ACCAP Pilot Project : Sea Ice

Sea ice conditions affecting Alaskan coastal communities, marine ecosystems and offshore transportation

Image from John Walsh

Diagram from Hajo Eicken

Alaska has 6,640 miles of coastline, more than that in the rest of the U.S. Alaska is also the only state in which large portions of the coastline are affected by sea ice. Sea ice is present along or close to the northern coast for 8-10 months of the year, and it affects much of the western coastline for at least several months of most years. The presence of sea ice is a major factor in the lives of many western and northern Alaskan coastal communities, for whom a stable ice cover is essential as a buffer against coastal storms, as a platform for offshore activity, and as a marine environmental feature essential for the survival of higher trophics such as walrus, polar bears and seals. Coastal flooding and erosion, exacerbated in recent years by the retreat of sea ice, has been highlighted in the Arctic Climate Impact Assessment (Impacts of a Warming Arctic, 2004, Cambridge Univ. Press, 140 pp.). In addition, information on present and forecasted sea ice conditions is vital for several of Alaska’s major industries: fishing, marine transportation and offshore resource extraction. These needs point to the importance of a synthesis of information on Alaskan sea ice conditions to serve the climate services and operational forecasting sectors, and, ultimately, stakeholders affected by sea ice.

Historical information on Alaskan sea ice conditions exists in a variety of forms and locations. However, the dispersed nature of this data calls for a coordinated synthesis into a comprehensive, well-documented database that is easily accessible by a variety of users. More importantly, the informational attributes most essential to users need to be determined and incorporated into any synthesis at an early stage. Our project begins with (1) stakeholder interaction to determine the sea ice information most important to users and to providers of sea ice products, followed by (2) identification of suitable component datasets on Alaskan sea ice, together with an assessment of their correspondence with user needs, and (3) synthesis of the appropriate subsets of sea ice information into a single Alaskan database that will best meet the user needs. Subtask (3) will be done in collaboration with database specialists and with appropriate NOAA partners (NWS/Alaska Region Headquarters, National Climate Data Center, and the Regional Climate Center serving Alaska). Subtask (4), collaboration with the National Weather Service and Climate Services sections to enhance forecast products pertaining to sea ice, will be undertaken as described below.

Photo from Canadian Coast Guard

User Needs

Users of sea ice information range from affected stakeholders (coastal communities, fishing and transportation companies, offshore mineral interests) to the NOAA agencies charged with providing real-time and forecast products pertaining to sea ice. While precise needs remain to be identified, it is readily apparent that needs vary widely among these different types of users. For example, coastal communities engaged in the hunting of whales, seals, walrus and polar bear are concerned about the extent and stability of fast ice attached to the coastline and shallow surface. Offshore oil and gas interests operate in a zone that includes not only fast ice, but also the areas farther offshore. Marine transportation interests are likely to be concerned with ice conditions well offshore of the coastline, particularly the distribution of thick and multiyear sea ice. Ecosystem studies require information on the character of the sea ice cover, especially as it pertains to marine productivity and to effects on the stratification of the water column. Most importantly, the seasonal (time) windows and specific locations of the required information will vary widely among users. We will hold two workshops: one workshop involving members of coastal communities of Alaska, and the other involving stakeholders in the commercial (including transportation) and industrial sectors. The product of these workshops will be a documentation of user needs, including types of sea ice information and the extent to which these needs are currently being met by NOAA and other agencies; specificity (location and timing); intensity of usage of existing and desired information; and potential value of enhanced information in planning activities.

Figure 1. Satellite passive-microvavederived sea ice concentrations in the Alaskan sector on 26 August 2008. Courtesy of Cryosphere Today, University of Illinois (http://zubov.atmos.uiuc.edu/CT)

Figure 2. Passive microwave-derived (SMMR/SSM/I) sea ice extent for the Northern Hemisphere expressed as departures from monthly means. Image courtesy of National Snow and Ice Data Center, University of Colorado, Boulder.

Identification of component datasets

It is well known that there exist significant sea ice datasets, ranging from single-point measurements to hemispheric products obtained by satellite remote sensing. Alaskan regional analyses produced by the NWS Alaska Regional Office (Anchorage). More recently, coastal radars have provided information on the characteristics of sea ice offshore of a particular location (e.g., Barrow, project leader H. Eiken). Perhaps the best known hemispheric products are the satellite passive microwave fields of sea ice concentration and the NOAA/Navy National Ice Center (NIC) analyses. Both these products date from the 1970s. Their resolutions are several 10s of kilometers. While issues of homogeneity are nontrivial, many of these issues have been addressed by quality-control activities at NASA Goddard and NIC. Both datasets have been used extensively by the scientific research community. However, accessibility by non-researchers is limited by the formats and the volume (global coverage) of the datasets. Figure 1 is a recent example of a satellite passive microwave depiction of sea ice concentrations in the Alaskan sector, while Figure 2 shows the monthly anomalies of total Northern Hemisphere sea ice extent extending back to 1978. The recent decrease of sea ice extent is readily apparent in Figure 2.

For applications pertaining to climate change, especially the placement of recent anomalies into a broader temporal context, the limitation of the global datasets to the post-1970s period limits the potential applications. A longer dataset would also enable planning and policy decisions, as well as actual sea ice forecasts, to be based on more robust statistics (trends, probabilities, ranges/extremes). Data for temporal extensions exist in the form of maps and charts compiled by NIC’s predecessors (Naval Oceanographic Office), by the private sector (e.g., the Dehn collection of charts now being digitized through NSIDC), and hemispheric analyses by international agencies (e.g., Danish Meteorological Institute, U. K. Meteorological Office). In addition, more local sea ice information for Alaskan coastal waters is available through archived break-up and freeze-up dates from coastal locations and ship reports; and other information may be held by Alaskan coastal residents. Sufficient information from these sources exists for a useful gridded database extending back to at least the early 1950s, thereby extending the database to more than 50 years. The adequacy and synthesis strategy for the first half of the 20th century remains to be determined.

We expect to hold a workshop of ice data specialists with the specific target of an integrated Alaskan sea database. Invitees will include sea ice data specialists from the National Ice Center (NIC), the National Snow and Ice Data Center (NSIDC), NASA Goddard, NWS Alaska Region Headquarters, the Alaska Ocean Observing System (AOOS), the National Climatic Data Center (NCDC), the Regional Climate Center responsible for Alaska, and others familiar with one or more databases. In addition, database specialists will be entrained to aid in the design of a database that is characterized by (i) multiple sources, (ii) varying temporal and regional specificity/resolution, and (iii) multiple measures of ice conditions (e.g., concentration, ice type, thickness, roughness, etc.). The objective will be to design a database structure that provides the optimum combination of accessibility and utility of information for users. We foresee this workshop occurring early in the second year of RISA activity, following the workshops convened to determine user needs.

Photos from Hajo Eicken