The Food of Two Predatory Demersal Fish Species and the Bottom Fauna in the Western Bay of Plenty, New Zealand

Abstract

The food and feeding relationships of 1,420 tarakihi Cheilodactylus macropterus (Bloch and Schneider) and 772 snapper Chrysophrys auratus (Bloch and Schneider) taken at seven sampling areas in the western Bay of plenty during 1969-70 are examined. Also examined are a further 154 tarakihi from Tasman Bay. A computer system for processing the food data from each fish species is described. Feeding of both tarakihi and snapper are analysed in relation to fish length, sex and to differences in depth, sampling area, season and time of day. For both fish species differences in feeding occurred with differences in fish length depth, sampling area and time of day. Only minor differences in feeding are apparent between the sexes of both fish species. Little seasonal change in diet is noted in either fish species. At six of the seven sampling areas in the western Bay of plenty, benthos samples were collected by an Agassiz trawl and anchor dredge just before or after each fish sample was taken. A sediment sub-sample was taken from each anchor dredge sample and analysed. Each of the six sampling areas had a variable sediment type. The benthos data from the Agassiz trawl, anchor dredge, tarakihi and snapper are analysed by numerical computer methods. The results of the two different dissimilarity coefficients (Canberra metric and Bray-Curtis) are compared using both Agassiz trawl numbers (abundance) and points (bulk) data with flexible sorting in normal (site) classifications. Decisions as to the "best" Agassiz trawl site classification differ with the number of site-groups (two or six) and the type of criteria (sorting within the site-groups by depth, sampling area or sediment type) used in the assessment. Site classifications using the two data forms (numbers and points) and the Bray-Curtis dissimilarity coefficient for each of the four benthos samplers are compared at the two or three and six site-group levels using sorting of site-groups by depth, sampling area and sediment type as criteria for assessment. By depth the sites are divided into two or three major groups: either shallow-intermediate and intermediate-deep or shallow, intermediate-deep and deep site-groups. At the six site-group level using separation of site-groups by sampling area and sediment type as criteria, points data separated a larger number of site-groups than did the numbers data. The anchor dredge data are believed under-sample since only three-fifths the number of site-groups were sorted by sampling area and sediment type when compared to the other three benthos samplers. Over twice as many site-groups were sorted by sampling are as compared to sediment type. Inverse (species) analyses divided species into three major groups, but only the snapper numbers species classification gave major groups which corresponded to species having shallow, widely occurring and deep distributions. The other species classifications were not very understandable. The two deep sampling areas (Alderman Is. and Plate Is.) and some Motiti Is. sites (intermediate depth) were impoverished in both numbers of species and individuals and by bulk compared to the shallow sampling areas, and this corresponded to tow faunistic zones: the outer shelf zone (27-70 m) and the upper continental slope zone (70-183 m). Most species taken by the anchor dredge were able to withstand changes of 10-70% in the weight of the mud fraction in sediments. In addition, most site and species-groups showed little relationship to sediment type. None of the seventeen previously described New Zealand infaunal, benthic, classical 'communities' of the continental shelf could be recognized in this classification, and the implications of this result are discussed. The selectivity of the two fish species for various benthos species was determined at six sampling areas in the western Bay of Plenty during two time periods (Oct. –Dec. 1969 and April-June 1970). The feeding relationships between each fish species and the benthos were constantly changing, as the food selectivity values, Food Similarity Index and the percentages of feeding overlap varied with both sampling are and sampling period. Food competition existed between the two fish species, but was probably only incipiently deleterious, since both fish species exhibited great diversity in their diets.