This project provides updated information about the harvests of salmon by the communities of King Salmon, Naknek, and South Naknek, Alaska. This project details the results of a household survey administered for the study years of 2017 and 2018 for harvests and uses of wild salmon by households in these Bristol Bay Borough communities. In addition to the results from the household survey data set, the resulting report includes information from in-depth interviews conducted with key respondents and during participant observation fishing trips. The three study communities are located along the Naknek River in Bristol Bay in Southwest Alaska. As in the past, during the 2017 and 2018 study years, many residents of these study communities relied on fishing for nutrition and to support their way of life. The household surveys found that, in both study years, subsistence harvests of salmon were important in the communities: more than 70% of the number of salmon harvested for King Salmon, and more than 80% of the salmon harvested for Naknek and South Naknek, were caught by subsistence net. Overall, the per capita harvests in 2017 were: 79 lb per capita in King Salmon, 138 lb per capita in Naknek, and 194 lb per capita in South Naknek. Sockeye salmon, followed by either Chinook salmon or coho salmon, composed the largest portions of salmon harvests (in pounds usable weight) for the three communities in 2017. In 2018, the harvests of salmon increased to 116 lb per capita for King Salmon, but decreased to 108 lb per capita for Naknek, and for South Naknek the per capita harvest decreased to 116 lb. Mirroring 2017, for study year 2018 sockeye salmon, followed by either Chinook salmon or coho salmon, composed the largest portions of salmon harvests for the three study communities. This study is part of the effort to collect data about the full range of wild salmon harvests and uses, and areas of harvest, to understand in all its complexity the importance of salmon as a subsistence resource. The project was funded by the Alaska Sustainable Salmon Fund (AKSSF). This information was collaboratively collected by research staff of the Alaska Department of Fish and Game (ADF&G) Division of Subsistence, and research staff from the Natural Resources Department of Bristol Bay Native Association.

The project was guided by the research principles outlined in the Alaska Federation of Natives Guidelines for Research and by the National Science Foundation, Office of Polar Programs in its Principles for the Conduct of Research in the Arctic, the Ethical Principles for the Conduct of Research in the North (Association of Canadian Universities for Northern Studies 2003), as well as the Alaska confidentiality statute (AS 16.05.815). These principles stress community approval of research designs, informed consent, anonymity or confidentiality of study participants, community review of draft study findings, and the provision of study findings to each study community upon completion of the research.

Systematic Household Surveys Data Collection

The primary method for collecting subsistence harvest and use information in this project was a systematic household survey. ADF&G finalized the survey instrument in December 2017. A key goal was to structure the survey instrument to collect demographic and salmon harvest and use data that are comparable with information collected in other household surveys in the study communities and with data in the Community Subsistence Information System (CSIS). Estimated salmon harvests by study community households are reported in numbers of fish and in pounds usable weight; the estimates include resources harvested by any member of the surveyed households during the study years. “Use” of salmon means any fish harvested, given away, or used by a household, and salmon acquired from other harvesters, either as gifts, by barter or trade, through fishing partnerships, or as meat given by fishing guides and non-local fishers. Additionally, the household survey included questions about gear types used to harvest salmon, harvest effort, and resource use assessment questions. The survey instrument also included a series of questions to gather observations about Naknek River salmon stocks and the household survey also included a series of questions about subsistence salmon permits for the Naknek River to address the study objective to evaluate the current harvest reporting and monitoring system that is based on subsistence permit returns.

Systematic Household Surveys Data Analysis

Surveys were coded for data entry following standardized conventions used by the Division of Subsistence to facilitate data entry. Information Management staff within the Division of Subsistence set up database structures within Microsoft SQL Server at ADF&G in Anchorage to hold the survey data. The database structures included rules, constraints, and referential integrity to ensure that data were entered completely and accurately.

Once data were entered and confirmed, information was processed with the use of Statistical Package for the Social Sciences (SPSS) software, version 21. Initial processing included the performance of standardized logic checks of the data. Logic checks are often needed in complex data sets where rules, constraints, and referential integrity do not capture all of the possible inconsistencies that may appear. Harvest data collected as numbers of fish were converted to pounds usable weight using standard factors.

ADF&G staff also used SPSS for analyzing the survey information. Analyses included review of raw data frequencies, cross tabulations, table generation, estimation of population parameters, and calculation of confidence intervals for the estimates. Missing information was dealt with on a case-by-case basis according to standardized practices, such as minimal value substitution or using an averaged response for similarly characterized households. Typically, missing data are an uncommon, randomly occurring phenomenon in household surveys conducted by the division. In unusual cases where a substantial amount of survey information was missing, the household survey was treated as a “non-response” and not included in community estimates. ADF&G researchers documented all adjustments. Harvest estimates and responses to all questions were calculated based upon the application of weighted means (Cochran 1977). These calculations are standard methods for extrapolating sampled data.

Mapping Locations of Subsistence Fishing

During household surveys, the researchers asked respondents to indicate the locations of their fishing activities during the study year. Project research staff established a standard mapping method. Points, lines, and polygons were used to mark fishing and harvest locations. Generally, points were used to mark harvest locations such as subsistence set gillnet sites. However, sometimes points were also used to designate a harvest effort location, especially if fishing from a riverbank. Some lines were also drawn in order to depict when the harvesting activity did not occur at a specific point; for example, lines were used to depict courses taken while trolling for fish. Polygons were used to designate areas where fish were harvested in a large area, such as while seining for spawned-out sockeye salmon in Naknek Lake. Harvest locations and fishing areas were documented on iPads using the Collector application (ESRI, or Environmental Systems Research Institute) customized for Division of Subsistence data collection needs. Once a survey was complete researchers conducted a quality control exercise by matching the map data to the survey form to ensure all map data had been documented. This was completed in the field before the surveys were submitted to the lead researcher. Once the data had been uploaded, researchers also verified that the household data were logged into the server. The data were first sorted by community, and then resource. Maps were then produced in ArcGIS 10.6.1 at the species-specific level for each study year separately.