GIS and Field-Based Spatiotemporal Analysis for Evaluation of Paleo–Ice Sheet Simulations*

*The author thanks his advisor, Dr. Jon Harbor, for support and guidance during this work. The author also expresses gratitude to his research team, including Dr. Yingkui Li, Dr. Alun Hubbard, Dr. Arjen Stroeven, and Dr. Johan Kleman. Without the efforts of this team, this work would have been impossible. The author would also like to acknowledge his committee members Dr. Bernie Engel and Dr. Ken Ridgway. This article was greatly improved due to the suggestions and comments made by four anonymous reviewers. This work was completed while the author was supported as a U.S. Department of Education GAANN fellow, and this material is based on work supported by the National Science Foundation under Grant OPP-0138486 to Harbor.

Abstract

Reconstructing paleo–ice sheets is significant for paleoclimate reconstructions and evaluations of sea level low stands. Accurate reconstructions of paleo–ice sheet dimensions and dynamics necessitate the combination of field evidence and process modeling. In this study, a GIS-based technique was developed to quantitatively assess model output against geomorphic data. However, implementation of this technique is not straightforward and requires consideration of time-space relationships, data representation, resolution, and analytical design. Combined use of two software tools holds considerable promise for the use, application, and interpretation of refined ice sheet models.

Key Words:

• GIS
• glacial geomorphology
• ice sheet reconstructions
• orientation
• spatial analysis

Notes

Notes: APCA=Automated Proximity and Conformity Analysis, LGM=Last Glacial Maximum, YD=Younger Dryas Period. The scores have been normalized so that a value of 1.00 indicates the best APCA score for that particular moraine, and 0.00 indicates the worst fit. APCA scores have only been summed for the estimated period of moraine development (see Figure 3 for locations of the moraines), as provided in the top of the table, and have been calculated using various dating techniques (e.g., see Boulton et al. 2001 and Rinterknecht et al. 2004 for more information on “time-slice” data).

*The author thanks his advisor, Dr. Jon Harbor, for support and guidance during this work. The author also expresses gratitude to his research team, including Dr. Yingkui Li, Dr. Alun Hubbard, Dr. Arjen Stroeven, and Dr. Johan Kleman. Without the efforts of this team, this work would have been impossible. The author would also like to acknowledge his committee members Dr. Bernie Engel and Dr. Ken Ridgway. This article was greatly improved due to the suggestions and comments made by four anonymous reviewers. This work was completed while the author was supported as a U.S. Department of Education GAANN fellow, and this material is based on work supported by the National Science Foundation under Grant OPP-0138486 to Harbor.