The U.S. Department of Defense operates numerous Arctic and subarctic installations, including Alaska. Climate warming is occurring and new methods are required for construction to insure longevity and robustness against degrading permafrost threats. This report outlines research conducted to establish new methods for ground ice detection and delineation, methods for early warning detection of thawing permafrost under infrastructure, and an outline of a decision support system to determine the most applicable foundation design for warming and degrading permafrost. These three tasks address the immediate needs to advance the ability for effectively constructing mission critical infrastructure on permafrost terrains, Arctic and subarctic.

The objective of this study was to provide a process or guide for site assessment and design selection using holistic methods. The methods required various steps and activities, as follows: (1) choosing the optimal location for infrastructure within a given site, (2) compiling existing information and acquiring data for determining promising foundation alternatives, (3) developing finite-element thermal modeling parameters for estimating foundation impacts to permafrost, and (4) assessing long-term monitoring options for identifying threats to infrastructure early on, while they can best be mitigated.

Technical Approach

To develop this systematic methodology, three test sites (Cold Climate Housing Research Center, Cold Regions Research and Engineering Laboratory (CRREL) Permafrost Tunnel, and CRREL Farmer’s Loop Test site) that represent various ranges of permafrost conditions and terrain in Interior Alaska were studied for site characterizations. Various tools and methods were used at the three sites to effectively characterize the subsurface information of the permafrost. Ancillary data for background information of the sites such as aerial photos and Light Detection and Ranging imaging revealed details of terrain, vegetation cover and type, and surficial features that suggest geologic origins, ground ice presence (polygonal ground), drainage, and other characteristics. Surface terrain features can indicate whether (or not) permafrost exists at depth. Geophysical instruments for continuous subsurface transects were conducted using electrical resistivity tomography, capacity coupled resistivity, and ground-penetrating radar. The borehole data and geophysical measurements provided the variation of permafrost or its ice content as a function of depth or areal extent. The ground soil type, the frozen or unfrozen state of the ground, and moisture content from the borehole core data provided the physical information to improve the correlation between ground resistivity and subsurface conditions at the specific location of interest.


This project describes and illustrates the Permafrost Foundation Decision Support System (PFFDSS) technological design, including the structure of the interface and the general system architecture. The current version of the PFFDSS is a fully constructed and functioning prototype, in which all three assessment phases are implemented and logically linked. The project team considers the tool to be an affirmative proof-of-concept that provides a solid foundation for continued refinement and expansion. It is the project team’s considered opinion that the prototype PFFDSS is not yet suitable for fully vetted site assessments. This is mostly due to the complex nature of the analyses and permafrost conditions at hand, coupled with the pure novelty of the approach. Nevertheless, iterative, long-term testing, refinement, and expansion of the tool that leverages progressive feedback from willing testers and beta-users will yield a reliable, practical utility that will serve a broad user base.


Based on the findings of this project, there is a critical need to do the following:

  • Refine and expand the PFFDSS. The current version of the PFFDSS is a fully constructed and functioning prototype that implements and logically links all three assessment phases. The tool is currently an affirmative proof-of-concept that accommodates continued refinement and expansion.
  • Generate and compile suitable parameters for finite-element thermal analysis. This work utilized finite-element thermal modeling to evaluate temperature in complex scenarios, such as beneath structures, to forecast temperature at various configurations. Thermal modeling is a complex and useful tool that provides feedback to the foundation designer to better account for the thermal impacts of their structure on frozen ground. Thermal modeling can be used to predict the permafrost table depth and to provide ancillary information to the PFFDSS tool.
  • Statistical analysis. This project defined the use of Bayesian-Gaussian statistical determination methods to progressively increase the probability of accurate permafrost characterization. The current method is computationally intensive, requiring much computational speed and time. Further refinement is needed to decrease computation time yet maintain the high degree of probability forecasting.
  • Develop a separate tool to assist engineers or designers to improve the long-term performance of infrastructure under conditions of increased future risk associated with climate change. This tool would be used to examine the Mean Annual Air Temperature (MAAT) value for predicting temperature projections and providing indicators of potential changes to permafrost conditions.