Data Management Plan

This dataset supports efforts to estimate the abundance and trends in population size of Alaska harbor seals. Annual surveys of harbor seal populations are fundamental to estimation of seal abundance, distribution, and trends, which in turn are essential for stock assessment, conservation, and management. The most feasible approach to determining harbor seal distribution and abundance is to use aircraft to count seals when they haul out of the water and are visible. Harbor seals in Alaska occupy a geographically extensive range from approximately long. 172ºE to 130ºW (over 3,500 km east to west) and from lat. 51ºN to 61.5ºN (over 1,000 km north to south). Estimation of the abundance of harbor seals statewide requires broad-scale aerial surveys and these surveys have been conducted by NOAA Fisheries, the Alaska Department of Fish and Game, and other collaborators since the early 1980s. This dataset reflects counts of harbor seals from surveys conducted in Alaska between 1998 and 2002.

During this period, the range of Alaska harbor seals was divided into five survey regions and only one region was surveyed each year. This approach allowed observers to survey the entire coastline of each region, sometimes twice per year, to identify new harbor seal haul-out sites, and for individual haul-out sites to be surveyed multiple times in a given year, weather permitting. This intensive effort produced a lot of data for each region in a given year, but surveying the entire range to produce state-wide abundance and trend estimates required a period of 5 years.

All coastal habitats (intertidal and offshore rocks and reefs) of Alaska south 61.5 degrees North or Cape Newenham in Bristol Bay

Lineage Statement:
During 1998-2002, aerial survey observers used paper maps and/or rudimentary GPS units for in-flight navigation and data recording purposes. Photographs of seal haul-out sites were primarily taken with SLR cameras shooting on rolls of 35 mm film which were later processed as slides. Observers were instructed to record the location of new haul-out sites and note when no seals were seen at known haul-out sites (i.e., record "zero" counts); however, this was not done in every case. Seal counts were obtained by projecting slides on a white board and counting each individual as it was marked. Count data for each year were recorded in separate spreadsheet files. During 2004-2006, an effort was made to retroactively fill in missing "zero" counts by comparing observer's notes and GPS tracks against locations of known haul-out sites and photo counts to identify sites that likely had been surveyed but had no data recorded. Count records for each haul-out site were also associated with newly-developed spatial survey units which were designed to simplify the data collection and analysis processes for this project. Count data for all years were given additional QA/QC procedures and combined into a single Access database.

Data represented in this dataset are the result of a culmination of multiple observations, data collection protocols, and spatial analyses. Throughout the process, various checks are performed to insure data are of the highest quality. Despite these quality control checks, users of these data must understand the data are often collected under challenging field conditions and additional processing and testing must be done before these data can be used appropriately. This dataset has been the foundation of Stock Assessment Reports published by NOAA Fisheries. As such, the data have been subjected to quality control and quality assurance procedures associated with the statistical analysis and both internal and external peer review processes.

The data set is in the process of being archived with the NOAA National Centers for Environmental Information. Once the archival process is complete and verified, the data set will be publicly available.

IT Security and Contingency Plan for the system establishes procedures and applies to the functions, operations, and resources necessary to recover and restore data as hosted in the Western Regional Support Center in Seattle, Washington, following a disruption.