We are beginning to get more information about invasive species on the African continent.
a flyer naming principal invasive ornamental plants in Kruger National Park
In several countries, the focus has been on threats to agriculture. Previous blog about horizon-scanning in Ghana. In Zimbabwe, N. Mudada and colleagues (2026; full citation and the end of this blog) found alarming, if not surprising, levels of risk to food production from introduced invasive plants. They investigated 1,668 human-aided transboundary plant introductions at 14 ports of entry and non-official crossing points over the course of four years.
They estimate that the 20,000 trucks that transported maize into the country over the four years carried over 20,700 metric tons of weed seeds and rubbish! They recorded detections of 11 species in eight orders. The pathways are familiar. As noted, several weeds were contaminants of grain shipments; Convolvulus arvensis in wheat for human consumption, Helianthus annus and Datura stramonium in maize for animal feed. Adenium obesum and Vitex agnus-castus were being smuggled for planting as flowers and ornamentals. (Vitex agnus-castus was also smuggled in passenger baggage for its medicinal properties). Several Lemna species (an aquatic plant) were also smuggled for planting as animal feeds.
In some cases, the focus is the threat to native ecosystems. I posted a blog the about threat of an introduced pathogen to trees in the remnant rain forests of Madagascar.
South Africa still has the lead in addressing invasive species. Regarding invasive plants specifically, the country has the benefit of more than 150 years of botanizing. The richness of the region’s flora is globally recognized. South Africa also has a long history of studying and managing invasive species, especially plants.
South African scientists and colleagues in Botswana, Eswatini, Lesotho, and Namibia have published four editions of the Flora of the Southern Africa region since 1984. In 2006, the PRECIS database of the South African National Biological Diversity Institute (SANBI) was combined with the Tropical African Plant Checklist published by the Conservatory and Botanical Garden of Geneva to create the African Plant Checklist and Database Project. It is continually updated. This is the first continental flora checklist for Africa; it fulfils countries’ obligations under the Convention on Biological Diversity’s Global Strategy for Plant Conservation.
The 2025 Checklist reports that 21,539 plant species are extant outside cultivation in the country; these comprise 20,204 indigenous species and 1,329 naturalized species. Thus, 6% of the total flora is non-indigenous. Of these, 649 (48.8% of the non-indigenous species, 3% of all plants) of them are invasive.
[Naturalized species are defined as species whose documented natural range does not include South Africa, but have overcome a biogeographic barrier and now sustain self-replacing populations for two or more life cycles or over a given period of time in the country. These populations are maintained without direct intervention by people, or despite human intervention. Invasive species meet the above definition plus produce reproductive offspring, often in large #s at considerable distances from the parent and/or site of introduction, and have the potential to spread over long distances.]
Since the previous checklist was published in 2006, botanists have identified 1,048 additional species – a 4.9% increase. Eighty-two percent of the newly identified species (865 species) are “naturalized”. Specifically, 414 new species are categorized as naturalized (a 31.1% increase), and 451 new species are classified as invasive (a whopping 69.5% increase). Le Roux and Klopper attribute these steep increases to active botanizing by SANBI’s Invasive Species Programme (begun in 2008), and the Southern African Plant Invaders Atlas Project (begun in 2010).
Of the 384 plant families present in South Africa, 350 contain at least some indigenous species. Thirty-four families contain only naturalized species. Among the 2,189 plant genera present, 459 (21%) contain only species that are non-indigenous.
Three families stand out because of the particularly high numbers of naturalized species: Fabaceae (143 species; 11% of all naturalized species), Asteraceae (140 species; 10%), Poaceae (123 species; 9%). Two of these families — Asteraceae and Fabaceae — are also the largest families among native South African plants. The third, grasses (Poaceae), ranks seventh in the list of most specious families indigenous to South Africa. The next group of families with high numbers of naturalized species has less than half as many invasive species: Myrtacae (55), Amaranthacea (52), Solanaceae (48). None of these families ranked within the top 20 families of indigenous plant species.
The genera with the most naturalized species were Solanum,Euphorbia and Acacia (all 24 or 23 species).
Acacia cyclops; photo by David M. Richardson
South African scientists are also exploring how to balance conflicting goals and perspectives when an invasive plant species has economic or social value. The example chosen by Mbobo et al. (2025) is guava (Psidium guajava) – a nutritious and popular tropical fruit grown commercially in South Africa, but also invasive along roadsides, watercourses and forest margins. Invasions are especially common in eastern parts of country; large monocultures are found in KwaZulu-Natal. Outbreaks have also been detected at five sites in Western Cape in riparian zones and at a hot spring. Mbobo et al. (2025) note that the microclimatic conditions at this last location differ from the broader conditions in the region – which are what most models would measure.
The scientists used models to predict where guava might invade – especially in large monocultures – and compared those areas to where the tree can be grown in cultivation with human inputs, e.g., irrigation. They then assessed whether six regulatory approaches would avoid restricting guava farming in areas at minimal or low risk while still protecting vulnerable locations. They also considered the amount of information required to implement the approach and costs of acquiring the information; and level of likely public acceptance. Mbobo et al. (2025) laid out the trade-offs between continuing to regulate planting of the species at the provincial level vs. at the municipal level. Prohibiting planting of guava in provinces where it is recorded as invasive allows some plantings near natural forests and riparian areas that are highly susceptible to guava invasions. On the other hand, nearly half of the prohibited area is outside the known or likely at-risk area. The provinces do allow exceptions through a permit process. Adopting more geographically limited rules by regulating at the municipal level would enable a tighter link to geographic areas at highest risk. However, this approach does not address long-distance seed dispersal by animals. Furthermore, the very detailed regulations might confuse stakeholders and complicate enforcement. Also, the models lack sufficiently fine spatial resolutions to predict invasible areas so accurately. Finally, the reduction in regulated area is minimal (~ 14%), so the economic benefits are unlikely to outweigh the significantly higher administrative costs and risk of allowing guava invasions in new sites.
Guava fruit on tree; Roenashy via Wikimedia
Gildenhuys et al. (2026) analyzed the factors that drive which non-native plants establish where. They assessed the roles of temperature, precipitation, urbanization intensity, urban area, travel time, year of city’s establishment, and human population density in determining which plant species are present in 54 urban centers in Western Cape Province. The cities have significant differences in climate: Mediterranean in the far southwest, warm temperate in the southeast, and semi-arid towards the interior. The expectation was that these drivers and assembly processes are influential at more advanced invasion stages when the species have already overcome some barriers to dispersal, so are now found in reasonably suitable habitats.
Gildenhuys et al. (2026) found temperature and precipitation were most important in determining plant species’ presence. This was especially true at the boundary between arid and mesic climates. These strong environmental gradients are the same ones which have driven high differences in native species presence across the province. [See pamphlet describing invasive plants in Cape Town.] This finding supports the “Goldilocks hypothesis”: that non-native plant species assemblages are driven by the same abiotic variables as native species assemblages. While did not directly study the “Biome decides hypothesis” (the composition of non-native flora is mediated by the biotic effects of native flora and fauna), Gildenhuys et al. (2026) doubt its applicability here because native species’ presence has probably been greatly reduced by the effects of urbanization.
Urbanisation intensity itself ranked third as a factor. Its effect was strongest at low to medium urbanization intensities. Because urbanization creates novel habitats, such as, “hardscapes” of paved surfaces that resemble deserts, their non-native plant assemblages are dominated by similar, urban specialist, species. At lower urbanization intensities a greater variety of habitats is available. Gildenhuys et al. (2026) conclude that urbanization acts primarily as a driver of opportunistic habitats for species at later invasion stages rather than as a filter of species introduction.
An earlier study found a similar effect from road density (often associated w/ urbanization) as an explanation for where specific woody non-native species establish. They do concede that larger urban areas might experience greater propagule pressure.
Gildenhuys et al. (2026) note that recent globalization of the plant trade has probably changed the specis planted in urban centers. For example, cities in the Western Cape are increasingly replacing English oak (Quercus robur) with more disease-resistant oaks. The change might reflect greater environmental awareness and regulations issued under the National Environmental Management: Biodiversity Act 10 of 2004. In newly established urban centers, fewer invasive species are being planted — at least among trees.
SOURCES
Gildenhuys, C.P., L.J. Potgieter, C. Hui, D.M. Richardson. 2026. Drivers of compositional turnover of the NIS urban flora in the W Cape, South Africa. Urban Ecosystems (2026) 29:51 https://doi.org/10.1007/s11252-026-01919-3
Le Roux, M.M., R.R. Klopper. 2025. Taking stock of South Africa’s flora. South African Journal of Botany 184 (2025) 571-579
Mbobo, T., D.M. Richardson, A. Datta, K.T. Faulkner, J.R.U. 2025. Wilson. Spatially-Differentiated Reg of NIS Can Be Improved Using Spp Distribution Models: Psidium guajava in South Africa as a Case Study. Diversity and Distributions. 2025 31:e70102 https://doi.org/10.1111/ddi.70102
Mudada, N., J. Chitamba, E. Nyangani, C. Chapano, N. Mapope,and W. Ngezimana. 2026. Weeds associated with cross border traffic, their approach and infestation rates in Zimbabwe. ISABB Journal of Food and Agricultural Sciences. Vol. 12(1) January-June 2026. DOI: 10.5897/ISABB-JFAS2025.0192
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
dying ash in Shenandoah National Park; photo by F.T. Campbell
The emerald ash borer (Agrilus plannipennis; EAB) was detected in North America in 2002. So both U.S. and Canadians have been motivated to evaluate the probable trajectory of the primary hosts – the ash genus Fraxinus – in the face of the ongoing invasion. See Deschênes et al. 2026 and Wilson et al. 2025 – full citations at the end of this blog. Both studies focused on white (Fraxinus americana) and green ash (F. pennsylvanica); they say next to nothing about black ash (F. nigra). I regret this silence because of the unique ecology of black ash swamps. Neither addresses the threat to Oregon ash (F. latifolia) in the West.
The two assessments have similar findings: high mortality of larger trees (canopy and “recruit” size trees); abundant regeneration (seedling and saplings sizes) after an initial period; and uncertainty as to whether persisting EAB populations will kill the saplings before sufficient numbers grow into reproductive size.
There are two conspicuous differences. First, the American study does not consider the possible impact of biological control – although USDA APHIS has placed all its effort on this approach since January 2021. The Canadians report that self-sustaining populations of the wasps are now found across the northern U.S. and eastern Canada. In their study, conducted in Ontario, they detected only Tetrasticus planipennisi; it was parasitizing 16% of the EAB larvae in dissected stems. This wasp’s affinity for colder climates and short ovipositor – which limits it to parasitizing larvae inhabiting small stems – are portrayed as positive traits under these circumstances.
Second, the Canadians did not find “lingering” adult ash trees as have the Americans. These trees indicate the probability of finding workable levels of genetic resistance to the EAB. USDA Forest Service scientists are pursuing a breeding program. While in south-central Michigan, where overstory ash mortality typically exceeded 80%, 46% of overstory ash and 82% of ash recruits were relatively healthy (Wilson et al. 2025), in Ontario none of 1,129 overstory ash trees survived beyond seven years after EAB detected. No trees exceeded 15 cm dbh (Deschênes et al. 2026).
ash resistance breeding trial at Holden Arboretum; photo courtesy of Jennifer Koch, USFS
Regeneration
As Deschênes et al. (2026) state, the future of ash stands depends on the complex of interactions among environmental conditions, management interventions, efficacy of natural enemies (natural or introduced), and life-history traits of the insect and its host. Coexistence might be possible if EAB larval densities remain sufficiently low to support survival of residual trees and successful seedling recruitment.
Larval densities in Ontario were said to be generally low, suggesting reduced carrying capacity in post-invasion forests, lower EAB fecundity, and higher EAB mortality in regenerating stems. Deschênes et al. (2026) note that in more northern areas colder temperatures are thought to slow larvae development. Perhaps these larvae might also be less vigorous, so they night disperse only over short distances. Still, there were sufficient EAB present after all the overstory trees had died to create 97% of the 298 galleries in regenerating ash stems (Deschênes et al. 2026). Furthermore, Wilson et al. (2025) say that EAB densities in infested trees in Michigan were similar to densities recorded during the initial invasion. This seems ominous to me – a solid foundation from which beetle populations could build up again as regenerating ash grow and provide more phloem for the insect to exploit.
Ash reproduce by both flowering/seeding and sprouting from the base. EAB predation is not the only complication. First, ash are dioecious so mature trees of both sexes must grow within a few hundred meters. Second, predation by the ash seed weevil (Lignyoodes helvolus) reduces seed supplies. Dense sedge mats can prevent germination (Wilson et al. 2025). Scientists generally believe that the soil seed bank is quickly exhausted, although Wilson et al. (2025) cite others’ conflicting findings. Neither article discusses predation by mammals, e.g., deer or rabbits. Wilson et al. (2025) mention attacks by beavers.
ash saplings felled by beavers; photo by F.T. Campbell
Wilson et al. (2025) did not study whether stump-sprouted ash were able to successfully recruit into the overstory. They do report that in one study in southeastern Michigan stump sprouts were the dominant form of green ash regeneration and about a quarter of these sprouts produced seeds at least once. Deschênes et al. (2026) found that on average 47% of regenerating stems at their Ontario research sites originated from stump sprouts.
EAB has been documented to attack and kill trees when the main stem is as small as 2.5 cm. While EAB probably prefer larger stems, Deschênes et al. (2026) suggest that stems become acceptable at the lower range of size required for seed production – 8–10 cm dbh. Reliable and abundant seed production doesn’t occur until white or green ash achieve > 20 cm dbh. At their Ontario sites, Deschênes et al. (2026) found that 42% of regenerating stems has been infested by EAB at least once; 14% had been attacked five or more times. They removed 74 EAB larvae from 28 stems; 49 (66%) were alive. Fifteen EAB (16% of current year galleries) had been parasitized — all by Tetrastichus planipennisi. They also observed numerous signs of defensive responses.
In Michigan, no ash recruits — living or dead – were found in plots in 28% of the cells. In the remaining 128 cells, an average of ~33% of ash recruits were infested by EAB, and ~21.4% of ash recruits dead. As is typical, white ash fared better than green ash. Recruit sized ash trees were twice as likely to die than to survive and mature into overstory size (Wilson et al. (2025).
In Ontario, as noted, all canopy ash had died. There were 119 live trees 5 – 10 cm dbh – a tenth as many “mature” ash as when EAB arrived, and all were smaller. There was abundant regeneration in most sites initially, but at the longest-infested sites in Essex County, regenerating ash stems were half as numerous as early after the transition (Deschênes et al. 2026).
The Canadians found it encouraging that some of the regenerating stems were vigorous despite containing EAB gallery densities greater than 20 larvae·m?2 of phloem. They did not know the mechanisms underlying survival of these stems. Possible explanations ranged from the low EAB carrying capacity of smaller trees to stronger host defenses in regenerating stems to EAB mortality due to parasitism.
Wilson et al. (2025) note that despite more than 20 years of EAB presence, densities of ash recruits, saplings, and seedlings were high relative to other species. However, they remind us, ~ one-third of the live ash recruits were infested so their survival into reproductive size was uncertain. The high mortality of overstory ash results in loss of seed resources, greater sun exposure, and cascading consequences for forest composition and function. In upland sites, cells with low ash basal area favored Quercus rubra and Tilia americana. They conclude that changes to forest composition is probably site specific — largely depend on what tree species are already present.
Despite the challenges described above, the Canadian scientists also believe that these findings demonstrate that ash has a capacity for long-term regeneration (Deschênes et al. 2026).
Changing Species Composition in the U.S. (Wilson et al. 2025)
Canopy gaps caused by ash mortality have largely been filled by lateral ingrowth of species already there — American elm (Ulmus americana), black cherry (Prunus serotina), and northern red oak (Quercus rubra). The regeneration strata (saplings and seedlings) is dominated by Fraxinus (white outnumbering green when differentiated), maples (Acer rubrum, A. saccharum), black cherry, Crataegus species and Carya ovata. Elms are consistently among most common non-ash taxa among overstory, recruit, sapling and seedling strata. At some Ohio sites there was also increased abundance of non-native tree and shrub seedlings. This is not surprising since invasive plants are widespread in the forests of Ohio and other eastern states. A decade ago 93% of Forest Inventory and Analysis (FIA) plots in Ohio had at least one of 50 invasive plant species.
In another paragraph they mention Tilia americana as one of the important species in these forests.
Situation in Canada (Deschênes et al. 2026)
Deschênes et al. (2026) express concern that the death of nearly all canopy-level trees will substantially reduce ash’ ability to fulfill its ecological roles in these ecosystems. Still, ash regeneration is persisting for decades following overstory mortality. The taxon’s continued presence is driven largely by strong sprouting, which has been observed in several locations in Ontario. In some areas, low EAB infestation rates and evidence that regenerating stems can withstand multiple infestations raises hope that some might reach maturity and produce seeds. This scenario would be similar to that of elms, in which surviving trees contribute to ongoing regeneration and might eventually facilitate development of some level of resistance to the invasive fungus. A second possibility is that ash’ high sprouting capacity might point to a scenario similar to that of American chestnut. This species has persisted for a century primarily as sprouting shrubs — although they rarely reach reproductive maturity.
white fringetree; photo by Ryan Somma via Wikimedia
White Fringetree
Scientists also reviewed the status of a secondary host of EAB in North America, white fringetree (Chionanthus virginicus). Earlier studies of this host-pest relationship had been conducted on ornamental plantings where the trees tend to be scattered across open lawns and actively managed – including protection from pests. The Cipollinis (see full citation at the end of this blog) believe they might be better able to ward of EAB attack than are wild, unmanaged trees in forests that must compete for resources. They wanted to assess the current status and likely trajectory of the tree species in the wild.
To do so they revisited a wild population of the tree in southern Ohio previously assessed 10 years earlier. White fringetree is a small multi-stemmed tree native to the southeastern U.S. It is widely planted as an ornamental in across the east. In Ohio, white fringetree grows wild in only a few southern counties, in small populations or as widely scattered individuals. The species is classified as “Potentially Threatened” at the state level.
In 2015, 30% of the white fringetrees at the site were infested. These trees had signs of stress but none had died. EAB larvae grow more slowly on fringetree than on North American ash species. Meanwhile, all mature white ash trees at the site had been killed by EAB. Smaller white ash trees more comparable in size to the white fringetrees had attack rates and impacts comparable to those on the fringetrees.
In their new study, the Cipollinis found that nine of 31 trees tagged in 2015 (29%) had died; 22 (71%) were alive. Of those 22 living trees, 12 (55%) stayed the same or improved slightly over the five-year period; 10 (45%) declined. Five of these 22 living trees (23%) had evidence of current infestation. Trees that had died had a higher incidence of old EAB galleries, adult exit holes, and woodpecker activity. This is interpreted as demonstrating that EAB must cause extensive damage to kill fringetrees.
In summary, fringetrees in a wild unmanaged population continued to be attacked by EAB over 10 years and suffered higher attack rates and more significant impacts than those previously observed in managed pops. The Cipollinis conclude that trees large enough to attract EAB oviposition will continue to decline in health and be killed as long as beetles are present. They expect that wild white fringetrees might meet the same fate as ash trees, but over longer time scales.
At the same time, this delay in complete mortality might create a refugium for remnant populations of EAB after most ash have been killed. This status would be exacerbated if it turns out that the biocontrol agents cannot find their target — EAB — in the alternative host. The Cipollinis found lower parasitism rates by Tetrastichus planipennisi in fringetrees, although this was not true for the egg parasite Oobius agrili and two Spathius larval parasites.
Whitebark fringetree populations can produce few adult EAB because the trees are small and contain low amounts of phloem. Still, as young trees grow into vulnerable sizes they might help sustain the EAB population – as young ash trees in the area appear to do.
While caution is appropriate in interpreting findings from a study of a single population, the Cipollinis argue that this population has been studied intensively: assessed six times over 10 years, beginning at the start of the EAB infestation. Therefore they think their analysis provides useful informative regarding the long-term impacts of EAB on fringetree.
They concede that larger populations in areas deep within the tree’s native range might experience different dynamics and impacts. So far, however, observations in Chattahoochee National Forest in Georgia and at Great Falls Park on the Maryland-Virginia border generally support their finding that wild fringetrees in natural landscapes will suffer higher attack rates and be more severely impacted by EAB than trees in managed landscapes.
Finally, the Cipollinis fear that a close relative, pygmy fringetree, Chionanthus pygmaeus, is at particularly high risk because it is endemic to only a few counties in the sandhills of central Florida. The species is already classified as endangered by both the state and the federal governments. The pygmy fringetree is smaller than white fringetree, so its size might help it escape attack. However, adults achieve sizes comparable to that of fringetree in some cases. So when EAB reaches Florida, the specie appears to be highly vulnerable.
SOURCES
Cipollini, D. and K. Cipollini. 2026 The Fate of a Wild White Fringetree (Chionanthus virginicus) Population in Ohio 10 Years After Invasion by Emerald Ash Borer (Agrilus planipennis) Forests 2026, 17, 712
Deschênes, É., C.J.K. MacQuarrie, L. Scott, C. Zimmerman, and I. Aubin. 2026. Ash population dynamics after two decades of emerald ash borer infestations in Canada. Canadian Journal of Forest Research. Can. J. For.Res. 56: 1–13 (2026) | dx.doi.org/10.1139/cjfr-2026-0075
Wilson, C.J, L. Labbate, T.R. Petrice, T.M. Poland, D.G. McCullough. 2025. Ongoing regeneration of ash and co-occurring species 20 years following invasion by emerald ash borer. Forest Ecology and Management 580 (2025) 122546
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
Photo of infested cactus at Cabo Rojo National Wildlife Refuge, Puerto Rico. Taken August 20, 2018 by Yorelyz Rodríguez-Reyes
For 15 years I have maintained a profile of the Harrisia cactus mealybug Hypogeococcus pungensbecause it threatens columnar cacti in the subfamily Cactoideae from the Caribbean basin and possibly in the American Southwest and Mexico. My recent attempts to clarify the current situation have been frustrated by the apparent collapse of funding support for scientists trying to conserve these cactus species.
The dry regions of the Caribbean Islands are home to about 100 native cacti, 75% of which are endemic. According to the Center for Plant Conservation, 20 species are listed as threatened by the IUCN. Puerto Rico specifically is home to 14 cactus species, at least three of which are endemic.
How and when the mealybug first invaded the Caribbean and North America is unknown. The presence of the insect now known as H. pungens Hyp-C on Puerto Rico was detected in the Guánica Commonwealth Forest and Biosphere Reserve on the island’s southern coast in 2005 (Zimmerman et al. 2010). However, the actual introduction probably occurred about ten years earlier, in about 1996 (Poveda-Martinez et al. 2022). See map of locations below.
In the 20 years since then, the mealybug has spread across the island’s dry districts. By 2010, it was estimated to be present on about 1,400 km2. By 2014 – nine years after detection — the mealybug had reached the small island of Caja de Muertos. The most recent survey of which I am aware (date unclear) detected the mealybug on 268 out of 445 cactus plants examined (60%) in 12 out of 39 sites examined (Poveda-Martinez et al. 2022). The mealybug is also killing native cacti on the nearby U.S. Virgin Islands (Poland et al. 2019), although I have found no data on this invasion or its impact.
Below – columnar cacti on St. John, US Virgin Islands; photos by F.T. Campbell
H. pungens Hyp‑C threatens seven of 14 native cactus species in Puerto Rico. Three of the cacti are endemic; two are federally listed as endangered species: Harrisia portoricensis and Leptocereus grantianus (USDA ARS). Since the mealybug’s detection in Puerto Rico, it has caused extensive damage to Pilosocereus royenii (Royen’s tree cactus), Leptocereus qaudricostatus (pitaya), Melocactus intortus (turk’s cap), and an introduced cultivar, Cereus hexagonus. It has caused minor damage to Stenocereus fimbriatus (Zimmerman et al. 2010). These cacti provide food or shelter for endemic bats, birds, moths and other pollinators (Segarra and Ramirez; USDA ARS).
The insect’s attack promotes abnormal gall-like growth on the stem and deformed flowers. These deformations severely affect infested plants’ reproduction and eventually survival (Poveda-Martinez et al. 2022).
Biological Control
When the mealybug was first detected Commonwealth and federal agencies tried to counter it. A search for possible biocontrol agents in the insect’s native range in Argentina began the 2010. While no funds have ever been appropriated for this activity, for several years the U.S. Department of Agriculture supported the work by allocating funds to the Agriculture Research Service Insect Behavior and Biocontrol laboratory in Gainesville, Florida, from broader programs. Dr. Stephen Hight took the lead, working with colleagues in South America. According to Dr. Hilda Diaz-Soltero, then a USDA official, these funds came primarily from the USDA Invasive Species Coordination Program and APHIS Eastern Region. In fiscal years 2017 and 2019, an additional ~$550,000 came in the form of grants under APHIS’ Plant Pest and Disease Management and Disaster Prevention Program. Link Scientists at the Center for Excellence in Quarantine and Invasive Species at the University of Puerto Rico devoted at least a decade to the search.
Scientists focused on two parasitoid wasps, Anagyrus cachamai and A. lapachosus (Hymenoptera: Encyrtidae). A third candidate, the predator Hyperaspis conclusa, was also assessed (Aguire et al.).
Research on the mealybug-wasps interaction uncovered troubling patterns. First, it has long been known that some mealybugs believed to be belong to the species Hypogeococcus pungens feed on columnar cacti while others feed on plants in two unrelated families, Amaranthaceae and Portulacaceae (USDA ARS; Zimmerman et al. 2010). Would the introduced wasps attack the cactus-feeding mealybug in sufficient numbers?
The confusion over how many mealybug species have been introduced – and where – severely hampered development of a program. (The mealybug has been introduced to control invasive cactus in Australia and South Africa. Most sources say it has been highly effective – prompting the initial concern when it appeared on Puerto Rico.) I have been unable to find any information about the status of the candidate biocontrol agents more recent than 2022.
Genetic Conservation
The USDA also partnered with the Naples (Florida) Botanical Garden to collect fruits and vegetative material for ex situ conservation. Rigorous phytosanitary procedures were followed to ensure the absence of the mealybug. Collections of fruits and vegetative material provided 1,298 cacti samples from 13 species, representing 1,173 maternal lines from 91 sites throughout Puerto Rico. A total of 90,720 seeds representing 8 species are banked at the NBG for long-term storage. Propagation of the vegetative material has 56% success, and plants are incorporated into the NBG’s living collections. (These figures include Opuntia cacti that are hosts of a second invasive insect, Cactoblastis cactorum.)
Genetic Concerns
Scientists now consider Hypogeococcus pungens (Hemiptera: Pseudococcidae) to be a species complex composed of at least five putative species. The species are separated in part by the plants they use as hosts. Two of the complex have apparently been introduced to Puerto Rico: H. pungens Hyp‑C feeds on cacti; H. pungens Hyp‑AP feeds on hosts in the Portulacaceae & Amaranthaceae. Both evolved from putative source populations in Brazil (Poveda-Martinez et al. 2022).
The two species H. pungens Hyp-C and Hyp-AP are currently separated on Puerto Rico by host preferences and climatic niches. They also occupy different geographic areas. Scientists fear that ongoing climate change could allow H. pungens Hyp-C to establish farther into the island’s interior and in a large area in the north. Such range expansion would end the geographic separation. Overlapping of the two species is likely to exacerbate the threat to Puerto Rico’s cacti. Most directly, it would complicate implementation of management strategies, especially biological control. Intermixing of the two species could also facilitate hybridization which might result in more vigorous attacks or a broadened host range. Hybridization is frequent in closely related species (Poveda-Martinez et al. 2022).
The Mealybug is Frequently Introduced
Mealybugs that feed on cacti and believed to be in the species Hypogeococcus pungens made multiple appearances in southern California between 2000 and 2018 – in gardens and in nurseries. Confusingly, CDFA reports interception of the mealybug on alternanthera and ludwigia plants shipped from Florida (CDFA 2018). I have no more recent data. The population in Florida was reported to be present in 16 counties in 2009); it might be the species that feeds on plants other than cacti (Poveda-Martinez et al. 2022). Other populations has been reported in the Dominican Republic (no date) (CDFA 2018); and in Hawai`i in 2005 (Hawaii Department of Agriculture new pest report). A mealybug that feeds on Amaranthaceae and Portulacaceae was detected in 2000 in San Juan, Puerto Rico (Poveda-Martinez et al. 2022).
In the absence of control measures, scientists expect H. pungens Hyp-C to continue decimating Puerto Rican cactus diversity and threaten other cactus rich ecosystems across the Caribbean islands, Central America and, potentially, North America (Poveda-Martinez et al. 2022).
saguaro and organ pipe cacti in Organ Pipe Cactus National Monument; photo by F.T. Campbell
North America has more than 500 columnar cactus species in the Cactoideae (Zimmerman et al. 2010). Some of these cacti are already endangered, e.g., several Pediocactus. Others are totems of the desert, e.g., the saguaro (Carnegiea gigantea) and organ pipe (Stenocereus thurberi) cacti. Picture The larger ones, particularly, play important ecological roles. It is not known how vulnerable individual species are to the mealybug (Golubov pers. comm. January 2011). In Mexico several mealybugs in the same genus are already present. The natural enemies of these mealybugs might be able to attack H. pungens Hyp-C if it invades the country (Zimmerman et al. 2010). Despite the well-founded concern, apparently no funds have been allocated by governments or conservation organizations to studying the vulnerability of these cacti to one or more mealybugs in the Hypogeococcus genus.
The most likely pathway by which the mealybug is spread is the trade in plants for planting (the horticultural trade) (Zimmerman et al. 2010). A decade ago APHIS reported intercepting mealybugs on cactus (primarily on roots) imported from Germany, Peru, and Puerto Rico. APHIS has also intercepted several other mealybugs in the same genus – on plants (including orchids and bromeliads as well as cacti) from Belize, Costa Rica, Ecuador, Guatemala, Honduras, Mexico, Panama, Peru, and Venezuela (USDA APHIS alert).
A decade ago NatureServe and IUCN found that 31% of Earth’s cactus species were threatened with extinction. They named overharvesting (often for the illegal horticultural trade) and destruction of habitat by smallholder livestock ranching and farming. Did not mention predation by introduced insects – although that is now manifest not only in the cactus mealybug but also the cactus moth.
Sources
Aguirre, M. G. Logarzo, S. Triapitsyn, H. Diaz-Soltero, S. Hight, O. Bruzzone. 2023? Effect of egg production dynamics on the functional response of parasitoids
California Plant Pest and Disease Report. 2005. Vol. 22 No. 1. Covering Period from July 2002 through July 2005.California Department of Food and Agriculture. 2018.California Pest Rating for Hypogeococcus pungens Granara de Willink | Harrisia cactus mealybug Hemiptera: Pseudococcidae Pest Rating: A California Pest Rating for Hypogeococcus pungens Granara de Willink | Harrisia cactus mealybug Hemiptera: Pseudococcidae Pest Rating: A
Poland, T.M., Patel-Weynand, T., Finch, D., Miniat, C. F., and Lopez, V. (Eds) (2019), Invasive Species in Forests and Grasslands of the United States: A Comprehensive Science Synthesis for the United States Forest Sector. Springer Verlag.
Poveda-Martinez, D. N.A. Salinas, M. Belen Aguirre, A.F. Sanchez-Restrepo, S. Hight, H. Diaz-Soltero, G. Logarzo, and E. Hasson. 2022 Geonomic & ecol evidence shed light on the recent demographic history of two related invasive insects. Scientific Reports.
Segarra-Carmona, A.E., A. Ramirez-Lluch. No date. Hypogeococcus pungens (Hemiptera: Pseudococcidae): A new threat to biodiversity in fragile dry tropical forests.
Segarra-Carmona, A.E., A. Ramírez-Lluch, I. Cabrera-Asencio and A.N. Jiménez-López. 2010. First Report of a New Invasive Mealybug, the Harrisia Cactus mealybug Hypogeococcus pungens (Hemiptera: Pseudococcidae). J. Agrie. Univ. RR. 94(1-2):183-187 (2010)
Srivastava, M., P. Srivastava, R. Karan, A. Jeyaprakash, L. Whilby, E. Rohrig, A.C. Howe, S.D. Hight, and L. Varone. 2019. Molecular detection method developed to track the koinobiont larval parasitoid Apanteles opuntiarum (Hymenoptera: Braconidae) imported from Argentina to control Cactoblastis cactorum (Lepidoptera: Pyralidae). Florida Entomologist 102(2): 329-335.
Triapitsyn, Aguirre, Logarzo, Hight, Ciomperlik, Rugman-Jones, Rodriguez. 2018. Complex of primary and secondary parasitoids (Hymenoptera: Encyrtidae and Signiphoridae) of Hypogeococcus species. mealybugs (Hemiptera: Pseudococcidae) in the New World. Florida Entomologist Volume 101, No. 3 411
USDA Agriculture Research Service, Research Project: Biological Control of the Harrisia Cactus Mealybug, Hypogeococcus pungens (Hemiptera:pseudococcidae) in Puerto Rico Project Number: 0211-22000-006-10 Project Type: Reimbursable
Zimmermann, H.G., M.P.S. Cuen, M.C. Mandujano, and J. Golubov. 2010. The South American mealybug that threatens North American cacti. Cactus and Succulent Journal. 2010 Volume 82 Number 3
Prostanthera cuneata – member of a genus endemic to Australia. Photo by Leonora (Ellie) Enking via Flickr
In 2023 a global meeting of plant conservation experts convened by The Royal Botanic Gardens, Kew (U.K.) released the 5th edition of a report on the State of the World’s Plants and Fungi.
Associate Professor of Plant Ecology and Conservation Science Rachael Gallagher from Western Sydney University had led the global evaluation of conservation assessments for unique flora species. She is also the lead author of an article (2023; full citation at the end of this blog) evaluating how well countries around the world met their treaty obligation to assess the conservation status of endemic plant species native to their territories. The analysis identified 221,399 endemic plant species in a total of 173 countries. The treasure is not distributed evenly. Five countries harbor a third of the endemic plant species: in descending order, Brazil, Australia, China, Mexico, and South Africa. (The United States, including its islands, ranks 8th.)
On average, countries completed assessments of just 34% of their endemic species. New Zealand and here and South Africa shone: they assessed 87% of their unique species. China assessed 71%. One of the world’s poorest countries, Madagascar, evaluated 42% of its ~10,000 endemic plant species. Reminder: tiny Madagascar ranks 6th in the number of endemic plants. Australia – one of the richest countries– carried out the process for 39% — slightly more than the global average. Other countries that are stewards of numerous endemic plants were below the average: Brazil reviewed 29%, Mexico assessed only 24%.
Rachael Gallagher and her colleagues in the Australian Biodiversity Council were quite critical of Australia’s low level of performance. They called on their countrymen to do much more to prevent the decline and extinction of the country’s unique plant species. Australia, as party to the Convention on the Conservation of Biological Diversity, has a treaty obligation to prevent extinction of species which occur nowhere else. Remember, Australia’s flora and fauna rank extremely high on a scale of phylogenetic distinctness as an heir of the isolated continent of Gondwanaland.
Gallagher and colleagues concede that many endemic plant taxa in Australia have huge ranges — averaging 235,829 km2. But these vast expanses do not prevent sudden population crashes caused by calamities. They mention the megafires of 2019–2020 and – over the longer term – climate change. I think of the invasion by the rust fungus Austropuccinia psidii.
When we think about Australia, we wonder at the kangaroos and koalas. I assume Australians consider their unusual fauna to be iconic symbols of their country. Why are they not equally committed to their flora – 88% of their plant species are endemic. Do they suffer from the same “plant blindness” I have encountered in the United States? South Africa undertook an assessment of her endemic flora that concluded that a quarter of these species are threatened. Sixty percent of the country’s 20,000 plant species are endemic.
a protea in South Africa’s fynbos; photo by Michael Wingfield
[I have found no parallel analysis of America’s endemic plant species. Our nation’s rank of 8th in number of endemic species is explained by the highly unique floras of the islands, especially the Hawaiian archipelago. More than 95% of native species on the Islands are endemic. This includes 67% of the large trees still present in the forests (Potter et al. 2023).]
This study reflects the findings of the International Union for the Conservation of nature (IUCN)’s 2024 Red List of Threatened Species. A decade-long global project had found that at least 16,425 of the 47,282 tree species (38%) assessed are at risk of extinction. Trees accounted for over one quarter of species on the IUCN Red List. Tree species are at risk of extinction in 192 countries around the world.
Sources
Gallagher, R.V., S. P. Allen, R. Govaerts, M.C. Rivers, A.P. Allen, D.A. Keith, C. Merow, B. Maitner, N. Butt, T.D. Auld, B.J. Enquist, W.L. Eiserhardt, I.J. Wright, J.C.O. Mifsud, S. Espinosa-Ruiz, H. Possingham, V.M. Adams. 2023. Global shortfalls in threat assessments for endemic flora by country. Plants, People, Planet. DOI: 10.1002/ppp3.1036
Potter, K.M., C. Giardina, R.F. Hughes, S. Cordell, O. Kuegler, A. Koch, E. Yuen. 2023. How invaded are Hawaiian forests? Non-native understory tree dominance signals potential canopy replacement. Landsc Ecol 2023 https://doi.org/10.1007/s10980-023-01662-6
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
In a press release on 31 March, 2026, the USDA announced major changes to the USFS structure. Agency headquarters will be moved to Salt Lake City. They point out that nearly 90% of USFS land is west of the Mississippi … but promise to sustain engagement in the Southeast (America’s “wood basket) by creating a regional office there. Furthermore, they will change the current regional organization to a state-based one; they plan to create 15 state directorships. State directors will serve as national leaders with primary oversight of forest supervisors, operational priorities, & relationships with states, tribes, & other partners. Each state office will include a small leadership support team responsible for functions such as legislative affairs, communications, & intergovernmental coordination.
There will still be some “operational service centers” in other cities; that for research will be in Fort Collins. The goal is to unify research priorities, accelerate the application of science to management decisions, & reduce administrative duplication. Information on which facilities will be retained or closed is available at this webpage. (I could not open this site.)
No specific information is provided re: forest health management program.
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
The US Department of Agriculture (USDA) and the North American Invasive Species Management Association (NAISMA) held the 34th annual forum on invasive species research at the end of February 2026. The agenda is available here. In this blog I summarize the presentations about invasive alien plants (IAS); a separate blog discusses findings on invasive plants. Formal proceedings will be available in some months.
The most important information from the meeting:
If NAISMA had not taken on the task of hosting the conference it would not have happened.
Government leaders allowed only 1 staffer per USDA Forest Service region to participate. Not allowed to come were people who had organized the whole meeting or individual sessions, and presenters discussing several topics, including preventing IAS plant spread, and progress on controlling cogongrass (major impediment to pine plantations, affecting harvests).
What do these decisions say about the genuineness of the USDA Secretary’s recent memorandum listing invasive species as one of four priority areas for the department’s research efforts?
The USFS International Program is one of the few sources of support for studying potential pests before they invade the US.
Early detection surveillance is undermined by reliance on deploying too few traps and in a too narrow, or the wrong, timeframe.
The Resistance Screening Center in Asheville, NC is no longer staffed, undermining breeding efforts in a region that reaches from Virginia to Texas.
A reminder to us all: Rebekah Wallace of the Center for Invasive Species and Ecosystem Health at the University of Georgia urged us all to provide citations for images used in informal materials – posters, presentations, outreach efforts, blogs, videos. Providing the citation increases our credibility and ensures that we avoid perpetuating mis-information!
an ash resistance breeding plot at the Holden Arboretum, Ohio
Summary of key research reports on tree-killing arthropods and pathogens
Jennifer Koch, researcher with the USFS, described the Trees in Peril program. TiP aims to increase the pace, scale and efficiency of resistance breeding programs for American beech; eastern hemlock; and green, white, and black ash. This includes integrating genomics with other approaches and strengthening partnerships. Partners are key to finding “lingering” trees, addressing some scientific questions, and possibly screening cuttings for resistance.
Koch first explained the value of resistance breeding for producing resistant stock for restoration and reducing habitat for pests. The goal is to develop resistance, which Koch defined as the ability of a tree to survive despite the pest. Full immunity is not required. TiP participants hope that by integrating breeding with other approaches, such as biocontrol, they can create a new ecological equilibrium in which the tree species will continue to play its ecological role. As Koch asserts, the public supports breeding more than some other approaches. Also, there is a record of success; she cites the USFS Dorena Genetic Resource Center, which has developed resistant seedlings for four five-needle pines and Port-Orford cedar.
The first step is to determine whether desired traits are inherited. Genomics and other tools can test cuttings while they are still young and small – a very important advance in efficiency. Still, once cuttings with the desired traits are identified, it often takes several rounds of breeding to raise resistance levels sufficiently high. Similar testing of immature clones later in the process also can accelerate creation of seed orchards.
Breeding programs also need to incorporate genetic diversity from across the species’ ranges. TiP partners are collecting genetic material from beech, hemlock, and ash trees across their extremely large ranges – much of eastern North America.
Finally, TiP is training additional people to contribute to these breeding efforts.
Progress on each taxon:
beech grafts in a breeding experiment at the Holden Arboretum
Beech – Breeders are dealing with two diseases. A decade ago they identified genetic markers associated with beech bark disease (BBD). Their efforts had led to orchards producing seedlings of which 50% are resistant. Then beech leaf disease (BLD) showed up! Early results of a pilot study suggest BLD symptom severity is under genetic control. Even better, some trees appear to be resistant to both diseases. Koch recommends that scientists first identify BBD-resistant trees, then test those trees for BLD resistance.
Ash – the emerald ash borer (EAB) is established in 40% of the range of ash species. (Note: I am not sure whether this statement includes Canada; I am fairly certain it does not include Mexico.) Nine of the 16 US species are vulnerable, five endangered – green, white, black, blue and pumpkin.
The process by which scientists determe that resistance traits are heritable and identifying promising genotypes is described in Mason et al. (2026). The effort to develop techniques to propagate rooted cuttings is described in Merkle et al. (2022).
Partners are helping to search for “lingering” ash. So far, 265 trees have been identified, and scion collected from 106 trees. Partners are also helping to plant cuttings for resistance testing.
The program has had to overcome several difficulties, including:
Black ash is dioecious, which complicates selection. Breeders are working on several approaches, but all are at early stages.
Many of the originally collected trees turned out to be unintended crosses of white and green ashes rather than pure species. This resulted in very low seed production.
Anticipating the introduction of ash dieback disease (caused by the fungus Hymenoscyphus fraxineus), TiP is collaborating with Europeans on searching for possible resistance to this threat as well.
Hemlock – the Hemlock woolly adelgid (HWA) causes mortality of 50 – 100% of overstory trees. TiP scientists are still trying to establish a test for heritability of HWA resistance. There are additional difficulties in propagating rooted cuttings. The University of Georgia, Holden Arboretum, and others are helping to resolve these issues.
Those who want to support this program by contributing funds, knowledge, facilities, or volunteer efforts should contact Dr. Rachel Kappler, Forest Health Collaborative Coordinator, Holden Forests & Garden.
One entity already actively helping the TiP program is the Ecological Research Institute through energizing citizen scientists. Radka Wildova described these efforts. The Monitoring and Managing Ash [MAMA] initiative has published detailed guidance on identifying “lingering” ash. For example, timing is crucial: searching too early points to trees that are not actually resistant. Searching too late means opportunities are missed (since “lingering” ash will die eventually because resistance is only partial) or a risk of confusing in-growth or regeneration for “lingering” trees.
The Institute could not create a similar action map for hemlocks because the adelgid has been present far longer. Recommends searching in sites where at least 80% of surrounding trees are dead or dying due to HWA or elongate hemlock scale. The program is also testing heritability of resistance among hemlocks on its own property, which was invaded 20-30 years ago.
[An unrelated initiative, the Hemlock Restoration Intiative, is pursuing protection and breeding efforts in the southern Appalachian mountains.]
Avalon Miller, Pennsylvania State University, discussed new techniques to detect American elm trees tolerant of this disease.
a healthy American elm in Fairfax County, Virginia; photo by F.T. Campbell
It is important to detect elm trees’ response to infection early in the infection process because the apparent mechanism of tolerance is some trees’ ability to limit growth and proliferation of the causal fungus Ophiostoma novo-ulmi in xylem vessels. Scientists sought to use spectral analysis to detect distal leaf stress as a signal of susceptible genotypes. The USFS has developed a small stem assay that is achieving 80% accuracy in identifying disease phenotype within two months of inoculation – before symptoms appear.
Future studies will focus on determining which metabolites vary in tolerant vs. susceptible trees, and whether that information suggests useful interventions. For example, it is thought that some trees respond too aggressively to the pathogen, thereby cutting off the flow of water and nutrients and killing themselves.
Meanwhile, continuing efforts to breed resistant elm are hampered by limited greenhouse space, the tree’s complex genetics, and vast geographic range, and great variation in trees’ responses.
Current USFS- and The Nature Conservancy-supported programs focus on the Northeast. I urge scientists in the Mid-Atlantic to engage; I have seen numerous healthy American elms in the Virginia and Maryland suburbs that could be included in a breeding program.
Courtney Johnson, North Carolina State University, described efforts to determine key aspects of the ALB invasion in South Carolina. First, the bad news: a second invaded site in the Charleston region was detected in 2025.
Because Charleston is much farther south than any other ALB infestation, questions have arisen about
its phenology (timing of development). Research has confirmed that the ALB in South Carolina has ~1 year development cycle, not multiple generations as some had feared. Beetle larvae stay in the phloem through the third instar. Adult flight season is from May – Sept; the peak is in July. Unlike earlier findings, adult beetles did not exhaust their natal tree before moving to a new tree to oviposit. (This is also true in the Massachusetts outbreak.)
Some of the beetles in South Carolina are larger. Outreach materials need to be amended to reflect this fact, e.g., much larger exit holes.
typical site of ALB infestation in Charleston South Carolina; arrows indicate infested red maple trees. Photo by David Coyle
Tree dissection and dendrology studies of the principal host, red maples, show that multi-stemmed trees and smaller branches are preferred. They also preferred vertical stems or bolts, although they did oviposit on horizontal bolts raised off the ground to mimic a tree branch. There was little oviposition on bolts on the ground. In practice this means managers can leave felled trees on the ground without prolonging the infestation. This is very helpful since swamps preclude using heavy equipment. picture
Chad Rigsby, Bartlett Tree Research Laboratory, described the results of testing the efficacy of several nematocides. A foliar spray, Bayer’s Broadform, has received emergency approval from many states. It suppresses nematode (Litylenchus crenatae mccannii ([LCM]) numbers when applied at very low rates. Trees can be treated as long as (green) leaves are present. Rigsby recommended not spraying until a tree displays symptoms.
Since foliar sprays cannot be applied in forests, near water, or on huge trees, scientists also sought a systemic injectable fungicide. Thiabendazole [TBZ] (commercial formula Arborjet 20-S) is available. Rigsby said applicators can avoid splitting of the bark by following protocols developed by the International Society of Arboriculture. Managers should inject a tree several times in the first year to get the disease under control; then they can apply less frequently.
injection of thiobenzadole into beech; photo by Matthew Borden of The Bartlett Tree Research Laboratories
Don Grosman of Arborjet believes mortality is the result of a disease complex, not just LCM. Any of three treatments containing phosphite greatly reduces nematode numbers and canopy symptoms. Low volumes of diluted product can be injected in a few minutes. However, Thiabendazole hypophosphite requires a high volume macro trunk injection. This is expensive and takes time
Testing shows potassium phosphite PHOSPHO-jet produced dramatic improvement in 1 year. There are early indications that one treatment might be effective for two years. Arborjet will test this finding again this year. The company is also testing another chemical – the name of which cannot yet be revealed.
Andrew Miles, Ohio State, described beech response to polyphosphate (PP30). This chemical is a biostimulant, not a treatment. It is used as a disease control agent in several crops, including woody species. Field observations indicate it does reduce disease severity. Scientists are trying to understand the mode of action. Experiments are under way in Cleveland MetroParks, where BLD was first detected. Miles called for experiments within buds as well as leaves, since LCM damages tissue while in the bud.
Scott Schlarbaum, University of Tennessee, collects butternuts; photo by F.T. Campbell
Anna Conrad, USFS, described ongoing efforts targetting this disease, which is present throughout the tree’s large range. A major challenge is distinguishing pure butternut from hybrids with Japanese walnut. Scientists have screened ~300 families from 22 states for possible resistance. At three sites in Indiana, the vast majority of highly resistant families are hybrids. Still, resistance was detected in up to 2.5% of pure butternuts; this level is sufficient to be enhanced through breeding. The program would benefit from genotyping across butternut’s range to identify lingering trees and confirm resistance.
Nicholas Dietschler, Cornell University, studies the relationship between western hemlocks and HWA in their shared native ranges in the Pacific Northwest. At all sites, lower numbers of HWA (of both PWN and Japanese lineages) survived on Western hemlock – in the absence of predators. Why? Dietschler believes western hemlock has better chemical defenses. For example, hemlocks exude pitch in response to adelgid herbivory. In eastern hemlocks, this induced resin might suppress the tree’s defenses. In addition, HWA also prompts greater suppression of phenolics in eastern hemlock. Dietschler concludes that bottom-up, tree-based defenses are a factor in the invasion and should be studied — while continuing efforts to find an effective combination of biocontrol agents.
Anne C.J. Peter, of Virginia Polytechnic Institute and State University, is comparing HWA chemical interaction with the most recent biocontrol agent, the silver fly Leucotaraxis argenticollis. (Scientists hope L. argenticollis will feed on summer populations of HWA; other biocontrol agents don’t suppress HWA at this stage.) The L. argenticollis population in the PNW feeds on HWA; however, its eastern North American relative L. rubidus feeds on pine adelgids, not the introduced HWA. It has been challenging to establish the PNW population in the East. One possibility is that the invasive HWA, which is from Japan, contain toxins that deter predators & parasitoids. Therefore, Peters is studying how both the western and eastern populations of Leucotaraxis deal with anthraquinones – compounds found in many plants and some insects, but not adelgids native to the eastern US.
Jian Duan, of the Agriculture Research Service Beneficial Insect Lab, summarized results of 15 years of biological control efforts. Over this period, four biocontrol agents have been introduced. I applaud APHIS’ rapid inclusion of this pest management approach; an egg parasitoid and two larval parasitoids were introduced before 2010, less than 8 years after the invasion was detected. Unfortunately, these agents proved less effective in northern parts of the EAB’s distribution. A fourth larval parasitoid was released in 2015. One or more of these biocontrol agents have been released in 479 counties in 34 U.S. states and three Canadian provinces.
To what degree have the wasps reduced EAB populations? Are those reductions resulting in regeneration?
Duan reported that at sites in Michigan, all four agents have spread rapidly. EAB populations crashed and recovered several times but overall numbers are lower. Ash saplings increased greatly after 2015; seedlings also increased. He concluded that the program has been successful but not spectacularly so.
Hannah Broadley, APHIS, described developments beginning with initial searches for possible agents in China in 2015 — just one year after the lanternfly was detected in Pennsylvania. The search has focused on agents that feed on SLF eggs and nymphs. Attention has narrowed to Dryinus sinicus. This wasp both preys on and parasitizes SLF nymphs – depending on the nymphal stage. Labs are developing a third colony and conducting host specificity testing. Scientist have begun drafting a petition for release; the review process will probably take more than one year. At the same time, scientists continue exploring other possible biocontrol agents – e.g., in Vietnam. The blizzard prevented this speaker from appearing.
Xingeng Wang, of the ARS Beneficial Insect Lab, described how Dryinus sinicus attacks SLF – with a graphic video! D. sinicus attacks on third instar are often unsuccessful. When it encounters a second instar nymph, however, D. sinicus switches from predation to parasitism: it lay an egg which then develops inside the SLF nymph. This parasitism kill seven times more nymphs than predation on older nymphs.
Individual D. sinicus wasps can live up to 60 days, lay an average of 175 eggs and parasitize ~137 nymphs! Since D. sinicus is most effective against just one instar, releases will need to be carefully timed.
Alex Wu, APHIS, discussed efforts to prevent establishment of four flighted spongy moth (Lymantria) species. APHIS seeks to improve the efficiency of trap analysis because states are submitting triple the number of trap contents of past years. The goal is to improve real-time qPCR efforts to distinguish the European species established in the East from the Asian flighted species, and to distinguish the several subspecies of latter taxon. Current qPCR results point to the wrong species ~ 5% of time. There are complexities: moths from Central Asia might be hybrids. Also Lymantria dispar japonica might be found in far southeastern corner of Korea – which is separated from Japan by a narrow strait.
Early Detection of Wood-Associated Beetles
Jiri Hulcr, University of Florida, discussed strengths and weaknesses of artificial intelligence (AI) in species identification of bark beetles. As he noted, differentiating a specific bark beetle species from among the more than 6,000 look-alike taxa is time-consuming. A properly trained AI program can help. Furthermore, no one can keep up with publications – in 2015 there were 432 discussing just bark beetles! AI can help researchers discover papers that they otherwise would miss and empower non-English speakers to search the literature.
Hulcr has created a website that now has 63,000 images of ambrosia and bark beetles to assist identification. This work has been funded by the USFS International Program – one of the few sources of support for studying potential pests before they invade. The website will be open source once it has been copyrighted to prevent “scraping” by bots. Hulcr invited participants to send more images to continue training the algorithm on more species.
In the discussion, Alain Roques noted that scientists in Europe and probably China are developing similar AI-assisted identification tools. He urged international coordination. Hulcr replied that scientists do coordinate – as long as funding is available. Jennifer Koch noted that historic collections have many taxonomic inaccuracies. She urged people to rely on genetics when trying to identify a species.
Hulcr says AI is much faster than people in completing some tasks. But managing bioinvasions continues to require trained people (taxonomists) to collect, detect and classify new species; and execute quality control. AI cannot do science, which Hulcr defined as generating new knowledge through observation, turning that information into data, and testing hypotheses, making assumptions based on that.
Hulcr says AI also cannot predict what the next damaging ambrosia beetle to enter the U.S. will be. He offered his predictions:
Euwallaceae destructans – from Indonesia – attacks live trees
Aggressive Platypdinae from Asia and South America (especially threatening to plantations where trees are stressed)
Cryphaus lipingensis (attacks pine seedlings)
Scolytus amygdali from the Meditteranean region – introduces pathogens during maturation feeding on living hosts; feeds on almonds and prunes – Rosaceae
Dryocoetes himalayensis – Asia and Europe; kills walnuts
Port of Marseille; via Wikimedia
Alain Roques, of Zoologie Forestiere in France, reported results of a beetle trapping study in France.
Since the European Union allows entry of species not listed as quarantine pests, it is vitally important to improve detection and analysis of the large percentage of detections that are “unknown” or “emerging”. Nearly 8,000 beetles have been trapped over five years; they belong to nearly 400 species, 35 non-native.
One approach is to develop more generic traps and lures. The EU is now using a blend combining 10 pheremones to trap Cerambycidae. Scientists are incorporating additional pheromones to the blend and to extend attractiveness longer than the current 10 days. There is still no generic lure for Buperstids.
Some species arrive regularly – is each detection a reintroduction? Or are these species established?
Roques asks whether we are trapping at the right sites. Half of Cerambycids are trapped only inside ports (of various types). Scolytids were trapped outside ports, at other “high-risk” locations– e.g., sawmills and recycling centers. In other words, they disperse more broadly. Roques wants to add the road network and to extend the survey to the entire European Union.
Davide Nardi, of the University of Padua, Italy, discussed results of his trapping program, which seeks to guide placement of traps. See Nardi et al. (2026) [full citation at end of this blog]. Important conclusions are:
Surveillance programs are probably under-sampling species. Halving the sampling effort (from 16 to 8 traps) resulted in failure to detect ~20% of the species at the site. Cutting the sampling effort to four traps resulted in missing ~ 40% of present species. This decline in catches is particularly severe in urban landscapes – the very places where insects are most likely to be introduced. Even when they deployed 16 traps per site almost 30% of total species richness was not detected, on average.
Urban landscapes might offer a higher diversity of potential tree hosts. They also have more barriers to insects’ spread, e.g., buildings. This means urban areas require a greater sampling effort.
Traps should be set near available forest patches or urban parks.
I was intrigued by Nardi’s suggestion that scientists use the data on native beetles included in the trap catches to alert countries receiving exports from these ports to which species might be transported to their shores.
Manoj Pandey, of Ohio State University, explored how environmental context shapes abundance and diversity of Scolytines caught in surveillance traps. His goal is to improve the efficacy of the USFS’ two- decade-old Early Detection Rapid Response trapping program, which targets bark and ambrosia beetles at high-risk sites. These include transit sites, destination sites, and wood waste treatment sites. Pandey analyzed program catches from 2010 through 2019.
He found that among native species bark beetles dominated catches; ambrosia beetles dominated non-native captures. Climate [minimum/maximum temperature and precipitation] was the most important factor determining which species were caught. Overall, both Scolytines and ambrosia beetles are governed more by ecological requirements than by human population levels. Among Scolytines, native species (which are adapted to stressed trees) are affected by precipitation; non-native species are favored by warmer temperatures. Ambrosia beetles – both native and non-native – are more affected by precipitation levels than bark beetles, probably because of the formers’ symbiotic relationship with fungi. Ambrosia beetles are also more likely to be generalists and to be attracted by deciduous forests.
The other influential criterion was landscape – whether forests are evergreen, deciduous, or diverse. Deciduous forests attract both types of beetles, but the influence is stronger for non-natives. Conifer (evergreen) forests attracted native species. Higher human population density was associated with higher trap catches. Propagule pressure – measured via human population density and per capita income – was less important, perhaps because the traps are always placed near population centers.
Xyleborus monographus; photo by U. Schmidt
I am concerned because this trapping program did not detect the Mediterranean oak borer (Xyleborus monographus) before it was detected in California and Oregon. The project also did not find the greater shot hole borer Euwallaceae interjectus on the West Coast before it was detected in Santa Cruz, California. This ambrosia beetle has been established in the Southeast for years (M. Pandey, pers. comm. 12 March 2026).
Other Pests and Pathogens
Thomas G. Paul, at Ohio State, explored whether understanding the temperature regime during transit can provide early warning of which wood-associated pests might arrive. He obtained ocean surface temperature data along shipping routes from China to the U.S. West Coast and across the Atlantic. He then related those temperatures to degree-days needed for development by Xylosandrus germanus (from Asia) and Ips typographus (from Europe). At present there is still lots of uncertainty, including how to factor in the insect’s stage at the time of departure, the relationship between ocean air temperature and temperature inside a container, and possible effects of a container left to sit for several days in the port of import.
Eliana Torres Bedoya, also of Ohio State, provided an update on spore trapping for improved detection of pathogens across large landscapes. In 2024 the project developed standardized protocols for surveillance. To learn what is going on in the region, one should sample many sites across the area of interest. To find a particular pathogen, officials also need to know which season to sample. Torres Bedoya notes that few states sample in the autumn, which probably results in biased results.
In 2025, the program was expanded to 10 states. Species searched for are chosen by participating states. They include the causal agents of oak wilt, thousand cankers disease, laurel wilt, annosum root disease, and the beech leaf disease nematode (Litylenchus crenatae mccannii). Participants – including state phytosanitary officials — are now asking how to respond to a detection. For example, DNA from Bretziella fagacearum, the cause of oak wilt, was detected in several states where no disease has yet been identified (New Hampshire, Massachusetts, West Virginia, and Ohio). DNA of Geosmithia morbida, the causal agent of thousand cankers disease of walnut, was detected in New Hampshire, Massachusetts, and Maryland. What should managers do in response to these findings?
Torres Bedoya explained that her team is now working to make the spore-trapping process more user-friendly. I noted that my poster previous blog discussed using these techniques at the interface of forests and agricultural land uses.
During other discussions, I learned that Jason Smith of the University of Mount Union is trapping for DNA from LCM in order to track the spread of BLD.
Brown spot needle blight
Several speakers addressed this disease, which is of increasing concern to pine timber interests in the American South and around the world. New Zealand is exploring resistance breeding of Pinus radiata in advance of introduction
The disease has long been known in long-needle pine – at the “grass” stage (early seedlings). In recent years needle blight has begun damaging loblolly and other pine species – in both plantations and natural forests. Jason Smith, from the University of Mount Union, was asked for help by the industry in 2016. He found that one factor is increasing reliance on herbicides instead of fire to control ground-level vegetation. The large doses of inoculum remains in the litter, rather than being killed by periodic fire – as in the past. Smith thinks it is also possible that the pines suffer subtle damage from herbicides. Other possible factors are the widespread planting of genetically identical monocultures and climate change.
Colton Meinecke at the University of Georgia reported that Lecanostica acidola has been confirmed as the disease agent at these sites by Koch’s postulates. Scientist at the University of Georgia, University of Mississippi, and other entities are collaborating on development of a predictive model. Work includes sampling needles from both the litter and canopy, tracking tree condition, destructive sampling of dead trees, and spore trapping.
In the discussion, Smith warned that dying pines are not being detected by aerial forest health surveys because they are conducted too late in the season. This is because the surveys focus on one specific pest, the southern pine beetle. He called for a more comprehensive survey program.
Meinecke reported that the disease is more severe in western parts of the Gulf Coast regions. It is also causing problems in Christmas tree plantations, especially Scots pine.
He has found evidence of some genetic resistance. He is trying to develop a rapid test of a tree’s vulnerability using spectral wave length. Meinecke is also experimenting with stand management approaches. He praised the close cooperation with experts from around globe and New Zealand’s pro-active preparation for combatting the disease before it arrives.
Kier Klepzig, of the Jones Center at Ichauway in Georgia, described establishment of a Pine Pandemic Preparedness Plan, stimulated by awareness that a non-native pest might be introduced that attacks loblolly pine (Pinus taeda) – the foundation of the southeastern “woodbasket”. [Of course, Sirex noctiliois already established in the eastern United States. Although it is a severe pest of loblolly growing in plantations in the Southern Hemisphere, industry and federal and state agencies have dismissed concerns in North America.] The Pine Pandemic Preparedness Plan has four components: communication, detection and diagnosis, delimitation and assessment, response.
As concern about brown spot needle blight grew, the Southern Group of State Foresters ask the “P4” team to engage. Klepzig and Kamal Gandhi pulled together a working group which has the goal of developing guidance for managing the disease within two to five years. The task force is developing a website for data-sharing. The task force is also studying genetics of the host and pathogen, fungicides, the role of fire, resistance screening, and spore trapping. Industrial concerns about coordinating with competitors cause challenges.
Ashley Schulz, of Mississippi State University, has reviewed experience with biocontrol for clues on species’ traits important for facilitating invasion. She analyzed information on 394 insects introduced to North America for biocontrol of invasive plant species (see other blog) and 87 agents targeting 325 insect pests. For each species, data was recorded on whether it established, level of impact, the insect’s feeding guild, climate matching, host specialization, and evolutionary history. For the 87 entomophagous insects, she also recorded host feeding guild and host specialization.
Schulz found that entomophagous insects introduced as biocontrol agents were more likely to establish if they are a specialist. Higher impact was also associated with specialization. Parasitoids had higher impacts than predators. What does this indicate re: invasive species? Schulz said that insects which can hide or defend themselves, i.e., specialists, are likely to be more successful invaders.
Schulz recommends more analysis of what can be learned from experience with biocontrol agents. However, such studies are challenged by poor records, lack of empirical evidence and quantitative data, the lower number of biocontrol agents introduced recently, and funding shortages that preclude post-release monitoring.
Schulz also mentioned that she worries that a proposal to drop the word “harm” from definition of invasiveness could result in biocontrol agents being lumped with invasive species. This would further hamper implementation of biocontrol. She considered this loss to have particularly bad affects at a time when there are growing restrictions on pesticide use.
SOURCES
Mason, M.E., Carey, D.W., Romero-Severson, J. et al. Select genotypes of white and green ash show heritable, elevated resistance to emerald ash borer. New Forests57, 12 (2026). https://doi.org/10.1007/s11056-025-10158-x
Merkle, S.A., J.L. Koch, A.R. Tull, J.E. Dassow, D.W. Carey, B.F. Barnes, M.W.M. Richins, P.M. Montello, K.R. Eidle, L.T. House, D.A. Herms, K.J.K. Gandhi. 2022. Application of somatic embryogenesis for development of emerald ash borer-resistant white ash and green ash varietals. New Forests https://doi.org/10.1007/s11056-022-09903-3
Nardi, D., D. Rassati, A. Battisti, M. Branco, C. Courtin, M. Faccoli, N. Feddern, et al. 2026. “Integrating Landscape Ecology into Generic Surveillance Plans for Bark- and Wood-Boring Beetles.” Ecological Applications 36(2): e70194. https://doi.org/10.1002/eap.70194
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
Callery/Bradford pear invasion in northern Virginia; photo by F.T. Campbell
The US Department of Agriculture (USDA) and the North American Invasive Species Management Association (NAISMA) held the 34th annual forum on invasive species research at the end of February 2026. The agenda is available here In this blog I summarize the presentations about invasive alien plants (IAS); a separate blog discusses findings on tree-killing pests. Formal proceedings will be available in some months.
The most important information from the meeting:
If NAISMA had not taken on the task of hosting the conference it would not have happened.
Government leaders allowed only 1 staffer per USDA Forest Service region to participate. Not allowed to come were people who had organized the whole meeting or individual sessions, and presenters discussing several topics, including preventing IAS plant spread, and progress on controlling cogongrass (major impediment to pine plantations, affecting harvests).
What do these decisions say about the genuineness of the USDA Secretary’s recent memorandum listing invasive species as one of four priority areas for the department’s research efforts?
A reminder to us all: Rebekah Wallace of the Center for Invasive Species & Ecosystem Health at the University of Georgia urged us all to provide citations for images used in informal materials – posters, presentations, outreach efforts, blogs, videos. Images grab attention, provide context for communication, and support data cited. Providing the citation increases our credibility and ensures that we avoid perpetuating misinformation!
Callery/Bradford pear in Kentucky; photo by Sherry Bailey via NARA archive
Two presentations focused on Callery / Bradford pear
Jess Hartshorn of ecoLogic described efforts to develop a remote sensing tool that will be as accurate as human surveyers — but faster. What scientists learned from this exercise will help build tools for other invasive plants. Hartshorn noted that while there are many no-cost sources of satellite imagery, no single source is sufficient. But integrating data from several programs, plus adding new criteria proved challenging. One setback was a surprise: the spectrum emitted from the tree’s most conspicuous feature, its early-season white blooms, is similar to that reflected from concrete! – with which the species is associated … The authors had to use data from several satellite systems to identify unique wavelengths from the leaves. Accuracy was lost when an individual pixil contain mixed “vegetation”.
Marcin Nowicki, of the University of Tennessee, explored the genetic changes that allowed a species that is rare in Asia to become a prolific continent-wide invader in North America. “Evolutionary overdrive” resulted from planting plants from several origins close together, thus promoting cross pollination. This led to exceptionally rapid diversification in nuclear and mitochondrial DNA. A bonus: once Sequencing the genomes of several cultivars have been sequenced, bans on sales of those hybrids that are most invasive can be enforced.
Becky K. Kerns, USFS Pacific Northwest Research Station reported on disturbing increases in invasive plants in forests of the Pacific Northwest. In the past, higher elevations, low light levels, and cooler temperatures appeared to protect the region’s forests from invasion. However, annual grasses, especially cheat grass (Bromus tectrorum), are now being found at unprecedented levels in forest plots that have been burned, grazed, or logged, burned, and grazed. This includes plots subjected to prescribed burns. Kern thinks the plant invasions are due to increased light, ground disturbance, changed competitive interactions, and potentially higher propagule pressure. Pyrophytic shrubs also of increasing concern; Kerns mentioned Scotch broom (Cytisus scoparius) in Douglas-fir forests. [I am uncertain how novel this threat is because academic scientists issued warnings about Scotch and other brooms in the mid-1990s.] [run together w/ following] She is working with the staff of the National Invasive Species Council’s task force on fire and invasives to increase attention to emerging threats and to encourage managers to prioritize managing known pyrophytic species along with fire.
Wavyleaf basketgrass infestation in closed-canopy forest in Maryland; photo by Kerry Kyde, Maryland DNR via Bugwood
Two speakers addressed aspects of the invasion by wavyleaf basket grass (Oplismenus hirtellus subsp. undulatifolius).
Wavyleaf basket grass was first detected in 1996 in Maryland. It is now widespread in the Mid-Atlantic and expected to spread along the Appalachian Trail and to other recreation sites. Thirty percent of public land in the East is considered vulnerable.
Carrie Wu of the University of Richmond is exploring the grass’ association with changes in the soil microbial community. She tested associated soil microbial communities in 12 locations with three types of soil. She found decreased fungal diversity but not homogenization of the fungal community. She is now constructing an invasion history to see how fast the changes occur, confirm the invaded range, and predict high-risk sites.
Michael Fulcher, of the USDA Agriculture Research Service’s Foreign Disease-Weed Science Lab, is concerned about the microbes associated with invasive plant species. We don’t know whether some of these microbes might be beneficial, perhaps as biocontrol agents? Or might they cause disease in desired plant species. He phenotyped 319 isolates from healthy leaves. This study detected two known crop pathogens on healthy wavy leaf basket grass plus an unknown species in a genus that includes some known pathogens. In lab tests, this organism stunted growth of wheat and tall fescue embryos
Fulcher emphasizes that even asymptomatic non-native plants can transport possible pathogens. Scientists should try to detect and analyze these as quickly as possible. I note that Eliana Torres Bedoya reported last year that healthy woody plants can also transport disease-causing fungi.
Fulcher is looking for collaborators to help collect plant samples
Other invading plants
Craig Barrett of West Virginia University seeks to answer questions related to “invasiveness” traits and whether selective pressures enhance those traits in the invasive range. To explore these topics, Barrett is mapping the invasion history of the widespread invasive species Japanese stiltgrass (Microstegium vimineum). He has found evidence of the grass’ rapid adaptation after introduction, including greater diversity in invasive populations in the Northeast than those in the Southeast. Barrett thinks it most likely that a genetic bottleneck at introduction was followed by mixing that created novel genotypes that might bridge gene transfer between larger populations. There is evidence of phenological adaptation to local climates and a genetic basis for whether a plant supports awns – which react to changes in moisture by “walking” across soil and burying themselves.
Elizabeth Ward, at the Connecticut Agriculture Experiment Station, documented how invasive plant species utilize forest gaps created by the death of ash caused by emerald ash borer (EAB). The progress of the EAB infestation across Connecticut is well-documented, so scientists can track plant responses to stages of canopy mortality. She found:
Larger canopy gaps contained more invasive plants and fewer native tree seedlings / reduced regeneration.
Higher soil nitrogen availability is also linked to higher non-native plant cover (all species) – including non-native tree seedlings.
Higher carbon availability led to lower non-native plant cover, including that of non-native tree seedlings.
Ward advises active management of EAB-invaded forests to reduce plant invasions and promote tree regeneration.
Ward is now comparing sites with passive management vs. salvage harvests. Early results find no difference in invasive plant cover. However, harvested sites had higher abundance of ash regeneration and and diversity of native plant species.
Jeremy Anderson, at the University of Massachusetts, discussed difficulties that have slowed the search for a biocontrol agent to control invasive knotweeds. North American scientists are collaborating with counterparts in Europe. Because knotweeds are related to rhubarb, scientists must ensure that any agent is host specific.
knotweed infestation in Maryland; photo by Will Parson, Chesapeake Bay Program
Initial surveys 20 years ago identified 180 candidate insects. However, the only speciesfound suitable for in- depth evaluation failed to establish. Why? First, there was apparently a climate mismatch: the insect is from southern Japan but the plant is from the North. Then a second difficulty was discovered: the target weeds are hybrids, not a pure species. Scientists are now testing a microbe that might overwinter on pine needles, so they are comparing needle chemistries of Japanese red pine with those of North American pines to determine whether there is a risk. In answer to a question, Anderson said scientists do not know how the microbe will respond to the warmer, wetter climate expected in New England in the future.
Ashley Schulz, of Mississippi State University, is continuing her efforts to identify clues to which newly introduced species might be most damaging. In this case she is analyzing efficacy of biocontrol agents to understand which establish and have significant impacts. Species with traits similar to successful biocontrol agents might be more successful invaders.
Schulz analyzed information from 394 insects introduced to North America to control 153 plant species and 87 agents targeting 325 insect pests. The data recorded on each species: whether it established, level of impact, insect’s feeding guild, climate matching, host specialization, and evolutionary history. For the 87 entomophagous insects, she also recorded host feeding guild and host specialization. See other blog.
Phytophagous insect biocontrol agents were more likely to establish if the insect is a generalist newly associated with the target plant species. The biocontrol agent is more likely to have a greater impact when released in environments similar to the agent’s native range. The introduced biocontrol agent will have less impact if it feeds on plant parts that the plant can easily restore (foliage, fruit/seeds).
What does this indicate re: invasive species? Schulz concluded that among phytophagous insects, generalists might be more likely to find a suitable host and survive. The “Goldilocks” premise applies: the host is sufficiently similar to the invader’s native host that it is recognizable but sufficiently distantly related to lack defenses effective against the invader. Bioinvasive phytophagous insects will have a greater impact when introduced to a similar climate and feeds on plant structures that are not easily restored – i.e., stem, root.
For traits of entomophagous insect biocontrol agents see my other blog here.
Schulz recommends more analysis of what can be learned from experience with biocontrol agents. However, such studies are challenged by poor records, lack of empirical evidence and quantitative data, the lower number of biocontrol agents introduced recently, and funding shortages that preclude post-release monitoring.
Schulz also mentioned that she worries that a proposal to drop the word “harm” from definition of invasiveness could result in biocontrol agents being lumped with invasive species. This would further hamper implementation of biocontrol. She considered this loss to have particularly bad affects at a time when there are growing restrictions on pesticide use.
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
As bioinvasions and their impacts continue to expand globally, managers and decision-makers charged with developing effective management and mitigation strategies urgently need tools that can assess and rank all impacts. These start with impacts on species’ populations … but go much farther, to the assemblage, ecosystem, and abiotic levels. Impacts at the “species and assemblage” level include species extinction (locally or more broadly), changes in species range, assemblage structure, successional patterns, and the soundscape. Impacts at the “ecosystem function” and “abiotic” levels include changes to primary production, food webs, water quality, and nutrient cycles. The analysis also addresses changes that do not affect native biota directly, although they present no examples.
For a decade, scientists studying bioinvasions have used the Environmental Impact Classification for Alien Taxa (EICAT) framework to standardize categorization of species-level impacts. One group that has not used this methodology is experts on tree pests. Why? Does the approach fail to describe the impacts of non-native arthropods and pathogens on tree species and forest ecosystems more broadly? Or is it simply because of academic silos?
Even more important: are the science and practical management of invasive species and forest pests losing valuable insights, resources, policy choices, … because of this schism? Would both groups gain from closer interactions?
In any case, the framework used by many scientists working on “invasive species” is undergoing a revision to better capture cascading and systemic effects from bioinvasion. A group of scientists has created the Extended EICAT (EEICAT) framework. (See the publication reference at the end of this blog to learn the process of development and details of the new system.) The proponents claim that the new system recognizes the functional interdependence of species in ecosystems, which means that alterations in species assemblages inevitably amplify throughout the system. E.g., alterations in physico-chemical characteristics or habitat structure. Impacts can even cross-ecosystem impacts between ecosystems that are often managed separately. An example is a change in the quality, magnitude, and novelty of resource flows between terrestrial and aquatic systems. To address these multifaceted effects, EEICAT integrates 19 impact types into the analysis. The intention is to improve communication about the complex ecological impacts caused by bioinvasions and facilitate prioritization of responses to competing bioinvasions.
While the various outcomes from bioinvasion can be positive or negative for nature and people, the EEICAT does not use value-laden distinctions. These determinations are left to stakeholders, managers, and community members, based on their own perspectives. Instead, it compiles and standardizes information about the measurable changes to species numbers (some decrease, others increase); to ecosystem processes (e.g., nutrient dynamics or hydrological regimes).
EEICAT incorporates the “reversibility concept”, which addresses the potential for a native sp (including individuals, pops, and assemblages), ecosystem function, or abiotic environmental to recover after removal of the bioinvader. The system developers distinguish “naturally reversible changes” and “naturally irreversible changes”. In the former case, the affected spp, ecosystem processes or abiotic conditions are thought likely to return to their original state within 10 years or three generations (whichever is longer) through natural processes or human-assisted actions that do not exceed what is already being done. This does not include reintroductions or restoration efforts that require new efforts. Instances of “naturally irreversible changes” are those in which the affected species, ecosystem functions, or abiotic conditions cannot return to their original state within that timeframe without significant additional human intervention, or even after intense human intervention. The system has reached a different, stable equilibrium. These “permanent” changes are the result of one or more species’ global extinction, or persistent environmental alterations, e.g., soil modification, altered hydrology, or irreversible changes in nutrient cycling.
The proponents assert that EEICAT allows multiple impacts reported in a single study to be classified independently at each impact level. Furthermore, the EEICAT analysis does not require extensive research on the assessed species or understanding of the mechanisms through which the invasive species affects native species or the environment. EEICAT framework is applicable to any amount of info available in each study. It also explicitly assesses the adequacy / reliability of evidence [data, methods, approach] used in studies of bioinvasions that are included in the analysis.
EEICAT framework enables researchers to evaluate how “ecosystem engineer” species influence key ecological functions by explicitly accounting for changes to ecosystem processes, e.g., nutrient dynamics or hydrological regimes. For example introduced bivalves increase water clarity in certain systems, triggering cascading effects on biodiversity and ecosystem functions.
The EEICAT framework also allows separation of the mechanisms of impact vs. attribution of impact. For example, when a non-native plant species alters nutrient availability, thereby changing the microbial community, EEICAT assigns separate impact categories to the two impacts.
Regarding cross-ecosystem effects, the proponents cite rats on islands. Their predation suppresses seabird pops; reduced guano alters the nutrient dynamics of adjacent coral reef ecosystems. Thus assign impact categories not only to the changes in nutrients, but also to ecological functioning. This provides a more comprehensive view of interconnected effects.
Proponents of the proposed new framework assert that the fundamental distinction between EEICAT and the earlier EICAT is that the earlier assessment is “species-based”, whereas the new one is “impact-based”. It is broader because it focuses on specific combinations of invading species plus the affected systems. It is better able, they assert, to account for contrasting impacts in different invasions.
EEICAT can be applied to any invasion event (i.e., a specific combination of invasive species, recipient system, and context). It broadens the range of evidence that can be integrated into the assessment. Decision-makers benefit from access to more information. The information can also be provided in more easily understood form through two visualization tools:
An “invasive species profile” aggregates all recorded impacts caused by a single invading species. This facilitates clear communication of the bioinvasion’s impact severity to managers and stakeholders, plus how those impacts vary by context.
An “invaded ecosystem profile” compiles impacts from different species to a site or location. This is particularly useful for synthetic analyses (e.g., meta-analyses), evidence syntheses, and manager assessments.
Resulting profiles can help stakeholders prioritize species or ecosystems for responses.
https://www.dontmovefirewood.org/pest_pathogen/phytophthora-root-rot-html/to are ants. No disease agent is discussed or even named. This gap is surprising given the devastating and geographically extensive impacts of e.g., avian malaria, chitrid fungi (Batrachochytrium dendrobatidis and Batrachochytrium salamandrivorans) on amphibians, and Phytophthora cinnamomi on the flora of western Australia.
One example in Table 3 pertains to native Hawaiian forests. The underlying study analyzed changes in ecosystem functions caused by the invasive nitrogen-fixing tree Falcataria moluccana. The EEICAT proponents say their analysis of this study would supports more informed decisions in conservation planning and ecosystem management. Indeed, the principal author of the underlying study has recently published a suggested method to manage the Falcataria moluccana invasions by replacing these trees with either native species or valued crops under an agroforestry program. Neither of the articles mentions that exactly this same area (the Puna District on the “Big Island) has suffered widespread death of the native tree ʻōhiʻa lehua (Metrosideros polymorpha) as a result of the invasive disease rapid ʻōhiʻa death (ROD). The more recent article does address the fact that native plant species are extremely rare in this region.
Would integrating studies of tree-killing arthropods and pathogens into the EEICAT system provide benefits? First, let’s consider analytical methodology. Many analyses of forest pests’ impacts already discuss at least some of the wider ecological (and economic) outcomes. (To explor this, visit www.dontmovefirewood.org and read some of the species profiles under the “invasive species” tab.) Would comparing these findings to an EEICAT analysis confirm the proposed methodology? Or would it instead suggest needed adaptations? In either case, the results should improve scientists’ work.
Second, would the science and practice of managing invasive species be strengthened by bridging the differences in methods and terminology between those focused on plants and vertebrates and those focused on tree-killing invertebrates and microbes? Would greater unity result in more attention to bioinvaders from policy-makers and/or conservation practitioners and advocates? Especially since (nearly) all the major forest pest invasions would qualify as “naturally irreversible changes” or even “permanent”: the affected species, ecosystem processes or abiotic conditions are thought unlikely to return to their original state within 10 years or 3 generations (whichever is longer) in the absence of intense human-assisted actions. If joining forces might bring about greater societal efforts, is the EEICAT methodology a promising tool to achieve this goal?
Finally, would applying the EEICAT system improve the analyses of tree-pest impacts? Would this approach result in incorporation of types of effects that would otherwise be missed – either often or in specific cases? Are there relationships among forest species, or between species and ecological functions, that might be discovered? Might preparation of “invaded ecosystem profiles” that include bioinvaders from earthworms to canopy foliage feeders provide an informative perspectives that is now lacking?
SOURCE
Carneiro, L., Pincheira-Donoso, D., Leroy, B., Bertolino, S., Camacho-Cervantes, M., Cuthbert, R.N., et al. (2026) Expanding invasive species impact assessments to the ecosystem level with EEICAT. PLoS Biol 24(3): e3003665. https://doi.org/10.1371/journal.pbio.3003665
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
SOD-infected rhododendron in a nursery; photo by Jennifer Parke, ODF
A group of scientists (See Khusnitdinova et al., 2026; full reference at the end of this blog.) contend that landscape interfaces—e.g., crop–forest edges, riparian zones, abandoned agricultural fields and orchards, and nursery–wildland transitions—are activezones of pathogen exchange. Biological and abiotic vectors collectively move pathogens from crops to wild plants, and vice versa. These exchanges create conditions speed up the evolution of pathogen aggressiveness and dispersal traits and promote the selection of generalist pathogen lineages capable of infecting both cultivated and wild hosts. In this way, crop-natural ecotones become not just passive transition zones but centers of adaptation.
The stronger or novel pathogens don’t stay in the specific local area; they are spread by a variety of human activities. Establishing large monocultures of crops and simplifying biological diversity at the landscape level boost inoculum production, limit host genetic diversity, and diminish natural regulation. Pathogens present in irrigation water can be spread during floods. Improperly sanitized green waste and compost can harbor viable oomycete propagules. Foot traffic and heavy equipment can move contaminated soil. Movement of infested plants for planting can transport the disease to a different continent. One example cited by Khusnitdinova et al. (2026) is the spread of numerous Phytophthora spp. from nurseries to forests and shrublands. A second example is rapid ʻōhiʻa death. They say it demonstrates that 1) a combination of human movement, forestry activities, and animal vectors can enable rapid local and landscape-scale spread; and 2) management measures (biobarriers, access control, restriction of animal movements, and phytosanitary inspection of planting material) can curtail that spread.
Meanwhile, the changing climate is causing shifts in the latitudinal and elevational distribution of plants and their associates; changing reproduction rates and latent periods; altering ranges and connectivity; and affecting disease incidence and severity. The direction is not always predictable; while drought or heat might reduce fungal and oomycete epidemics, the same conditions increase host stress and so might worsen disease outcomes.
Plant health scientists can use these concentrated geographic areas to focus plant disease surveillance. By integrating molecular and genomic tools with remote sensing and Geographic Information System (GIS)-based monitoring, plant health agencies can more quickly detect newly emerging diseases and implement effective action to counter the threat.
However, Khusnitdinova et al. (2026) warn that surveillance employing these technological advances can reduce the risk that a pathogen will “spill over” from an anthropogenic to a natural ecosystem or vice versaonly if pertinent sectors are transformed. Yes, they need resources: funding, staff, facilities. Also required is unification – or at least coordination. Khusnitdinova et al. (2026) advocate abandoning the compartmentalization that currently separatesforest health studies from invasive-plant and infectious-disease ecology studies. Instead, agencies should consider managed and natural systems together. They should conduct joint surveillance programs, share data standards, and coordinate management of the transition zones. In other words, apply a “One Health” landscape-based approach to the entire landscape.
Khusnitdinova et al. (2026) add that implementing such combined surveillance programs is especially vital in biodiversity-rich regions which have limited monitoring capacity. Might I suggest Hawai’i?
ohia trees killed by ROD; photo by J.B. Friday, UH
Other facts that challenge traditional phytosanitary practices
Khusnitdinova et al. (2026) provide strong evidence that pathogens change – sometimes quickly. Is the current regulatory system sufficiently flexible and agile to effectively address these developments?
First, pathogens’ host range is not fixed. Instead, it is a trait that changes quickly under the influence of alterations in effector repertoires, plant immunity genes, and environmental conditions (including those driven by human actions). Even small genetic changes—such as mutations, gene losses or gains, or horizontal gene transfers—can enable pathogens to infect new hosts or weaken previous infection barriers. They suggest that plant pathogens with broad host ranges, e.g., Phytophthora cinnamomi, can easily move between hosts in agricultural plantings, ornamental landscapes, and semi-natural vegetation within a relatively small region. Such frequent spillovers maintain inoculum in landscape mosaics and complicating eradication or containment efforts.
Khusnitdinova et al. (2026) note that host-range expansions have especially long-term consequence in forest ecosystems, where loss of a single tree species can change understory makeup, light and moisture patterns, related fungi and invertebrate communities, and ultimately, landscape diversity and function. They cite chestnut blight and sudden oak death in North America and ash dieback in Europe as examples.
In addition, Khusnitdinova et al. (2026) maintain that genetic recombination is now recognized as a fundamental driver of innovation in plant pathogen populations. Table 2 of their publication lists pathogens exhibiting well-documented and experimentally confirmed cases of recombination, hybridization, or other forms of genome exchange. Forest-related examples include several Phytophthora hybrids and the ash decline fungus, Hymenoscyphus fraxineus.
Phytophthora dieback in Western Australia
Khusnitdinova et al. (2026) add their voices to a growing chorus decrying a global forest health crisis. They say that repeated pathogen introductions—often via trade in plants and wood—have shifted many temperate and boreal forests into states characterized by higher tree mortality, increased dominance of opportunistic or disturbance-adapted species, and reduced functional diversity. These changes lead to reduced resistance [defined as the capacity to limit damage during a new outbreak] and resilience [defined as the speed and trajectory of post-disturbance regeneration and ecosystem reorganization]. They note that increasing tree species diversity is one of the few management interventions that succeeds in strengthening both forest resistance and resilience to pathogens—by decreasing host density for specialist pathogens and reducing continuous “fuel” for epidemics.
One step toward improving scientific understanding on the scale they advocate, in their view, is the European Holistic Management of Emerging Forest Pests and Diseases (HOMED) effort. HOMED combines plant pathology, forest ecology, and biosecurity. The emphasis is on early detection, risk assessment, and management of human-mediated pathways, incl plant trade and nursery systems. The initiative aims to limit pathogen establishment and spread while strengthening forest resistance and resilience under global change. Participants also try to provide practical solutions for stakeholders to manage emerging native and non-native pests and pathogens threatening European trees not only in forests, but also in nurseries, urban and rural areas.
USDA Secretary Brooke Rollins
I am inspired by the proposals in Khusnitdinova et al. (2026). In hopes that USDA will explore how to implement them, I presented a poster presentation at the annual USDA Research Forum on Invasive Species. In that poster I suggested that these ideas complement USDA Secretary Rollins’ Memorandum on departmental research priorities. The need for research to clarify scientific puzzles is particularly acute regarding tree-killing pathogens nematodes, etc.
I suggested prioritizing research on the following issues:
Setting up intensive monitoring programs targetting the agriculture/natural system interfaces, as recommended by Khusnitdinova et al. (2025). These authors describe useful technologies in molecular diagnostics, genomic surveillance, environmental DNA, and remote sensing to detect fungi, oomycetes, rusts, bacteria, and viruses. Kantor et al. (2025) define techniques applicable for nematodes.
Rapid analysis of potentially invasive species and their pathways of entry revealed by “early warning” systems [e.g., APHIS’ “PestLens” website; “door knocker” introductions; academic studies; and “unimportant” species introduced to the U.S. (e.g., Leptosillia pistaciae in California)].
Exploring ways (in addition to those suggested by Khusnitdinova et al. 2025) to shorten the time lag between introduction of a pathogen and its detection.
Incorporating into risk analyses information from sentinel garden program. Fund expansion of data collection and analysis to address asymptomatic plants, sampling techniques, and seasonality, as outlined by Drs. Eliana Torres Bedoya and Enrico Bonello (at the 2025 USDA Research Forum) and Raffa et al. (2023).
Over a somewhat longer-term, I suggested that research address these topics:
Find techniques to speed up determination of disease causal agents – which often remain obscure for years or decades. The International Plant Protection Convention (IPPC) link requires countries to name the causal agent before regulating disease hosts and vectors.
Determine which components of a “systems approach” are most effective against each type of pathogen – fungi, oomycetes, rusts, bacteria, viruses, nematodes, etc.
With state counterparts, explore ways to better curtail domestic spread of organisms once they have established in the United States.
Integrate socio-economic drivers of pest introductions into studies. E.g., why do some organisms suddenly spread to numerous countries over a period of a few years?
Greatly expand efforts (in house and by collaborators) to breed trees resistant to established and newly detected pathogens.
Increase research supporting biocontrol.
As I have frequently complained in the past, the international phytosanitary system has failed to protect Earth’s forests and other natural ecosystems from non-native plant pests (or invasive plants). This failure has been documented by Weed, Ayres, and Hicke (2013), Fei et al. (2019), Quirion et al. (2021) for North America; and Gougherty (2023), Wu (2023), Sitzia et al. (2021), Martinac et al. (2025) and Khusnitdinova et al. (2025) from a global perspective.
Challenges:
Most microorganisms are unknown to science – “unknown unknowns”.
Scientists usually cannot predict the impact of known micro-organisms on new hosts under novel environmental conditions.
The World Trade Organization’s SPS Agreement and the International Plant Protection Convention (IPPC) demand unachievable levels of specificity re: a potential pest’s impact.
Most tree-killing pathogens are detected after they have entered the forest.
Agencies assign a low priority to protecting natural ecosystems from bioinvasion.
Resources (funds, staffing, etc.) are unreliable for agencies carrying out the full range of efforts, from assessing various risks to restoring pest-resistant trees to the forest.
SOURCES
Fei, S., R.S. Morin, C.M. Oswalt, & A.M. 2019. Biomass losses resulting from insect & disease invasions in United States forests
Gougherty, A.V. (2023) Emerging tree diseases are accumulating rapidly in the native & non-native ranges of Holarctic trees. NeoBiota 87: 143–160. https://doi.org/10.3897/neobiota.87.103525
Kantor, C., Teixeira, M., Kantor, M., and Gleason, C. 2025. Tiny Invaders, Big Trouble: Emerging Nematode Threats in the United States. Phytopathology 2025 115:587-595 https://doi.org/10.1094/PHYTO-09.-24-0290-IA
Khusnitdinova, M., V. Kostyukov, G. Nizamdinova, A. Pozharskiy, Y. Kydyrbayev and D. Gritsenko. 2026. Cross-Ecosystem Transmission of Pathogens from Crops to Natural Vegetation. Forests 2026, 17, 76
Martinac, M-L., F. Ningre, A. Dowkiw, N.Le Goff, B. Marcais. 2025. High host density favour ash dieback Preprint Plant Pathology
Quirion BR, Domke GM, Walters BF, Lovett GM, Fargione JE, Greenwood L, Serbesoff-King K, Randall JM & Fei S (2021) Insect and Disease Disturbances Correlate With Reduced Carbon Sequestration in Forests of the Contiguous United States. Front. For. Glob. Change 4:716582. [Volume 4 | Article 716582] doi: 10.3389/ffgc.2021.716582
Sitzia, T., T. Campagnaro, G. Brundu, M. Faccoli, A. Santini & B.L. Webber. 2021. Routledge Handbook of Biosecurity & invasive species. Chapter 7. Forest Ecosystems. ISBN 9780367763213
Weed, A.S., M.P. Ayers, J.A. Hicke. 2013. Consequences of CC for biotic disturbances in North American forests. Ecological Monographs, 83(4), 2013, pp. 441–470
Wu, H. 2023/24. Modelling Tree Mortality Caused by Ash Dieback in a Changing World: A Complexity-based Approach MSc/MPhil Dissertation Submitted August 12, 2024. School of Geography & the Enviro, Oxford University
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
Falcataria moluccana tree; photo by Forest & Kim Starr via Flickr
Nitrogen-fixing tree species have been recognized as damaging to invaded ecosystems for decades. These trees increase soil N availability through increased N content in litterfall. The elevated soil N availability might persist long after the mature individuals responsible for creating such litterfall have ceased to exist. When this happens, some plant species able to exploit increases in nutrients and light, e.g., non-native grasses and forbs, might quickly dominate post-control succession.
In Hawai`i one of the worst nitrogen-fixing tree species is albizia (Falcataria falcata) [formerly Falcataria moluccana, Paraserianthes falcataria, or Albizia falcataria]. This fast-growing species has aggressively invaded across the archipelago, transforming composition, structure, and function of remnant lowland wet forests. There are an estimated four million F. falcata trees across the Hawaiian islands; 720,000 large trees (i.e., > 25 cm DBH). The trees spread rapidly once established because the small seeds remain attached to the lighweight pods, which can be blown for long distances in wind storms (J.B. Friday, University of Hawaii, pers. comm.).
Stands with contiguous overstory F. falcata canopies reduce light availability to 20% of ambient levels; adding in understory vegetation further reduces light to ~5% of ambient levels. Albizia’s abundant and persistent seedbank promotes its return to dominance after mature individuals controlled.
understory of an albizia-invaded area; invasive plants: forbs along roadside; Miconia calvescens in the shade. Photo by F.T. Campbell
Beyond the conservation threats, albizia also poses a threat to residential communities & agricultural lands. The trees are some of the fastest growing species in the world, easily growing 5 m in height annually over the first few years and reaching up to 40 m. When their brittle branches fall they crush structures and entire trees can topple during windstorms. The damage is exacerbated by trees’ widespread presence. When Tropical Storm Iselle hit Hawai‘i island in 2014, over 10,000 people were stuck in their subdivisions or on their farms because fallen albizia had blocked all their access roads (Friday, pers. comm.).
Until recently control efforts have relied largely on clearing the land using large machinery (e.g., bulldozers). This is expensive and – worse – not very effective because the magnitude of disturbance to the soil disturbance often leads to explosive germination of the trees’ seeds.
There has been success recently through application of a target-specific herbicide (aminopyralid) at low doses (Leary et al. 2014). Hughes et al. (2025) found that herbicide-killed F. falcata quickly lost their leaves. This litterfall increased litter inputs of N and P that translated to increased soil nutrient availability that is exploited by extant understory vegetation (non-native grasses and forbs). These plants formed a continuous layer that severely limited germination of F. falcata seeds. In their study plots the number of saplings per ha after three years was only 18, despite the presence of perhaps 8 million seeds!
As an early successional pioneer species, F. falcata requires high light conditions to germinate, persist, & grow. The rapid growth & thorough occupation of the understory by other species prevents the species’ re-establishment. However, these aggressive non-native plants also prevent restoration of native Hawaiian species. There is little to no regeneration of native plants under albizia, either on stands that established on abandoned agricultural or ranch lands or under trees that spread into native forests.
Hughes et al. (2025) suggest manipulating the succession trajectory by planting desired species – either native species or species that have cultural importance to native Hawaiians – under albizia stands before herbicide treatment. If the land is to be restored to agricultural use, mechanical clearing would be used rather than herbicide used as felling the brittle dead trees is hazardous to equipment operators, and standing dead trees would pose a risk to farmers. In a forest setting, understory planting before herbicide treatment of the canopy-forming F. falcata stands would allow desired species to take maximum advantage of the increased resources (i.e., light and nutrients) (Friday pers. comm.).
Even after invasive N-fixing trees have been physically removed, the soil legacy effects of transformed microbial communities, depleted native seedbanks, increased available soil N, and dominance by undesirable weed species are daunting barriers to restoration of native species. With intensive management, though, these lands can be restored to agricultural production. Dozens of acres of papaya farms have been established on areas in the Puna district of Hawai‘i island on lands formerly occupied by albizia (Friday, pers. comm.).
In this case, re-establishment by native species is not expected due to their scarcity in study areas. These areas had experienced significant disturbance (i.e., fire, and/or conversion to agriculture) before albiziast and establishment. Instead, the proposal’s objective is primarily to understand whether, how, and to what extent F. falcata stands could be eliminated from areas in a manner that constrains the species’ seedling recruitment and subsequent re-establishment leading to overstory dominance once again (Friday, pers. comm.).
Hughes et al. (2025) emphasize the need for long-term follow-up to ensure that F. falcata does not re-establish later on. The species’seeds retain 70 – 90% viability following 18 months in storage; possibly some much longer. Also, a few saplings did still establish. The non-native grass invasion might lead to declines in soil N availability that provide opportunities for secondary invasion by N2-fixing treesin light gaps. Dr. Friday reports that practitioners revisit treated areas to kill these seedling while they are still 10 – 20 feet tall.
Conclusions
Hughes et al. (2025) assert that management of this large, fast-growing, & disruptive invasive tree is possible by exploiting its weakness of shade intolerance. Dr. Friday agrees that fast-growing timber species, e.g., Eucalyptus, could outcompete regenerating albizia. However, will there be a market for locally grown timber? Dr. Friday doubts the possibility of agro-forestry plantings of smaller or slower-growing species because of the danger that the overtopping dead F. falcate would fall on and crush agricultural workers or structures.
The fall hazard would presumably apply in other parts of the Pacific & elsewhere where F. faclata poses the same invasiveness problems.
ʻōhiʻa trees killed by ROD in the Puna District of Hawai`i Island; photo by F.T. Campbell
Hughes et al. (2025) do not mention that the native tree that was probably most widespread before the disturbances is ʻōhiʻa lehua (Metrosideros polymorpha). In precisely the same lowland region of the Big Island where they conducted their study, ʻōhiʻa has been killed by a newly introduced disease, rapid ʻōhiʻa rust (ROD). This new invader greatly complicates any effort aimed at restoring native plant species.
healthy ʻōhiʻa in Hawaii Volcanoes National Park; photo by F.T. Campbell
SOURCES
Hughes, R.F., C. Morrison, E. Bufil, J. Leary. 2025. Ecosystem response to management of an invasive N-fixing tree in Hawai`i. Trees, Forests and People 21 (2025) 100932
Leary, J., J. B. Friday, S. Kaye, and F. Hughes. 2014. Proper technique of injecting albizia (Falcataria moluccana L.) with the herbicide Milestone ® (active ingredient aminopyralid).
Dr. Friday provided the following more local references:
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
