I have seen greenhoods countless times since, but not once, in years of studying weed ecology in New Zealand, have I encountered them among gorse. Is this just chance? And what other differences are there between gorse and native successions? As a clay-streaked kid in Lower Hutt, of course, I never considered such matters, but now that the impacts of invasive species are coming under increasing scrutiny, and revegetation projects are popping up everywhere, these are questions worth asking.
Before European settlement, the dominant native shrubs in formerly forested sites stricken by fire in lowland New Zealand were manuka and kanuka. With settlement, these were replaced over large areas by naturalised woody legumes, particularly gorse (Ulex europaeus), before the large-scale conversion of lowland country to forestry plantations, mainly pine. Early observers noted the role of manuka and kanuka as precursors to forest, and we have long known that if gorse is kept free of fire, and a seed source is available, native broad-leaved species such as coprosmas (Coprosma spp.), mahoe (Melicytus ramiflorus) and five-finger (Pseudopanax arboreus) dominate after about 40 years.Kanuka/manuka and gorse are thus alternative post-disturbance systems across large parts of the New Zealand landscape. Which species dominates in a particular area is determined largely by the history of fire and grazing there.
By apparently facilitating the restoration of hillsides from failed agriculture to native vegetation, gorse has been viewed as a beneficial “nurse crop” and often managed through “benign neglect”. The unstated assumption is that gorse on the one hand and manuka and kanuka on the other are of comparable value as regards conservation of native biodiversity and will lead to the same kind of forest.
But this assumption is wrong. Recent scientific evidence shows gorse and kanuka scrub differ in many respects, including microfauna in the soil and insect and bird communities above ground. Studies by the author and colleagues on seed fall and native-seedling survival in gorse and kanuka stands near Nelson suggest that successions through gorse and successions through kanuka may actually lead to different forest types. Strong evidence for this comes also from Hinewai, on Banks Peninsula, where Hugh Wilson found kanuka scrub gives way to the original cover of beech forest whereas, on similar sites, gorse gives rise to native broad-leaved forest.
Recently I wandered again through the eastern Hutt Valley hills above Epuni, and, sure enough, large areas of my prickly childhood play area are now covered in mahoe, with not a beech tree in site, despite beech having been the dominant cover in pre-European times.
Studies further south, in Dunedin, have shown the diversity of native woody species, including lianas (climbers and twiners), to be less under gorse scrub up to 30 years of age than under similar-aged kanuka.
No podocarp seedlings were found under gorse, whereas there were a few under kanuka. This suggested gorse might be less favourable for podocarp establishment, at least for the first 40 years of woody plant cover. Small-leaved coprosmas, mingimingi (Cyathodes juniperina, also small-leaved) and native ground species were also more common under kanuka than under gorse. Green-hood orchids were absent from gorse but common beneath kanuka, both in Dunedin and at Taita, Lower Hutt, where an earlier seminal study of gorse successions was conducted.
These apparent differences between kanuka and gorse communities, and the differing abilities of the two kinds of community to support native birds and invertebrates, would be of only minor importance were it not for the fact that little remains of weed-free early successional vegetation in lowland New Zealand.
Jon Sullivan, of Lincoln University, Susan Timmins, of DOC, and I chose the Wellington and Nelson regions to test if kanuka and gorse scrub did indeed have similar plant-biodiversity values, and whether they led towards the same kinds of forest. We sampled vegetation at four successional stages, namely “young” gorse with native shrubs or weeds protruding, “old” gorse with native shrubs predominant, short “young” kanuka and taller “old” kanuka.
Differences in the composition of plant communities in stands in the two regions were related primarily to whether the stands were of kanuka or gorse. Overall there were more native woody species on kanuka sites than on gorse sites, especially among seedlings.
The main native species more common in kanuka than in gorse were heketara (Olearia rani), jasmine (Parsonsia heteropylla), mingimingi, kamahi (Weinmannia racemosa), rangiora (Brachyglottis repanda), supplejack (Rhopalostylis sapida), hook grasses (Uncinia spp.) and various ferns, in particular bracken (Pteridium esculentum), common shield fern (Polystichum richardii), hanging spleenwort (Asplenium flaccidum), kio kio (Blechnum capense) and shining spleenwort (Asplenium oblongifolium). We found only one native orchid species—a greenhood (Pterostylis graminea)—and only in the Nelson region, where it occurred more often among kanuka than among gorse.
Naturalised woody species, too—some of them weeds controlled by regional councils and DOC, such as blackberry (Rubus spp.) and wild cherry (Prunus spp.)—were more prevalent on gorse sites than on kanuka sites. This was particularly so in Wellington, where weeds were more abundant than in Nelson, all of them horticultural escapees.
Unfortunately, the lack of mature podocarps near the test stands in either region meant we were unable to resolve the question of whether kanuka or gorse is more favourable for their regeneration.
Compositional differences between old-gorse plant communities and old-kanuka plant communities were just as great as those between young-gorse and young-kanuka communties. This suggested kanuka and gorse were leading to different vegetations. We do not know how long it will take for these vegetations to become indistinguishable.
Indications are that in the presence of adequate seed sources, gorse scrub is more likely than kanuka scrub to lead to vegetation dominated by weeds such as barberry (Berberis glaucocarpa), a preferred food source of exotic birds and silvereyes (recent colonists from Australia) rather than endemic species such as bellbirds. Where gorse gives way to a weed species in this manner—which it doesn’t do universally—the adverse effect on native biodiversity may be compounded.
In summary, the plants associated with kanuka scrub and gorse scrub are different, as are their immediate successors. Fewer species—in particular, fewer orchids and small-leaved shrubs—tend to grow among gorse than among kanuka. While both gorse and kanuka lead to native forest, gorse-originated forest is different from kanuka-originated forest and is unlikely to include beech. Gorse is not therefore a substitute for native successional species, and kanuka is worth preserving and encouraging even where gorse seems to be giving rise to the regeneration of native plants.