Central Amazon fish populations : biomass, production and some dynamic characteristics / Peter B. Bayley.

Tese (Ph.D.) - Dalhousie University, Halifax, Nova Scotia, 1982

A 2.5 year field study designed to estimate production of fish in the Amazon river provides a basis for understanding the fishery and certain features of floodplain ecology. Intensive sampling with a fine-mesh seine net throughout the hydrological cycle in a variety of floodplain habitats revealed the seasonal spawning sequences of many species, allowing identification of cohorts sampled routinely, the efficiency of the seine net was estimated for various species groups and sizes. Thus seine catches could be converted to biomass density estimates. The overall estimate of fish and decapods was 16Ø gm-2 with 95% confidence limits of ±24. Growth of individual species appeared unaffected by the biomass of potentially competing species when the water was rising, and only two out of eleven species suggested density-dependent growth (at p 05) during falling water. A strong seasonal variation in growth rate was observed. A multispecies growth function which depends on weight, maximum weight of species, and hydrological season was derived from the growth rates of 14 common species. Fish plus decapod production averaged 28Ø gm-2yr-1, based on the biomass estimates and growth function. The estimated phytoplankton primary productivity of 29Ø gCm-2yr-1 is insufficient to sustain the fish productivity, and the importance of macrophytes was also evident from the gut contents of common species. The biomass distribution when plotted in logarithmic length intervals increased sharply with length. Non-conformity with the Sheldon spectrum for pelagic systems could be explained by many fish feeding on extremely small particles (fine detritus feeders) or very large particles (macrophyte herbivores). However, the prey to piscivore biomass ratio varied from 1.2 to 1.3, wich was similar to Kerr's (1974) interpretation of the Sheldon spectrum. Transfer efficiencies of 37% and 39% were estimatd for two successive trophic levels, indicating a high predation efficiency and grouth efficiency. The similarity in biomass ratios and predation efficiencies between the central Amazon and those estimated for pelagic systems is not paralleled by similar transfer efficiencies which are apparently much lower in pelagic systems. This implies higher gross growth efficiencies in Amazon fish which may be due to the higher prey/predator length ratio of 25. The system thus appears to be very efficient, yet only 2.7% of the total productivity is taken as yield by man in a more exploited part of the basin. It is concluded that the high predation efficiency leaves limited surplus production from lower trophic levels to support an expanded fishery, unless the latter can be managed in such a way as to exploit all piscivores, many of which have low market value at present.