in some cases impossible (gigabytes), amounts of space. Even seaming lower resolution 1:100,000 and 1:250,000 maps still created huge bitmap files. As a result, I turned to the National Elevation Dataset and downloaded digital elevation data covering part of southern Colorado and derived topographic contours from these data using the Spatial Analyst extension in ArcMap. I was initially turned off by the frustration of downloading 10 Mb parcels of data through a fairly convoluted interface and ordering process. However, I recently discovered an new enhanced version of the Seamless Data Distribution System that allows downloading of unlimited quantities of DEM, SRTM and NLCD data (as big as you are willing and able to download and store) in real time. I was able to download a split block of data that covers my entire area (~100 Mb), whereas previously it would have taken 10 or more 10 Mb parcels of data and the associated download problems. I also downloaded a small block of Shuttle Radar Topography Mission (SRTM) elevation data in an area of the Sangre de Cristo Mountains where I am doing my dissertation research to qualitatively compare the quality of the two datasets. Click on the thumbnail below to view some examples of the DEM data and Spatial Analyst products with a brief discussion of the two datasets.
Southern Colorado digital elevation data and ArcMap Spatial Analyst products
OUTPUT AND ANALYSIS
One benefit of the project that became clear rather quickly was its flexibility. As I mentioned earlier, I want this map document to include the most current information. If used as a work in progress for researchers working in the Precambrian of southern Colorado, a geodatabase could conceivably contain the most up-to-date information available, even if some or much of it has yet to be published. The published Geologic Map of Colorado (Tweto, 1979) has a number of features that are out of date, and these features were easily changed and updated based on published information in ArcMap using the editor toolbar. For example, the Oak Creek pluton in the northern Wet Mountains was originally mapped as a Paleoproterozoic intrusion (~1.7 Ma) based on its texture and appearance. However, it has since been dated by Bickford et al. (1989a), and they demonstrate that the pluton crystallized at 1436 Ma. Simply changing the value in the "UNIT" field of the attribute table from Xg (Paleoproterozoic granite) to Yg (Mesoproterozoic granite) made the update complete and changed the symbology of the polygon representing the granite to the appropriate color.
Once all of the different data components were created and properly formatted, I merged them all into one map document and began testing for consistency among the various datasets and to see what was possible within the ArcGIS framework. To a first order, the map itself is incredibly interesting itself regardless of the amount of data embedded or contained within it. With a simple look at the map, one can tell whether exposures occur in steep cliffs or subtle hills and where the best point of access might be. Published regional-scale maps rarely include detailed topographic information. The variety of contour intervals produced also allows for analysis at multiple scales. Because the elevation data are contained within the map document, one could easily perform a number of different analyses. For example, if someone were interested in geomorphology rather than Precambrian geology, that person could conceivable analyze the relationship between rock type or rock age with average elevation or slope. The data also allow field geologist to recognize more practical mapping issues like whether outcrops occur along grassy slopes or in steep cliff faces.
When all of the components of the project came together, I was thrilled to discover what was possible in ArcGIS. By using the identify tool in ArcMap, one can click on a particular geologic feature and immediately see what unit it is mapped as. Where there is a data point, one can click on the point and quickly see the sample information, rock type, outcrop descriptions, and the rock's age and associated errors. If there is isotopic data associated with the sample, this information is related and shows up in the identify box as well. Ideally, with a few clicks one could see the age, initial Sr ratios, epsilon Nd values, and Nd model ages for one sample all in the context of a geologic map. One can also query the map to isolate features of interest and export them as individual map layers or query the attribute tables to isolate particular values or check the statistics. Some sample results of such analysis is discussed briefly below. For a graphical example of this discussion, click the thumbnail below.
Identifying map features and related data in ArcMap
I was able to take the project a step further when I finally figured out how to hyperlink in ArcMap. I built web pages containing photographs and data plots for various samples across the area, and I was able to link those through the identify table in ArcMap. As a result, for certain samples, one can right-click on the sample identifier in the identify box and see the hyperlinks that are associated with the feature. So not only does the map contain abundant data, but it is also linked via the web to resources like field and thin section photographs, data plots, and other useful types of information. If all of the features in the map do not have hyperlinks, it can be difficult to recognize which features do have associated web pages. However, there is a hyperlink tool in the identify toolbar that looks like a lightning bolt that one can use to identify which features in a map document have associated hyperlinks. For a graphical example of hyperlinking in ArcMap, click the thumbnail below.
Linking features in ArcMap to the world wide web.
Finally, I want to include a hint of the capability of this project to answer particular geologic questions. I mentioned earlier that I was interested in identifying and constraining the occurrence of tectonic events during the Paleoproterozoic (ca. 1700 Ma) and the Mesoproterozoic (ca. 1400 Ma). By running a quick statistical analysis of the attribute "AGE_MA" of the geochronology features, it becomes clear that there are distinct pulses of magmatism and/or metamorphism during these two time periods with a distinct hiatus in between. It also is clear that the peaks of activity are centered around ~1435 Ma and ~1670-1690 Ma. This is very powerful information gathered rather quickly from the attribute table. Furthermore, by querying the data by age distribution, it is possible to select and export rock units with these ages to look at their distribution across the region.
Distribution of Paleoproterozoic (ca. 1700 Ma) rocks across Colorado.
Distribution of Mesoproterozoic (ca. 1400 Ma) rocks across Colorado.
These analytical outputs are just a hint of what is possible with such a data-rich and interactive map document, and I look forward to exploring its capabilities in more detail.
SUMMARY and CONTINUING WORK
I hope it is clear from the discussions above that ArcGIS enables the production or compilation of up-to-date, data-rich, and flexible map documents. There is a tremendous amount of effort that goes into creating such a project, but its benefits are endless. The ability to display analytical data in the context of geologic mapping is critical in itself, but the added abilities to display the data and mapping with elevation data and to link the data and maps to other types of information via the World Wide Web provides endless opportunities for enhancing research capabilities and geologic mapping and modeling. Not only does this database contain the current state of knowledge, but I also believe that it will be relatively easy to update as new data and more refined studies emerge. This map and database and others like it can provide user-friendly access to a wealth of geologic information that otherwise would have to be culled from the published literature and allow construction of maps that can be themed to address particular research questions.
This project will continue to be a work in progress as new data and interpretations will continue to emerge. I hope that it can also be expanded to include the Precambrian of the entire state of Colorado and perhaps ultimately the Rocky Mountains and southwestern United States. Furthermore, I hope to add other types of data including radiometric age data for all systems available (e.g., U-Pb zircon, U-Pb monazite, Ar-Ar, Pb-Pb), geochemical and additional isotopic data, and perhaps some interpretation of tectonic or geologic environment. I would also like to explore ways to incorporate more detailed mapping (i.e., 1:24k or 1:12k scale mapping) into regional-scale geologic maps. Finally, another critical issue involves the dissemination of these types of projects. This type of data product will be most useful if anyone interested has access. Perhaps publishing on CD-ROM or DVD-ROM might allow enough access, and these products could be updated and re-released as significant advances are made. The World Wide Web is probably the ideal venue for publishing and maintaining these types of databases, but we're not there yet. This project certainly demonstrates (at least to me) that field geologists should head in the digital direction.
Bickford, M. E., Cullers, R. L., Shuster, R. D., Premo, W. R., and Van Schmus, W. R., 1989a, U-Pb zircon geochronology of Proterozoic and Cambrian plutons in the Wet Mountains and southern Front Range, Colorado, in Grambling, J. A., and Tewksbury, B. J., eds., Proterozoic Geology of the Southern Rocky Mountains, Volume 235: Special Paper: Boulder, Geological Society of America, p. 49 – 64.
Bickford, M. E., Shuster, R. D., and Boardman, S. J., 1989b, U-Pb geochronology of the Proterozoic volcano-plutonic terrane in the Gunnison and Salida areas, Colorado, in Grambling, J. A., and Tewksbury, B. J., eds., Proterozoic Geology of the Southern Rocky Mountains, Volume 235: Special Paper: Boulder, Geological Society of America, p. 33 – 49.
Green, G. N. , 1992, The Digital
Geologic Map of Colorado in ARC/INFO Format: U.S.
Geological Survey Open-File Report 92-507, U.S. Geological Survey,
Sabin, T. G., 1994, Geochronology and isotope geochemistry of Early Proterozoic rocks exposed in portions of the Twin Peaks and Blanca Peak quadrangles, Alamosa, Costilla, and Huerfano counties, Colorado [M.S. thesis]: University of Kansas, Lawrence, KA, 94 p.
Tweto, O. L., 1979, Geologic Map of Colorado: Colorado Geologic Survey, scale 1:500,000.
Tweto, O. L., 1987, Rock units of the Precambrian basement in Colorado: U. S. Geological Survey Professional Paper, 54 p.
if you have any questions or comments, please email me.
back to GIS index