MnModel Phase 4 Goals
Since 1998, the Minnesota Department of Transportation has been using MnModel's Phase 3 predictive models to identify the potential for archaeological resources in its project areas. However, these models have known shortcomings, primarily because of deficiencies in the data used. The primary goal of Phase 4 modeling is to improve model performance through the use of better data.
The primary objective of the MnModel project was to create a series of accurate digital maps capable of alerting planners to the presence of potential precontact archaeological properties in accordance with the identification requirements set forth in Section 106. The use of the Phase 3 models has allowed MnDOT and other agencies to conduct advance planning and base decisions on sound scientific findings. In addition, the digital environmental and cultural resource data developed for MnModel have contributed to streamlining of environmental reviews, planning, and other activities at MnDOT. Phase 4 updating of MnModel's data and models will not change these activities, but should make them more effective.
Perhaps the most significant goal of Phase 4 is to improve the accuracy of the predictive models. Because of the availabiity of new and improved data, both environmental and archaeological, it should be possible to develop better models for much of the state. More accurate models will be characterized by one or more of the following:
- a higher proportion of sites predicted.
- a smaller area of high and medium site potential.
- a smaller area of unknown site potential.
- greater model stability.
MnModel's overall accuracy goal remains unchanged: to predict 85 percent of known archaeological sites while having 33 percent or less of Minnesota's area classified as high or medium site potential. However, the goal for each region will be to improve model performance based on the four criteria above. An additional goal for Phase 4 is to develop better methods for measuring model accuracy and performance.
More accurate models will provide Culture Resource professionals with a higher degree of confidence in the models, reduce the number of surveys required by decreasing the unknown area, and reduce the potential of finding sites during construction that were not predicted by the model.