4-30 



earlicjr, (Jl arko-HumiJus (CB) mctliod was considerably Jess accurate. (!B 

 determinations were characterized by a high threshold density for taxa 

 detection (100 to 200 per m ) . Below threshold, taxa densities were 

 underestimated: when organisms were most abundant, densities were 

 overestimated. In addition, though the reverse also occurred, there 

 were far more instances when the CB yielded no organisms of a particular 

 taxon while the half-meter net (1/2 meter) indicated organism presence. 

 Lower sample volumes, higher clogging rates and, perhaps, higher avoidance 

 rates combined to make the CB method inconsistent in estimation of 

 organisms present at low densities, and less accurate overall. 



Comparisons of data derived from the two methods must be made 

 with care, and any comparisons of study results to be made which require 

 cross-method comparisons should be based on sets of data and not on 

 individual samples or stations. In broad terms, regression equations 

 show that CB estimates ranged up to 1.5 orders of magnitude higher than 

 1/2-meter estimates at the lower end of the density range observed in 

 the harbor. At highest observed densities, 1/2-meter estimates were in 

 the same range as those derived from CB samples . For copepod nauplii 

 and polychaete larvae, CB estimates were nearly always higher than 1/2- 

 meter estimates over the observed density range. For certain taxa, over 

 the observed range of values, CB estimates were generally higher than 

 1/2-meter at lower observed densities: at highest observed densities 

 estimates were about even with the two methods. These taxa included 

 copepod copepodites, cirripedia nauplii, and gastropod veligers. 

 Acartia spp. seem to have been equally estimated by the two methods at 

 lower observed densities but at the upper range of observed densities 

 1/2-meter sampling yielded higher densities than CB sampling. 



From 1973 through 1976, total zooplankton abundances in New 

 Haven Harbor were usually at lowest seasonal levels in December and 

 January (Figures 4-12 and 4-13) . Abundances generally increased to 

 seasonal peaks by March or April (1974 through 1976) ; in 1977, this peak 

 was delayed until May/June. This delay may have been related to the 

 exceptionally cold winter of 1976-1977, although no definite connection 



