Minnesota Department of Transportation
Minnesota Department of Transportation

511 Travel Info

golden field


Minnesota Statewide Archaeological Predictive Model

MnModel Home | Data | Procedures | Results | Implementation | Terms of Use | Contact


Quick Links





















Mn/Model Research Design
Appendix C: Archaeological Field Survey Methodology


Mn/Model logo



The first phase of development of a predictive model is the acquisition of additional archaeological site survey data collected within a probability based framework. This new data must be compatible with the Minnesota Statewide Archaeological Survey conducted during 1977-1980, yet rectify earlier field sampling and survey problems. The field procedures are designed to survey the required sample parcels cost-effectively and efficiently, while meeting state guidelines for field survey, and recovering all necessary information for the predictive model. Several crews of experienced and qualified personnel from three institutions will conduct the survey. Standardized field methods and documentation will be employed to ensure comparable results, with quality controlled and assured through continuous feedback to a central Survey Coordinator.


Archaeologists from three institutions, BRW, the Mississippi Valley Archaeology Center (MVAC), and the Leech Lake Heritage Sites Program, will conduct the survey. Dr. James Gallagher, MVAC, will serve as the Principal Investigator for the field survey. Dr. Constance Arzigian, MVAC, will serve as Survey Coordinator and Laboratory Director, responsible for establishing and monitoring survey procedures and standards to ensure that comparable, appropriate and adequate data are being collected by all crews. She will be available on a daily basis throughout the field season to deal with problems that may arise. Mark Cassell, MVAC, will serve as Quality Control Monitor, responsible for monitoring survey crews, evaluating survey execution and maintaining the comparability of data acquisition. To provide a control on the variability of survey procedures, he will conduct an evaluation of the relative efficiency of different methods.


Field and Laboratory Procedures

Prefield Investigations

The project team will select survey parcels to meet the demands for the predictive model. Each parcel will consist of a quarter-quarter section of land totaling 40 acres, and will be assigned to a particular crew based on geographic proximity. The state site files, historic maps and documents, and SHPO historic contexts and other records will be consulted to identify any known prehistoric or historic sites in the sample parcels. However, the identification of known sites would not preclude field survey to verify the information or expand knowledge of the site. Each parcel will be identified on topographic maps and air photos to evaluate surface conditions and exposure. Based on county soil surveys and the geology and geomorphology of the parcel, soil types and the depth to the subsoil can be identified. This initial examination of parcels will determine vegetation cover and permit identification of the areas to be examined through surface reconnaissance and those that will require shovel testing. The fieldwork can then be scheduled appropriately. Surface reconnaissance will be most efficiently conducted in the spring or fall when vegetation is low, but with shovel-testing appropriate throughout the field season. The Survey Coordinator will review each parcel with the crew chiefs before the fieldwork to identify any potential problems and ensure that uniform procedures are being followed.


The landowners and tenants of the selected parcels will be identified through plat books and county records. Early in the spring, they will be contacted by a letter explaining the purpose of the survey and requesting written permission to survey and collect artifacts. A self-addressed stamped return envelope will be included. If the landowners request that artifacts be returned to them, arrangements will be made for that disposition at the end of the project. A phone call will follow for landowners that do not respond to the letter, requesting access. If permission is given, the crew chief will discuss surface exposure, accessibility, any special conditions on access, and identification of the specific plot in the field. The appropriate time for the survey will be arranged with the landowner. Identification of specific 40 acre parcels within largely undeveloped or forested land may require an advance visit by the crew chief to locate the sample parcel. GPS will help in locating parcels in such areas.


The landowner will be asked for any information he/she may have on prehistoric or historic artifacts found in the sample block. Sites identified only through collector information would not be included in the predictive model, although the information might be useful as non-probability based data. Each landowner will receive a follow-up letter acknowledging their support of the project. Those landowners that request to keep the artifacts from their land will receive them back at the end of the project, after full documentation has taken place.


Field Survey Standards Procedures

The Survey Coordinator, in conjunction with the Minnesota State Historic Preservation Office, will establish survey standards that will meet at least the minimum guidelines for archaeological survey in Minnesota according to the SHPO Guidelines for Archaeological Projects in Minnesota. As with the 1977-80 survey, even isolated single artifacts will be considered archaeological sites for this phase of the project (Minnesota Historical Society 1981:11). In areas of adequate surface exposure, such as in cultivated fields, surface reconnaissance will be conducted along transects no more than 30 meters apart, examining all the surface visible from that vantage point.


Shovel testing will be conducted at 15 meter intervals within transects 15 meters apart, where vegetation cover prohibits adequate surface visibility. Shovel test holes will be approximately 40 cm on a side, with all sediments screened through 1/4 inch mesh screens. Holes will be excavated to sterile soil, and then backfilled. Standard information will be recorded about each shovel-test hole, including depth of excavation, whether the soil profile matches that defined as typical for that parcel, and artifacts found. All cutbanks and other exposures within each sample parcel will be examined for evidence of cultural material, but systematic ground testing for deep, geologically-buried sites will be conducted in a later phase of the project. Caves and rockshelters will be examined for evidence of occupation and of rock art.


Site boundaries will be defined based on the distribution of surface material or positive shovel test holes. Wherever appropriate, site boundaries will be correlated with natural features such as knolls. Separate topographic features will generally be identified as separate sites, while a continuous scatter of cultural material on one topographic landscape surface would be considered one site. Site boundaries will be drawn in the field on topographic maps and on air photos if available at a scale of 8 inches to 1 mile.


Field crew composition will be appropriate to the institutions involved, but each crew is envisioned to be about six members, including the supervisor. The total number of crews will be decided by the number of parcels to be surveyed and the time available.


A variety of standardized forms will document the survey procedures and results. One form will document each parcel surveyed, and will include the methods used for each portion of the parcel and information on the ground cover, topographic setting, and identified sites. It will also document reasons for any areas not surveyed, such as steep slopes, evidence of historic disturbance, or burial by thick colluvial or alluvial deposits. Maps will show the location of all positive and negative shovel test holes. Each parcel will be described through a range of environmental parameters appropriate for the predictive model such as landscape setting, sources of water or other raw materials, and soil type. Many of these are available from the topographic maps but can be verified in the field with any additional qualifiers provided. Other information may not be available except from field observation. For example, the presence of lithic raw materials in the vicinity can be noted.


Identified sites will be reported on a field form that will then be used to complete the Minnesota Historical Society site reporting form.


Field computers will be used where appropriate to expedite data recording and transmittal. For each identified site the survey will document the size of the site and its relation to topographic features, and, if possible, the historic context of the site. All sites will be plotted on USGS topographic maps.


Integrating the work of multiple field crews will be critical for maintaining the quality and reliability of the survey results. To deal with this situation, during the first week of survey, all of the field crew supervisors and senior crew members will participate in a Procedures Workshop. Together they will survey a variety of sample parcels with different vegetational and environmental parameters. This joint survey will allow the supervisors to standardize field procedures and become familiar with the survey forms, customize the procedures for different field conditions, and resolve potential problems. Laboratory procedures will be standardized in the same way through completion of a standard inventory form, using a standard manual of terms. Weekly throughout the field survey, copies of all forms will be submitted to MVAC for final review. In the lab, all artifacts will be examined by the Survey Coordinator to ensure that they were correctly identified, that there is uniform interpretation of terms, and that all documentation is complete. This will be an ongoing process so that any problems can be identified and corrected while crews are still out in the field.


Artifact Collection, Analysis and Curation

To prevent false positive records, that is, to avoid reporting sites where none exist based on inappropriate artifact identification or field judgment, to ensure that site historic contexts are correctly identified from recovered artifacts, and to minimize the time in the field spent examining and documenting potential cultural materials, a sample of artifacts will be collected from each site. To reduce the volume of artifacts recovered and the time expended in laboratory analysis, while maximizing the number of parcels that can be surveyed, ten artifacts will be collected from each potential site. These will be the artifacts that convey the most information about the site, such as diagnostic lithics or ceramics. Where diagnostic artifacts are not present, a sample of the cultural materials will be collected to verify the actual presence of a site. The artifact sample will convey the variety of cultural items present, such as the range of lithic raw materials.


All artifacts recovered from the shovel tests will be identified by shovel test number, collected, and returned to the laboratory for processing. Information on the total abundance and range of cultural materials present at each site will be documented in the field, with photos of other artifacts if necessary. If nonportable features are encountered, such as an historic house foundation, black and white photos and color slides will be obtained as documentation. In the lab, artifacts from each site will be cleaned, inventoried, analyzed, and their cultural affiliation identified when possible. Artifacts will be acquisitioned and documented as appropriate, for final curation at the Minnesota Historical Society. Those artifacts that will be returned to landowners at the end of the project will be documented and photographed.


The survey data will be available immediately for use in formulating the predictive model. A final report will document the survey methods and results. In addition, there should be considerable public interest in the results of the survey, and local press releases and public presentations to local interest groups such as county historical societies will be scheduled. Local press releases early in the project will alert local residents to the survey, and prepare them for the letters requesting access to their land. A brochure will be prepared for wide public distribution at the completion of the survey, briefly explaining the survey, its purpose and results, and the contributions made by the study.


Quality Assurance and Control

Archaeological data collection and the manner in which it is conducted form the empirical basis for the entire project. Consistency and comparability in data collection methods are key factors in achieving the predictive model generation goals of the project. It is important to have mechanisms in place from the beginning to permit ongoing assessment of methodological consistency and comparability. This is particularly important given the large scale, broad scope, and potentially complex levels of data integration of the project. Many factors can compromise the high level of comparability and consistency required. The goal of quality assurance and control, then, is to eliminate as much as possible these factors, identify sources of unavoidable variability in data collection and develop strategies to explicitly deal with them. Whatever variability may be found in cultural resources between regions or environments should be a product of a diversity of past human land use, not a product of a diversity of current archaeological methods.


To ensure the quality, reliability and consistency of all data acquired during the field survey, several procedures have been established. First, uniform and acceptable standards and procedures will be established and monitored throughout the survey by the Survey Coordinator and the Quality Control Monitor. Consistency in data collection will be initiated with a prefield Procedures Workshop for all senior field and laboratory personnel to develop and learn standardized survey methods and recording formats. This workshop will ensure that established procedures will be appropriate to the diversity of crews, environments and field conditions anticipated throughout the state survey. All personnel will be made aware of the need for methodological homogeneity. The Quality Control Monitor will visit each crew in the field at least once a month throughout the survey to monitor consistency and reliability of field data acquisition. The Survey Coordinator will be available on a daily basis to help resolve problems.


Second, a series of standardized forms for parcel survey and site description will be used to ensure that all appropriate data are collected in a consistent and timely fashion. This will minimize variability in individual recording styles (sample forms will be included in the first summer survey's field report). Third, the field records will be prepared on manual paper copy and transferred to electronic medium (computer) as soon as possible. This guarantees as much as possible that data will not be lost should one or the other medium fail. It will also make data manipulation and verification easier and more timely. Fourth, to ensure efficient, controlled, and consistent acquisition of data in the field, copies of all field records will be mailed weekly from the field to the Survey Coordinator at MVAC for immediate review.


Fifth, to ensure consistent interpretation of recovered artifacts and to avoid reporting "false positives" (recording sites where none exist on the basis of inappropriate/inaccurate artifact identification or field judgment), a sample of ten artifacts will be collected from all potential cultural loci found during pedestrian survey. All potential cultural material from shovel test pits will be collected. Material from all the potential sites will be examined in the lab by a single qualified person. This will minimize differences in individual interpretations in the field.


The final aspect of quality control addresses the inconsistencies between different field methods. It is expected that the differences in survey method based upon ground surface visibility will lend the most artificially introduced variability to the database. Survey methods will involve pedestrian walkover surface reconnaissance or shovel test pit excavation. Pedestrian survey is recognized as appropriate to areas with moderate to high surface visibility (e.g., plowed fields), while shovel test pit excavation is appropriate where surface visibility is low (e.g., forest floor). Both are guided toward resource identification. However, pedestrian survey under optimal conditions provides a complete (100 percent of the area covered) sample in a given transect over a straight-line distance of 15 meters plus a conservative estimate of one meter viewed on either side of the transect (30 m2), while shovel test pits provide only a 0.3 percent sample of the same 30 m2. Despite the fact that shovel test pits sample that small space very intensively through excavation below the surface, the likelihood exists for the identification of more and smaller resources with pedestrian survey in plowed fields than with shovel test pit excavation in forested areas. This would potentially skew the results in terms of resource size and frequency to those areas where pedestrian surveys were conducted and away from those areas where shovel test pit surveys were the norm. On a statewide basis, this could impose a skew to the south and central regions.


In order to compensate for this potential resource identification skewing, an appropriate sample of parcels in each region and environment will be surveyed twice, using both pedestrian walkover and shovel test pit excavation. The second survey will be conducted by a single, consistent roving MVAC crew directed by Mark Cassell. This will permit an evaluation of the relationship between resource identification rates and survey methodology, and, if necessary, provide a weighting figure to weight resource frequencies when appropriate to compensate for underreporting sites through shovel test pit excavation or poor surface visibility.


In order to make earlier survey data, collected under different conditions and standards, more comparable to the current survey, it would also be useful to reexamine a sample of parcels initially examined during the Minnesota Statewide Survey of 1977-1980. Although providing an abundance of useful information, the Statewide Survey 1) used a 50 m interval between survey transects and shovel-tests along transects, and 2) employed a variety of stratification procedures as described in Minnesota Statewide Archaeological Survey Summary: 1977-1980 (Minnesota Historical Society 1981). Since earlier field sampling strategies were not as rigorous as those proposed for this survey, and the potential exists for different site identification rates, particularly within environments not included in the present project. The results would further provide consistency and comparability between the 1977-80 database and the data derived in this project.


Because only a small sample should be required, it is not expected that the entirety of the fieldwork period will be consumed by this roving crew in the quest after quality assurance and control; they will be engaged in basic parcel survey for part of the time. It is expected that the contributions of this work will be felt most strongly when the survey results are integrated into a GIS framework and subsequently into a usable predictive model.


References Cited

Minnesota Historical Society
1981 Minnesota Statewide Archaeological Survey: Summary 1977-1980. Minnesota Historical Society, St. Paul.





MnModel Orange Bar Logo


MnModel was financed with Transportation Enhancement and State Planning and Research funds from the Federal Highway Administration and a Minnesota Department of Transportation match.


Copyright Notice

The MnModel process and the predictive models it produced are copyrighted by the Minnesota Department of Transportation (MnDOT), 2000. They may not be used without MnDOT's consent.