Protocol filters Cape Town trees for invasion and pest risks

A multi-criterion protocol applied to 444 tree taxa planted in Cape Town identified 49 taxa that met the study’s central conditions for urban planting: they were not regulated as invasive in South Africa, were not known to have naturalised in South Africa or elsewhere, and were not known hosts of the polyphagous shot hole borer, an invasive ambrosia beetle. The authors describe the result as a decision-support tool rather than a permanent guarantee of safety. It combines national regulations, global records of naturalisation, and current evidence on tree susceptibility to the beetle and its symbiotic fungus, Neocosmospora euwallaceae.

Among the 49 taxa were 37 native species and 12 non-native species not known to have naturalised, while a survey of South African nurseries and online plant retailers found 32 of the recommended taxa commercially available. The case study was developed for the City of Cape Town Metropolitan Municipality, where the need for safer replacement planting is immediate. Polyphagous shot hole borer, or PSHB, was first detected in the city in 2019.

By March 2025, more than 10,000 trees had been infested and approximately 4,000 affected trees had been removed from public land, according to the City of Cape Town Invasive Species Unit. The municipality’s tree canopy covers only about 6% of its area, and further losses could accompany urban expansion, infrastructure development, climate change and pest damage. Although the inventory and protocol describe planning conditions, they do not test whether the selected trees will prevent infestations or invasions; they provide a transparent way to classify candidate species and identify where planting, removal and monitoring are warranted.

The protocol begins with a comprehensive inventory of every tree taxon known to occur in a study region, including trees in natural, semi-natural, urban and cultivated settings. A second list then identifies which taxa are already planted locally. The Cape Town analysis drew on a South African inventory containing more than 35,000 records of 805 planted tree taxa.

By considering only taxa already established in the country, the researchers avoided actively recommending new introductions, which would require separate screening for unforeseen invasion potential. Each candidate is then evaluated against criteria chosen for the local ecological, regulatory and management context. The first principal criterion in the Cape Town case was regulatory status: taxa listed under South Africa’s National Environmental Management: Biodiversity Act alien and invasive species regulations cannot be planted unless the relevant environmental authority grants a permit or exemption.

The authors also advise consulting local practitioners because official lists may not identify emerging or localised invasions. The second criterion concerned PSHB host status. A Neocosmospora-colonised host permits transmission of the fungus but does not support reproduction of the beetle.

A competent host supports the establishment of a natal gallery and production of offspring. A kill-competent host has at least one documented individual that died from the combined effects of the beetle and fungus infestation. The researchers integrated several independent datasets rather than relying on a single source.

South Africa’s 2020 regulatory lists established legal status, while a national planted-tree inventory classified taxa as native or non-native. The Global Naturalised Alien Flora database supplied evidence of naturalisation within and outside South Africa. A global PSHB host list, compiled using molecular records confirming the beetle’s identity, and host-status information were used to assess susceptibility.

To avoid underestimating potential risk, taxa with unknown statuses were conservatively assigned the highest known risk level among their congeners. Cross-referencing these attributes produces eight recommendations grouped into scenarios involving tree planting, tree removal and tree monitoring. Taxa regulated as invasive fall into the most restrictive category and must not be planted without authorisation.

Naturalised kill-competent hosts should not be selected for planting; where a site is close to a PSHB-infested or ecologically sensitive area, their removal is recommended where feasible. Native kill-competent hosts and naturalised competent hosts are also unsuitable for planting in relevant situations, while less severe combinations may be considered only with restrictions on scale, site choice or continued monitoring. Outside infestation and sensitive zones, the protocol prioritises taxa that are not known hosts and have no evidence of naturalisation.

These recommendations encode precautionary assumptions rather than demonstrated outcomes, and uncertainty requires consultation of global and region-specific evidence. Among the 444 planted taxa assessed for Cape Town, 85, or 19%, were regulated as invasive and could not be planted without permission. Nine were kill-competent hosts, five of them non-native and naturalised.

Another 57 were competent hosts and were predominantly non-native and naturalised. Beyond confirmed hosts, 81 taxa belonged to genera containing competent hosts, while 68 were Neocosmospora-colonised hosts. Of the 115 taxa not known to be PSHB hosts or closely related to known hosts, 66 were non-native and naturalised and were therefore not recommended for planting under the central scenario.

The final set of 49 included native taxa such as Dais cotinifolia, the pom pom tree, and Aloidendron dichotomum, the quiver tree, as well as non-native taxa not known to have naturalised. Commercial availability may determine how readily the protocol can influence planting programmes. Searches of a representative selection of printed nursery catalogues, public online plant retailers and targeted South African sales queries showed that 32 of the 49 recommended taxa were offered as seeds, seedlings, saplings or mature trees; 29 of those available taxa were native.

The findings therefore identify a gap between ecological screening and horticultural supply. They do not show that commercially available species are inherently less secure than unavailable ones. Rather, the authors propose using the screened list to encourage nurseries to propagate suitable alternatives and help municipalities and public buyers identify lower-risk options.

Lists can also be tailored to particular functions, such as street planting, riparian buffers or allergen-free school environments, and should be reviewed as evidence changes. Several limitations prevent the output from being treated as universally applicable. The underlying national inventory does not include every tree planted in South Africa or taxa that have never been trialled, so potentially suitable species may be absent because they are poorly represented in the available data.

Global and citizen-science databases can be incomplete, inconsistently standardised or locally unverified, while some regions lack the institutional capacity for taxonomic verification and pest surveillance. Native status alone does not eliminate all risks: habitat disturbance, propagule pressure and local environmental conditions can permit native trees to expand beyond their normal range. Factors such as prolific seed production, long-distance dispersal, clonal reproduction and vigorous resprouting may also require assessment.

Municipal or provincial application can incorporate climate, soils, nearby populations and provincial rules that differ under national regulations. The authors also stress that PSHB host knowledge is a temporal snapshot. Because the beetle is relatively recent in South Africa and Australia, some tree taxa may have had insufficient exposure for colonisation to be recorded, creating what the authors describe as host debt.

The known host range can expand as the invasion unfolds, requiring periodic review, long-term surveillance and adaptive management. Visual inspections can look for staining around entry holes, sawdust-like frass, gumming and discoloration, while baited traps, community reporting and regional networks may improve early detection. The protocol is intended to be transferred by replacing Cape Town datasets with local inventories, regulations and pest-host records, adding criteria such as drought tolerance or fire susceptibility where relevant.

Its value lies in making the evidence, exclusions and uncertainties behind tree-selection decisions visible rather than in claiming that a static list can guarantee ecological security.

Subject of Research: Evidence-based screening of tree taxa for invasive-species and shot hole borer risks

Article Title: An evidence-based protocol for developing lists for tree planting

Article References: Potgieter, L. J., Cadotte, M. W., Kumschick, S., Paap, T., Roets, F., Wilson, J. R. U., & Richardson, D. M. (2025). An evidence-based protocol for developing lists for tree planting. NeoBiota, 104, 113-137. https://doi.org/10.3897/neobiota.104.156206

Image Credits: AI Generated

DOI: 10.3897/neobiota.104.156206

Keywords: tree planting, biological invasions, polyphagous shot hole borer, species selection, Cape Town, urban forestry, pest risk, tree nurseries, ecological resilience, evidence-based, protocol, developing