Distribution, growth, and condition of salmonids in the central California Current Ecosystem.

The Fisheries Ecology Division of NOAA’s SWFSC conducted annual surveys of salmon and their ocean habitat in the coastal waters of northern California and southern Oregon from 1998-2016. We used a surface trawl to collect juvenile and subadult salmonids, including several ESA-listed populations of Chinook and coho salmon and steelhead. We also quantified other coastal pelagic fish and invertebrates that co-occur with salmon, and we measured spatially matched biological and physical oceanographic variables. Juvenile salmon were frozen at sea and transported back to shore for further analysis. Scales, DNA, otoliths, stomach contents, blood plasma, and implanted tags (if present) were retained. The majority of older salmon and bycatch species were released alive at sea. Additional data recorded during our survey included seabird counts, plankton samples, echosounder readings, and CTD profiles of temperature, salinity, chlorophyll, transmissivity, and PAR.

The broad objective of this project was to measure the spatial distribution and physiological condition of salmonid stocks, including ESA-listed and non-listed populations, in the central portion of the California Current Ecosystem. Additionally, we measured oceanographic conditions and nekton community structure and related salmon abundance, migration, and condition factors to ocean habitat.

The raw data from this project are stored on desktop computers and servers at the FED laboratory in Santa Cruz, California. Available data formats are Microsoft Excel, Microsoft Access, and SQL. The original paper datasheets used to record data at sea are all still available and stored in archives at the SWFSC FED lab.

Contact Jeff Harding for direct access via ODBC to the database.

None

None

Data reside on an SQL Server 2008 database.

Our coastal ocean survey was based on samples collected annually at a small number of stations (60-80 fixed, predetermined locations at sea). Most stations were visited only once or twice per year, usually without repeated or replicate sampling on any given cruise. Our study area was large: appx. 750 km north to south, with widely spaced transect lines running from appx. 2 to 30 km out from shore. Therefore, our study design provided a long time series of coarse-grained snapshots of general ocean conditions (biological and physical) over many years. Due to the large distances between stations, the lack of replication, and the infrequency of cruises, the accuracy and precision of our population size estimates are low to moderate (the confidence intervals are large), however the effort employed to obtain these data is highly comparable to most other biological surveys of this type conducted by large research ships at sea. The accuracy and precision of physiological data obtained from the specimens themselves is higher, because most stations yielded numerous individual specimens (i.e. we have replicate samples and higher statistical power). The accuracy and precision of physical measurements is high, because the instruments themselves are accurate, reliable, and calibrated regularly.

As mentioned above, the accuracy of some biological data from this survey is low to moderate. Due to the large distances between sampling points, the lack of replication, and the long intervals between sampling (usually just one or two cruises per year), the confidence intervals on population size estimates are large. These data would probably not be sufficiently accurate to estimate a true population size for any of the species sampled. However, this was never the goal of the study. The sampling effort and protocols were highly consistent among years, and the data can be used appropriately to detect interannual trends (increasing, decreasing, or stable) and broad spatial patterns of abundance. The accuracy of physiological data is higher, because individual stations often yielded many specimens, providing replication and allowing for better estimates of statistical parameters, such as means and standard deviations.

Some level of bias is unavoidable in any study of this kind. For example, the catchability of different species by rope trawl and plankton nets varies depending on the size and swimming ability of the target animals, the mesh size and tow speed of the net through the water, sea conditions, and variation among vessels, captains, and crews. Bias was minimized by consistent use of the same sampling gears and operating protocols throughout the survey.

Our trawl nets and other sampling equipment is standard issue for numerous research surveys of this kind. We report catch abundance in standardized units (CPUE) allowing for limited comparisons even to other surveys using different gear types. Our laboratory analytical techniques are well described in the scientific literature and they are similar or identical to those used by other researchers in this field. Our instruments are commonly used standard tools for this field.

We completed most of our sampling activities as planned. We obtained a sufficient number of specimens (e.g. numbers of juvenile salmon, number of water samples, number of DNA samples) to conduct all planned laboratory analyses. On occasion, sampling stations were dropped due to unfavorable sea conditions. In some years (e.g. 2006, 2008, 2009) lack of funding and no allocation of sea time on research vessels caused gaps in the time series.,We completed most of our sampling activities as planned. We obtained a sufficient number of specimens (e.g numbers of juvenile salmon, number of water samples, number of DNA samples) to conduct all planned laboratory analyses. On occasion, sampling stations were dropped due to unfavorable sea conditions. In some years (e.g. 2006, 2008, 2009) lack of available sea time on research vessels caused gaps in the time series.

Due to general limitations on available ship time and days at sea, we did not usually conduct replicate trawls at individual stations, therefore our ability to measure the precision of population estimates with standard statistical tests is limited. Instrumentation used at sea (CTD profilers, EK60 sonars) was calibrated and tested regularly, and is considered accurate and precise. Physiological data were obtained under controlled, consistent laboratory conditions using carefully frozen or preserved specimens. Statistical measures of precision are presented, whenever possible, in our published research findings.

Analytical precision is generally high because measurements were made in controlled laboratory conditions using industry-standard equipment (e.g. scientific balances, flame-ion lipid chromatograms, etc.).

Due to general limitations on available ship time and days at sea, we did not usually conduct replicate trawls at individual stations, therefore our ability to measure the precision of population estimates with standard statistical tests is limited. Instruments used to measure physical oceanographic properties at sea (CTD profilers, EK60 sonars, water filters) were calibrated and tested regularly, and these data are considered accurate and precise.

Information about omissions, selection criteria, generalization, definitions used, and other rules used to derive the data sets can be found in our published research findings.

Data are collected at sea and recorded on paper datasheets by observers, or recorded directly to digital data files in the case of electronic instruments. Back on shore, paper datasheets are transcribed to computer files. All data files are proofread and checked for errors and completeness. Final versions are error-free.

Data checking and loading into final database.

Awaiting guidance from headquarters.

Data is backed up weekly and stored in a secure location.

Data are collected at sea and recorded on paper datasheets by observers, or recorded directly to digital data files in the case of electronic instruments. Back on shore, paper datasheets are transcribed to computer files. All data files are proofread and checked for errors and completeness. Final versions are error-free.