short-term composition shifts during six years; three of those were El Niño years.
However, the abundance of dominant forms in her data, and thus copepods overall,
did decline in correlation with the Roemmich–McGowan pattern, especially in the
prolonged period of high ocean temperatures in the early 1990s. The same array of
zooplankton species appeared to have sustained substantially smaller stocks.
(^) Lavaniegos and Ohman (2007) again revisited samples from the Bight, just those
from one spring cruise each year. They combined splits of all night samples from the
mid-spring cruises of 45 springs, subsampled that mixture, then counted and
measured plankton groups. After converting to carbon biomass by length–mass ratios,
they concluded that the only statistically significant drop in abundance was due to
salps and doliolids. Possibly, but most of the detailed features of the displacement-
volume time-series recur in their total carbon estimates and those for copepods,
euphausiids, and chaetognaths. The year-to-year troughs and spikes mostly
correspond, respectively to warm El Niño years with advection of southern plankton,
and cold years with more extended advection from the north. The nature and
persistence of the stock rebound in the years after 1999 remains to be evaluated.
(^) Shorter time-series of nearshore zooplankton samples from Oregon and Vancouver
Island show both faunal and abundance shifts correlated with the PDO, even the PDO
variations within single years (Mackas et al. 2004). At these more northerly sites, the
shifts are caused by changes in the oscillation between southern zooplankton species
in winter, northern species in summer.
The Odate Project
(^) Chiba et al. (2006) report the results from a taxonomic analysis of the Odate
collection. A 1433 sample subset of the Odate collection, and some other collections
from similar net hauls, was selected representing the Oyashio (sites deeper than 500 m
in the vicinity of 42°N and colder than 5°C at 100 m) in months from February
through October of each year from 1960 to 2002. The wet-weight biomass (Fig.
16.23) of the whole samples varied by a factor of approximately three almost exactly
at the periodicity of the PDO. Copepods were identified and 59 of the more abundant
species were evaluated statistically. The species divide (by cluster analysis) into five
seasonal “communities”: species peaking in April (including Neocalanus cristatus
and N. flemingeri), May–July (N. plumchrus, Eucalanus bungii, Oithona similis,
Pseudocalanus minutus), July, September, and October. The last three groups have
warm-water associations and are likely to represent the frequent summer–fall
overwashing of the Oyashio by Kuroshio loops and rings. Chiba et al. characterize the
shifts of faunal composition as follows (paraphrased):
(^) The abundance of the spring community gradually increased from 1960 to 2002.
Changes of life-history timing coincided with the climate regime shift in the mid-