Hay, D. E., McCarter, P. B., Daniel, K. S., and Schweigert, J. F. 2009. Spatial diversity of Pacific herring (Clupea pallasi) spawning areas. - ICES Journal of Marine Science, 66: 1662-1666.Eastern Pacific herring spawn in intertidal and shallow subtidal areas. Spawning sites are conspicuous: milt turns coastal waters white, sometimes for distances of many kilometres. This attribute has enabled biologists to document spawning distributions for more than 70 years throughout the 29 500 km coastline of western Canada. Spawning distributions and spatial diversity have varied over time. When aggregated over 70 years (1938-2007), spawning occurred along 5574 km or ∼20% of the total coastline. Cumulative annual spawn length ranges from 131 (in 1966) to 770 km (in 1992). We examined annual changes in spawn distribution using spatial units of variable size, ranging in area from a maximum of >1000 km 2 to a minimum of 0.3 km 2 ). The choice of spatial scale can affect the results from analyses of other factors, such as SSB, that could affect spatial diversity of spawning areas.
An approach of relating Pacific herring (Clupea pallasi) tag-recovery data to dispersal distances is presented. Observations from coded wire tag sampling (1999-2006) were used to represent adult herring interannual spawning patterns on the British Columbia coast. Six datasets were applied, differing by number of years-at-liberty (1, 2, or 3) and gear type of recapture (purse-seine or gillnet). In total, 227 tag-recovery samples, consisting of 5687 tag recoveries, were used. Distances were approximated to the shortest paths through water between the release and recapture sites. Recovery rate and distance relationships suggest that exponential models fit the data reasonably well, with average rates of change in recovery rates (slopes) varying from approximately -0.009 to -0.005. A combined slope estimate of -0.007 is similar to four of the six estimates. Using these models, the intensity of movement among five stock-assessment regions was estimated by applying distances relative to their centres. Fidelity estimates range from 53 to 90% across all models and regions, which is consistent with previous findings and premises that influence resource management. Interpretation and application of the modelling exercise are discussed in terms of previous and future work.
Nearly 1.6 million tagged herring (Clupea pallasi) were released in two separate programs (19361967 and 19791992) in British Columbia. Several thousand tags were released in each of 955 release sessions. Over 85% of the release sessions had subsequent recoveries. Almost 43 000 tags were recovered over all years. We re-assembled the tagging data into an electronic database, geo-referenced all tag release and recovery data, analysed spatial movements, and estimated straying and fidelity rates. The analyses do not wholly support the conclusions of previous work indicating high homing rates to local coastal areas. Estimates of fidelity, defined as the proportion of tags recovered in the same area as released, varied with the size of the geographic area used in the analyses. Fidelity rates are high for large areas, such as the Strait of Georgia (~10 000 km2), but lower for small geographical areas, such as inlets or bays (~100 km2). High fidelity is not necessarily evidence for "homing." Homing and fidelity are different biological processes and tagging cannot necessarily distinguish between them. Although fidelity rates for small areas are generally low, there are exceptions that may be evidence for the existence of biologically distinct populations in certain areas.
