Language selection


Research Document - 2011/064

Investigating Changes in Pacific Herring Spawn Density (layers)

By D.E. Hay, C. Fort, J.F. Schweigert, L. Hamer, and P.B. McCarter


This report examines biological, ecological and methodological factors that affect estimates of egg density, or egg ‘layers’ in spawning areas of the British Columbia coast. Recent literature indicates that the density of spawning herring depends on intrinsic behavioural mechanisms that apply to all pelagic schooling fish, including herring. If so, egg density may not vary directly with the spawning stock biomass (SSB). This report examines the relationship between egg density and SSB by analysis of the metrics (i.e. measurements) of individual spawning events, of which there are several hundred per year in BC. The important metrics are spawn length, spawn width and number of egg layers. Spawn metrics were assessed and compared with abundance trends in the five main herring populations in BC. In general, the metrics do not decrease or increase in unison with the abundance of populations. Even as all BC populations were in a period of contraction for the last 5 years, the dimensions of individual spawn events remained relatively consistent, although there were fewer spawn events. An exception to the apparent consistency of spawn metrics was egg density or egg layers. In BC the mean number of egg layers declined significantly in all coastal areas over the last two decades but the decline was especially sharp during the last 5 years. Mean egg layers declined in some regions even as the total spawning area increased. This decline in layers might have biological or methodological explanations that are examined in this paper. There is no obvious biological explanation and analysis in this paper shows that survey effort has remained relatively consistent. An unexpected observation was that in the last decade there has been an increase in the incidence of ‘trace’ observation – recorded arbitrarily and unrealistically as 0.01 layers. The incidence of the trace category was relatively rare in the early years of diver surveys but it has increased and now accounts for more than 40 percent of all observations in most regions in recent years.  It seems probable that this change in estimated egg layers is associated with a subtle and unintentional change in dive survey protocols. A recommendation is that the rationale for using the trace category as true estimate of layers should be re-examined. Other recommendations refer to potentially important details about the computational procedures used to derive the herring spawn index used in support of the assessment model.

Date modified: