Gorse biocontrol began in the 1930s with the release of the gorse seed weevil (Exapion ulicis), which was intended to reduce the spread of gorse but not affect its use as a hedging plant. Six more agents were released during the 1980–90s, including the gorse pod moth (Cydia succedana), the gorse soft shoot moth (Agonopterix umbellana), the gorse colonial hard shoot moth (Pempelia genistella), gorse thrips (Sericothrips staphylinus), the gorse spider mite (Tetranychus lintearius), and the gorse hard shoot moth (Scythris grandipennis). The gorse hard shoot moth failed to establish.
Various investigations have been conducted to assess the individual impacts of some of these biocontrol agents for gorse in New Zealand and Australia, but no formal, wide-scale assessment has been completed to assess agent occurrence and prevalence, impacts on seed production, and gorse condition and growth.
This study, completed between 2022 and 2024, measured current levels of seed predation and impacts from foliage feeding by gorse biocontrol agents throughout New Zealand. “We wanted to determine if impacts of biocontrol agents released since the 1930s have been sufficient to reduce gorse cover on a landscape scale”, said Paul Peterson, the senior technician who led this work.
The investigation was split into two components: one was an intensive, focused study to measure the impacts of the two seed predators (the gorse seed weevil and gorse pod moth) on seed banks in the soil at two sites; the other was a non-intensive survey to get a snapshot of impacts from the four established foliage-feeding biocontrol agents (the gorse spider mite, gorse thrips, gorse soft shoot moth, and gorse colonial hard shoot moth), covering all regions of New Zealand.
“We chose two sites for the first study – one in the North Island and one in the South Island – to measure annual addition to the seed bank after accounting for seed predation,” said Paul. “We tagged individual seed pods at various times throughout the year on 30 plants and very carefully followed the fate of each.” Pods were opened and seed predation by the agents assessed.
Image: gorse pod moth larva.
Image: gorse seed weevil larvae.
Data from each site were compared with a model produced by Rees and Hill in 2001, which predicts gorse cover based on seed fall and the probability of disturbances, such as fire and overgrazing. At the site in Christchurch, with seed predation at 60%, the model predicted that gorse cover would decline over time if the site remains undisturbed and seedling recruitment is low. However, at the Palmerston North site, where seed production was higher and seed predation only 24%, gorse cover was likely to increase over time regardless of disturbance regimes.
Paul and his team then focused on the four established foliar-feeding biocontrol agents. Seventeen sites, including one on the Chatham Islands, were visited and measurements taken to assess the occurrence and prevalence of biocontrol agents, the general condition of gorse, gorse growth rates, and soil seed banks. The study also included a questionnaire for landowners to find out about changes to land management that could affect gorse – aside from biocontrol impact. They were able to replicate growth rate measurements that were carried out intensively in the Dunedin area in the mid-1980s and compare the current growth rates to the ‘pre-biocontrol’ snapshot for that region.
Weeks of scrambling around through gorse taking measurements and speaking to landowners made one thing clear: despite the biocontrol agents being present at most of the sites, gorse is as healthy and as dense as it has ever been. In fact, dead gorse made up just 8.8% of gorse stands on average.
There were some exceptions. For example, at the Palmerston North site the gorse spider mite appeared to be severely damaging and/or killing up to 18% of gorse, and a potentially damaging outbreak of soft shoot moth was also seen.
Two native insects that bore into the stems of gorse were also commonly found on dead or dying plants. Lemon tree borer (Oemona hirta) damage was found at nearly all North Island sites, and gorse stem miner (Anisoplaca ptyoptera) damage at nearly all South Island sites. In fact, the amount of native borer damage found in live plants was the only predictor of the percentage of dead gorse.
Other species found inside dying gorse included the moth species Erchthias capnitis and Barea confusella; however, these species are recorded as only feeding on dead plant material, and they are unlikely to be contributing to plant mortality.
Seed banks varied by region but were particularly high in Otago, where a record 32,000 seeds/m2 were collected. On the flip side, the lowest seed bank density was measured from the Chatham Islands at 204 seeds/m2. It is possible that a lack of pollination by honeybees at this site was the cause. Honeybees have previously been identified as a critical driver of pollination and seed production for broom, which has a ‘spring-loaded’ flower, very similar to gorse. This pollination mechanism can only be tripped by large exotic pollinators.
“Insights from landowner questionnaires agree with our field measurements”, said Paul. Landowners were generally not aware of any impacts from biocontrol agents (except a mention of seed feeders and gorse spider mites), and gorse was mostly described as getting denser over time.
“The repeat sampling at gorse sites in and around Dunedin added even more interesting data to our study”, explained Senior Researcher, Simon Fowler. Not only did gorse appear to be unimpeded by biocontrol, but there was a significant increase in the stem diameter of plants compared with data from the 1980s, suggesting, if anything, that gorse growth rates have increased over time! “Whether this is due to increased temperatures or carbon dioxide concentrations, or some other factor, is yet to be determined, but we do know that the impact biocontrol agents are currently inflicting on gorse is insufficient to overcome those factors,” said Simon.
Our results confirm that the gorse biocontrol agents are struggling to have a meaningful impact on gorse in most circumstances. New biocontrol agents for gorse can be pursued if there is interest and funding. It won’t be a quick fix, though, and land management practices such as minimising disturbance and controlling seedlings following disturbance will remain significant tools in the medium term, alongside minimising gorse pollination.
Further reading:
Lee WG, Allen RB, Johnson PN 1986. Succession and dynamics of gorse (Ulex europaeus L.) communities in the Dunedin ecological district South Island, New Zealand. New Zealand Journal of Botany 24(2): 279–292.
Rees M, Hill RL 2001. Large-scale disturbances, biological control and the dynamics of gorse populations. Journal of Applied Ecology 38(2): 364–377.
This work was funded by the National Biocontrol Collective and the Ministry for Primary Industries’ Sustainable Food and Fibre Futures Fund [Grant #20095] on multi-weed biocontrol.
