An important outcome of Arctic Coastal Dynamics I was the segmentation and characterization of the entire circum-Arctic coastline by regional experts which is presented in this dataset. This dataset contains data on coastal morphology, composition, dominant processes, ground ice, and environmental forcing parameters such as wind speed, storm counts, melt season, and wave energy. A listing of the variables included in the coastal classification can be found in Appendix A of the ACD II Science and Implementation Plan (2006). This information is available for over 800 segments, covering the coastline of all eight regional seas of the Arctic Ocean. The length of individual segments is variable (median length is 38 km), and depends on classification parameters and data availability. The segmentation format is scalable, allowing the adoption of future digital coastlines and the integration of additional data at higher spatial resolution. An assessment of the data quality for the more important quantitative variables has just been completed and the data will be publicly released on an internet map server (IMS). The goal of the IMS will be to allow individual users to prepare their own maps displaying the region and variables of interest.The ACD Classification was conceived as a broad enough framework to encompass existing classification schemes while capturing fundamental information for the assessment of climate change impacts and coastal processes. The implementation of the classification was done by so-called "regional experts", who, based on digital and paper products and personal knowledge provided information which was subsequently gathered into a circum-Arctic coastal database. The classification was primarily geomorphological in nature and considered: (1) the shape or form of the subaerial part of the coastal tract, (2) the marine processes acting upon the coast, (3) the shape or the form of the subaqueous part of the coastal tract and (4) the lithofacies of the materials constituting the coastal zoneThe beta version of the classification is made of 1331 segments each characterized by a series of geomorphological quantitative and qualitative variables. The classification is stored as an ISO 19115-compliant personal geodatabase and is therefore mappable in off-the-shelf Geographical Information Systems (GIS)