Ecology of urban streams

Factors influencing benthic communities and colonisation in Christchurch’s urban stream

Worldwide, increasing numbers of stream restoration projects are being initiated to rehabilitate waterways modified by urbanisation. However, many of these projects have limited success in restoring stream communities. Okeover Stream, on the University of Canterbury campus in Christchurch, New Zealand, has been the subject of restoration efforts since 1998. As part of her BSc Hons research, Tanya Blakely focused on quantifying the response of this urban stream to current restoration efforts. Initially, physico-chemical conditions and biological communities at three sites along the Okeover Stream were compared with three physically similar sites on each of nearby Waimairi Stream and Avon River. General physical and chemical parameters were similar in all streams with circum-neutral pH, specific conductivity ranging from 167 to 173 µS / cm, dissolved oxygen ranging from 9.0 to 9.2 mg/litre, low turbidity, and similar hydrological conditions. However, analysis of heavy metals in the sediment showed mean lead (Pb) concentrations in Okeover and Waimairi Streams exceeded ANZECC ISQG-low trigger values (86.9 and 83.7 mg / kg, respectively), whereas Avon River sediment Pb levels (27.3 mg / kg) were below trigger values.

Benthic taxonomic richness did not differ significantly among the three streams, but the benthic community in Okeover Stream was dominated by the amphipod Paracalliope fluviatilis, whereas the gastropod snails Potamopyrgus antipodarum and Physella acuta were the dominant benthic fauna in Waimairi Stream and the Avon River. A further assessment made at six sites along 1200 m of Okeover Stream showed no distinct longitudinal patterns in physical or chemical conditions, but there was a strong longitudinal change in benthic macroinvertebrate communities. Taxonomic richness and caddisfly diversity increased downstream, with twice as many taxa at the most downstream site than the uppermost sampling site (Fig. 1; Blakely & Harding 2005, pdf, 700kb).

Figure 1. Total number of benthic macroinvertebrate (black bars) and caddisfly (white bars) taxa collected for six sites along 1200 m of Okeover Stream. Figure is modified from Blakely & Harding 2005.

Prompted by this longitudinal trend in decreasing caddisfly richness upstream (Fig. 1), Tanya investigated previously unrecognised barriers to aquatic insect colonisation in urban streams (Blakely et al. 2006, pdf, 780kb). To investigate whether the availability of suitable substrata for oviposition limited the longitudinal distribution of caddisflies, large boulders were added to the upstream reaches of Okeover Stream. Prior to the addition, more egg masses were observed downstream and this longitudinal pattern persisted subsequently (Fig. 2).

Figure 2. Boulders were added to four reaches along Okeover Stream and after two weeks we counted the number of Hydrobiosis parumbipennis egg masses on these. Significantly more egg masses were found at sites after boulder had been added, but there remained a distinct longitudinal pattern with fewer egg masses found upstream. Figure is modified from Blakely et al. 2006.

Malaise trapping also revealed that adult caddisfly diversity and abundance was greater downstream than upstream. The only obvious potential obstructions between study reaches were roads beneath which the stream flowed through culverted crossings (Fig. 3).

Figure 3. Urban streams are often piped under road crossings through concrete culverts, which may act as barriers to adult aquatic insect dispersal

Tanya used Malaise trapping to examine the effect of road culverts and bridges on caddisfly dispersal. Numbers of caddisflies caught declined upstream of culverts and about 2.5 more individuals were captured in traps immediately below than above five culverts (Blakely et al. 2006, pdf, 780kb). Furthermore, bridges (which had a more open structure than culverts) had no significant effect on the size of catches made above and below them.

In an attempt to determine the processes by which adult caddis are restricted by culverts Jon Harding investigated the influence of culvert mouth and dimensions, and the influence of spider predation. Spiders living within the culverts (Fig. 4) markedly reduce the number of adults passing though, while the size of the downstream culvert mouth also plays an important role in determining the number of adult caddis that enter the culvert (Harding et al. 2005, pdf, 200kb).

Figure 4. Spiders living within the road culverts are likely to prey on adult aquatic insects, thereby reducing the number of adults passing through culverts and negatively impacting upstream recruitment.

People involved in this project

University of Canterbury
Tanya Blakely
Jon Harding
Angus McIntosh
Mike Winterbourn

Publications

Blakely & Harding 2005. Longitudinal patterns in benthic communities in an urban stream under restoration. New Zealand Marine and Freshwater Research 39: 17-28 (pdf, 700kb).
Blakely, Harding, McIntosh & Winterbourn 2006. Barriers to the recovery of aquatic insect communities in urban streams. Freshwater Biology 51: 1634-1645 (pdf, 780kb).

Posters and newsletters

Blakely & Harding 2004. Restoring Okeover Stream - what factors affect in-stream recovery? Poster by the Freshwater Ecology Research Group, University of Canterbury (pdf, 1mb).
Blakely, Harding & McIntosh 2004. Road culverts - unrecognised barriers to upstream caddisfly dispersal. Poster by the Freshwater Ecology Research Group, University of Canterbury (pdf, 900kb).
Harding, Neumegen & Smith 2005. Spiders, culverts and urban streams. Newsletter of the Freshwater Ecology Research Group, University of Canterbury (pdf, 200kb).