Size and configuration of freshwater ecosystems

Drivers of community structure in steams: could ecosystem-size be the answer?

One of the main challenges confronting ecologists is understanding and predicting the composition of biological communities across the landscape. Abiotic conditions have long been recognised as one of the major determinants of community composition, but only recently have patterns in community structure been thought of as the product of multiple interacting drivers acting at several ecological scales. As a result, Phil Jellyman's PhD research has been focused on two potential key drivers of food-web structure in stream systems; ecosystem-size and disturbance. Whilst the influence of flow-related disturbance on stream community properties such as food and habitat availability is well established, the importance of ecosystem size and its interaction with disturbance is largely unknown.

Phil has been investigating the effects of these processes on New Zealand fish communities by firstly analysing the abundance and biomass of fish faunas across gradients of stream size and disturbance for more than 50 sites over a 10-year period (and 25 sites for stream invertebrates). Analyses showed that the amount of fish biomass at a particular site was significantly affected by invertebrate biomass. In addition to this, both fish and invertebrate biomass were significantly influenced by the level of disturbance at a site. However, these relationships were confounded by stream size and further analysis revealed a relationship combining all these factors to explain how food web structure changed across gradients of ecosystem size.

More recently Phil has been investigating relationships between prey availability and fish assemblage structure. Both prey community assemblage and biomass were strongly related to fish community structure and fish biomass within stream reaches. Salmonid abundance, in particular, was strongly affected by prey biomass, and a mesocosm experiment indicated that the type and quantity of prey affected trout growth. These results (and drift survey work) indicated that the biomass and composition of fish communities was being affected via disturbance-mediated changes to prey communities. Current research is now focussing on understanding the interaction between disturbance regime and prey community dynamics and how these alter stream fish communities (Fig. 1) for improved fisheries management.

Figure 1. Hypothetical model of fish community structure in relation to disturbance and prey gradients.

People involved in this project

University of Canterbury
Phillip Jellyman
Angus McIntosh
Jason Tylianakis

Stroud Water Research Center
Dave Arscott