PDF of this article (277 KB)
Paula Reeves, Mark Meleason, and Fleur Matheson have studied fifteen stream-restoration projects to learn the keys to success.
Streamside – or riparian – vegetation can influence the health of waterways and how they function. Over the last 10 to 15 years, there has been a surge in stream-restoration projects by a range of agencies, community groups, and rural landowners. These projects are strongly focused on re-establishing or enhancing riparian vegetation. In rural areas, the main objective of planting riparian buffer strips tends to be improving water quality. In urban areas, riparian planting projects also seek many social and cultural benefits including aesthetics, recreation, and flood control.
Riparian planting and enhancement can be costly; Christchurch City Council estimates costs from $30 to $1000 per metre for its projects, depending on site conditions and objectives. To ensure funds are spent effectively, it’s worthwhile to aim for plantings that are self-sustaining, that is, self-seeding with minimal weed growth.
Here we examine some of the issues for establishing self-sustaining plantings of native species in urban and rural landscapes. Our findings are based on interviews with agencies and landowners involved in riparian planting, and on field assessments of 50 riparian buffers planted in urban catchments in Waitakere City and Christchurch City, and in rural catchments near Raglan (Waikato) and Lake Ellesmere (Canterbury). Our field assessments included measuring species composition, planting density, canopy closure, weedy ground cover, and seedling densities.
Key factors: weeds and seeds
Two main factors contribute to the ability of a riparian planting to sustain itself over the long term (more than 50 years): the plantings must out-compete the weeds and new seedlings must establish under the plantings through natural regeneration. To achieve both factors, plants must form a closed canopy that shades the ground. This reduces and prevents weed dominance and creates conditions that favour the establishment of new seedlings.
Reducing weedy ground cover
At the sites we assessed, canopy closure usually occurred within four years of planting. However, the elimination of weedy ground cover often took much longer and was most strongly related to planting age and buffer width. At almost all sites, herbicides were used to control weeds for at least the first two years. When active weed control had ceased, weedy ground cover was usually least within the interior and greatest near the edges of the planting, where canopy cover was low and light levels were high. Weedy ground cover at the edges intruded from about 1 to 5 m into the buffer. Planting densely leaved shrubs such as manuka (Leptospermum scoparium), lemonwood (Pittosporum eugenioides), and kohuhu (Pittosporum tenuifolium) at the edges helped reduce light penetration into the buffer.
In some sites, such as the urban parks in Christchurch, planting projects that emphasised social benefits ('nature in the city') tended to be less effective at achieving a self-sustained planting due to the narrow buffer width and low planting densities. In such cases, active management (such as weed control) may always be necessary to maintain the plantings.
Planting density was a crucial factor at some sites, especially in the early establishment stages. In Waitakere City, plants spaced about 1.5 to 3 m apart were commonly dominated by weeds, and the shrub plantings frequently failed to dominate the site. Densities were subsequently reduced to 90 cm for shrub species, increasing the survival and growth of the new plantings. At the sites most affected by weed invasion, canopy closure needed to be achieved within two years to ensure dominance of the native species. This was accomplished by planting clump-forming species, such as the native toetoe (Cortaderia fulvida) and mapere (Gahnia setifolia), spaced 50 to 70 cm apart.
Seedling establishment
The number of self-sown seedlings was highly variable at the sites we investigated, with no significant difference between urban and rural sites. The density of native seedlings across all sites averaged 0.88 seedlings per m2. This average is a little higher than seedling densities recorded in intact native forest in Waikato hill country (0.64 seedlings per m2) and much higher than in nearby forest remnants (0.004).
Of concern for the sustainability of these sites was the equal split between native and weedy seedlings. Even at the oldest site (30 years), where seedling densities were highest, there was no significant difference between weedy and native seedling densities.
We found that, in general, the wider the buffer the greater the density of seedlings, although other factors may be more important at different stages in the development of the riparian planting. For example, seedling density was particularly low in 15 to 20-m buffers, where planting densities were consistently high (80 to 125 cm apart) and relatively young (5 to 8 years old). These plantings typically had a very dense canopy cover and had only just begun the process of self-thinning, which leads to a more favourable light climate for seedlings to germinate.
Sustainable widths
In general, dense riparian plantings with widths of 15 m or more provide the best environment for reducing weedy ground cover. If this amount of space is not available, then planting dense shrubs on the edges to reduce light entering the buffer can help. Self-seeding was generally higher in riparian plantings wider than 10 m; however, it is likely to take a number of years before densely planted riparian buffers begin to thin out, providing a suitable light climate for seedling germination.
Riparian recipe
- Riparian buffers of native plants are known to improve rural and urban streams, affecting water quality, flood control, and aesthetics.
- A survey of 50 buffers in 15 New Zealand planting projects has found two key factors that determine whether the buffer will become self-sustaining:
- Weedy ground cover must be reduced to the point that the natives outgrow the weeds.
- Self-sown native seedlings must outnumber weed seedlings.
- The optimum width for riparian buffers is 15 m or more.
Paula Reeves and Dr Fleur Matheson are freshwater ecologists based at NIWA in Hamilton; Paula is on maternity leave during 2006. Dr Mark Meleason, formerly at NIWA in Hamilton, is now a private consultant in Portland, Oregon.
Thanks to the following people who volunteered their time to talk about their riparian planting experiences: Fred Lichtwark (Whaingaroa Harbour-Care), Peter Chamberlain (Lake Ellesmere farmer), Chris Ferkins (Waitakere City Council), Kerry Bodmin (ex-Waitakere City Council), Rachel Barker (Christchurch City Council), and David Hewson (Environment Canterbury).
Teachers’ resource for NCEA AS: Biology 90165 (1.5), 90167 (1.7), 90460 (2.4), 90461 (2.5). See other curriculum connections at www.niwa.co.nz/pubs/wa/resources