Horizon Scanning – 2 experiences

sorting coffee beans; photo by Niels Van Iperen via Wikimedia

Many have recognized that preventing introduction of invasive species is the most efficient approach to minimizing their ecological and economic impacts. Prevention requires many capacities, including control over a country’s borders, strong border biosecurity agencies and policies, and foreknowledge of probable pathways of introduction and high-impact species that might arrive.

Horizon scanning is one tool for gathering information about non-native species likely to enter, how they might arrive, and their probable impact. Horizon scanning involves a systematic search for potential invaders, assessment of their potential to harm BD, economic activities and human health, and opportunities for impact mitigation. It thus supports choice of prevention policies, targetting of efforts, and implementation of early identification and eradication procedures (Kenis et al. 2022; Martinou et al. 2026)

I have reviewed two case studies of the application of horizon scans.

Plant Pests in Ghana

One horizon scanning exercise aimed to identify and rank potential invasive non-native plant pest species that could be harmful to agriculture, forestry, and the environment in Ghana. The ultimate objective was to enable prioritization of actions aimed at preventing their introduction. As the participants in this exercise note (Kenis et al. 2022), the resource-poor farmers of Sub-Saharan Africa are particularly vulnerable to invasive pests that attack their crops, both those grown for subsistence e.g., maize and sorghum, and those grown for the international market, e.g., cacao and tomatoes. The continent’s vulnerability is increased by porous borders, weak cross border biosecurity, and inadequate capacity to limit or stop invasions. This exposes Africa both to repeated invasions and to continued spread across the continent once they have arrived.

Marc Kenis and 21 others assessed 110 arthropod and 64 pathogenic species using a simplified pest risk assessment. This set had been winnowed from an initial list of 1486 arthropods, nematodes and pathogens. Unfortunately, assessors were unable to agree on confidence levels for the assessments.

Sixteen of the assessed species – 14 arthropods and two pathogens – were thought at the time to not be on the African continent. Another 19 arthropod and 46 pathogenic species had been reported established in the neighboring countries of Burkina Faso, Côte d’Ivoire, and Togo. Seventy-seven species [62 of them pathogens] were recognized as established elsewhere in Africa.

Ninety-five percent of the arthropods were considered likely to arrive as contaminants on commodities, i.e. on their host plants; 23% were also likely to arrive as stowaways; some good fliers already present in neighboring countries could also enter unaided.

The 64 pathogen species included 14 bacteria, 16 fungi, 14 nematode, seven water moulds (Kingdom: Chromista), and 13 viruses. Sixty-two of these species have been detected on the African continent; 46 are reported in neighboring countries. Thirty-one (48.4%) of the pathogenic organisms were considered likely to arrive both as contaminants on commodities and/or as stowaways; Twenty-six (40.6%) probably arrive only as contaminants; five could arrive exclusively as stowaways. Kenis et al. (2022) specify which of the fungi, nematodes, viruses, bacteria, and water moulds fall into which category.

The most important input in the threat scoring process was likelihood of entry. The unsurprising result was that species known to be in neighboring countries or spreading rapidly in Africa received the highest overall scores. The likelihood of establishment was less important because the assessors had already excluded species they thought would encounter an unsuitable climate or absence of host plants. The impact score played an important role in the overall score; it was based primarily through their potential economic impact. There is little information about or attention to the potential threat of non-native plant pest species to non-commercial plants. Kenis et al. (2022) cite well-known examples to remind us that invasive plant pest species have had “huge impacts” on native tree species and biodiversity in North America and Europe. On the African continent, most non-native pests attack mostly concern exotic trees. They note one exception, Euwallacea fornicatus, DMF a wood-boring beetle from Asia killing many native trees in South Africa.

*Bemisia tabaci*; one of the arthropod pests in a country bordering Ghana; photo courtesy of INCTELUNI

Kenis et al. (2022) state that some of the several alien arthropods and pathogens identified in neighboring countries might already be present in Ghana although not yet recorded or identified to the species level. They say it is essential to clarify these species’ status by enhanced surveillance and applying morphological and molecular methods. Some of these possibly introduced species received high scores in the assessment. They threaten cocoa, a key crop in Ghana, and vegetable crops.

I am disappointed that Kenis et al. (2022)’s main actions suggested for both arthropod and pathogenic species that scored highly are to ramp up surveys and to conduct full pest risk analyses. It is true, as thy point out, that such assessments are required by international regulations before a country may implement phytosanitary measures. [See discussion of the requirements of the International Plant Protection Convention here.]

To some extent, the horizon scan echoed the obvious: most of species ranked high are already on the African continent, including 19 arthropod and 46 pathogenic species known to be established in neighboring countries. Plus, the recommended actions are minimal. Since Kenis et al. (2022) is essentially the scan itself, it provides no information on whether Ghana has implemented the recommendations. Still, given what I assume is lagging preparation across most of Africa, the horizon scan might be useful in encouraging countries to set priorities and take some action.

Cyprus

The second case study of applying horizon scanning is more encouraging. Scientists on Cyprus tried to assess the efficacy of their own horizon scanning exercise. I applaud their decision to do so. The horizon scan itself might have been undertaken on their own initiative? Or it might have been taken on in response to European Union regulations, which oblige Member States to enact measures to prevent or manage introduction and spread of invasive species designated as of Union Concern. The Union also encourages development of national invasive species lists and provides a legal basis for emergency measures in response to a detection.

Scientists carried out two horizon scan workshops in 2017 and 2019. The two workshops evaluated 225 and 352 species, respectively, to predict which are most likely to arrive and the level of provable impact to Cyprus’ biodiversity, human health, and economy. In 2023, four to six years after the workshops, scientists evaluated the listed species to reveal the accuracy of the predictions and actions taken so far (Martinou et al. 2026).

During the period 2017 – 2023 there were 183 Martinou et al. (2026) found publications naming 183 non-native species not previously officially detected in Cyprus. (As I will discuss later, a significant number of these species had been present on the island in 2017 but knowledge of their presence did not reach the assessors.) Of the 183 newly reported species, 31 had been included on some list of invasive species (e.g., EPPO or European Union list of species “of Concern”) or predicted by the horizon scanning exercises to rank amongst the top 100 riskiest species.

Cyprus’ horizon scans highlighted the risk posed by 10 of these 26 species. Martinou et al. (2026) focused on seven of them as having been ranked as high risk to the nation’s BD, human-health or economy. They added an eighth species, a venomous marine fish.

A further 10 species that were detected in the country had received lower impact scores, so they had not been included on the high priority lists of the horizon scans.

One of the species allotted a lower impact score, Spodoptera frugiperda, is under eradication, although it is widely distributed on the island. This action might be in response to the species’ inclusion on the EPPO A2 list.

As I noted above, scientists learned that 17 of the species had been present in Cyprus before the scanning exercises were undertaken but since their presence was then unknown to the participants, they were assessed as if still had not been introduced. This points to the country’s non-native species checklists not being fully up to date at the time.

Nine plant species common in the plant trade were most certainly present on Cyprus before the horizon scans (2017), but there were no published reports of their escape from cultivation. Nevertheless, they might have already been present in the wild. It is also possible that at least some escaped since the scans. Always tricky; always depends on who looking where.

Actions upon detection of specific taxa

Detection of the common myna (Acridotheres tristis) – a species widely recognized as invasive – occurred in January 2022, close to a port. Eradication measures were implemented by the wildlife agency. Martinou et al. (2026) believe the introduction was facilitated by shipping. They think there is an extremely high risk of repeated introductions of mynas.

*Aedes aegypti*; photo by James Gathany via Flickr

Two mosquitoes were detected in 2022. A pilot project to eradicate The yellow fever mosquito, Aedes aegypti, was begun in 2023. There is no information about its success. The Asian tiger mosquito, Aedes albopictus, has been documented by citizen scientists as spreading rapidly in the suburbs of Limassol and Nicosia. To date the proposed interventions have been unsuccessful, possibly due to focusing on public land while the mosquitoes can also breed on private properties.
Detection of the little fire ant Wasmannia auropunctata (in 2022) was not surprising since it had already invaded other regions of the Mediterranean. Martinou et al. (2026) believe the introduction was probably facilitated by the plant trade. The scientists note that ant management and eradication efforts are both challenging and costly, but do not report whether any has been initiated.
Detection of several marine invasive species was reported, some by citizens, e.g., divers or fishermen.
Among the 17 species determined to have been present on the island since before 2017 were some fairly conspicuous vertebrates: brown rat (Rattus norvegicus), raccoon Procyon lotor, two tortoise species, house crow (Corvus splendens) ruddy duck (Oxyura jamaicensis). Also two more ant species, Solenopsis geminata and Trichomyrmex destructor. There were also several non-native plant species, including the notorious seaweed Caulerpa taxifolia.

Value of the Horizon Scan

I am surprised that Martinou et al. (2026) do not explore why so many detections were published in 2022 since they assert that horizon scanning helped raise awareness amongst the authorities, scientists and the public. They do note that this awareness led, in some cases, to a rapid response by the competent authorities. Martinou et al. (2026) assert further that the exercise facilitated communication between invasive species experts, policy makers and society, encouraged active engagement and raised awareness regarding the importance of early warning, rapid response, and management of IAS. They therefore propose that the horizon scanning process for the island of Cyprus be repeated regularly – every five to 10 years – since new introductions continue. These efforts should include development pathway management plans and contingency planning that would be shared with local authorities and stakeholders.

Martinou et al. (2026) note two detections that have not, apparently, resulted in establishment. A dead specimen of brown marmorated stink bug (Halyomorpha halys) was reported in luggage in May 2022, the result of ‘Bug Alert Cyprus’ awareness campaign. The Colorado potato beetle (Leptinotarsa decemlineata) was detected in 2010 by Department of Agriculture inspectors in a consignment of potatoes. The agency ordered immediate destruction. Imports of potatoes are subject to special phytosanitary requirements for protected zones. It is not clear that this measure was implemented by Cyprus or is a European Union decree.

brown marmorated stinkbug; courtesy of Oregon Department of Agriculture

Martinou et al. (2026) are worried that no introductions have been reported at border crossings across the ‘Green Line’ [the United Nations-controlled buffer zone between Greek and Turkish portions of the island]. They call for enhanced cross-community collaboration and improved information and data sharing for border control staff and customs officers about invasive species. They suggest that border order inspections and pathway monitoring could be supported by local experts offering identification services for a variety of taxa. They suggest that the horticultural industry is a major pathway for the introduction of plants and insects such as ants.

Martinou et al. (2026) also advocate efforts to improve communication among the various institutions and authorities that discover bioinvasions and are responsible for taking action. While researchers + experts from government departments involved in the horizon scans are informed, the findings of the horizon scanning needs to be provided to e.g., customs officers, fishers, ship crews, pet shop owners, and school teachers. Much of this information might be exchanged through informal networks and through a growing body of web-based databases and other resources.

Early detection and rapid response depends increasingly on efforts by citizen scientists to report observations of IAS of concern. Martinou et al. (2026) note that six of the invasive species identified in the horizon scanning exercise were reported by citizen scientists. They express the hope that artificial intelligence and deep learning models could help identify species from photographs collected by citizen scientists on platforms such as iNaturalist. Such platforms also facilitate rapid dissemination of information to decision-makers who can take appropriate action. Martinou et al. (2026) also hope eDNA can help detect cryptic bionvaders, including freshwater or marine taxa.

As I blogged earlier, Mark Hoddle had endorsed several components of prevention programs:
* Early research to identify natural enemy species that might “self-introduce” along with the invading host.
* Collaborating with non-U.S. scientists to identify and mitigate invasion bridgeheads.
* Sentinel plantings. These plantings can also support research on natural enemies of key pests. [A year ago, Eliana Torres Bedoya of Ohio State alerted participants in the annual USDA research forum on invasive species that fungi, including potential pathogens, were isolated from asymptomatic plants;
Detection of the full range of fungal pathogens requires that samples must be collected throughout the growing season; microbes present differ.
Need to expand surveillance beyond symptomatic plants – at both sentinel gardens and plant health border inspection stations.
*Integrating online platforms, networks, professional meetings, and incursion monitoring programs into “horizon scans” for potential invasive species. He mentions specifically PestLens, (https://pestlens.info/); online community science platforms, e.g., iNaturalist; international symposia; and official pest surveillance, e.g., U.S. Forest Service’s bark beetles survey and surveys done by the California Department of Food and Agriculture and border protection stations.

That blog also cites Weber et al.’s support for sentinel plant nurseries because accidental plant and herbivore invasions often occur at the same points of entry.

At the 2026 meeting of the annual USDA Research Forum on Invasive Species, Ashley Schulz (Mississippi State) reported findings of study analyzing establishment of insects imported deliberately as biocontrol agents as clues to bioinvasion. She found that generalist phytophagous insects might be more likely to find a suitable host and survive after introduction. The “goldilocks” standard applies: the host must be sufficiently closely related to the insect’s native host to be recognizable but sufficiently distant so that it lacks defenses. Considering impact, phytophagous insects that feed on structures not easily restored – e.g., main stem or root, cause more damage than those that feed on easily replaced leaves. Entomopagous insect, on the other hand, must be able to find hosts that can hide or defend themselves. This means that highly specialized insects might be more likely to establish.

SOURCE

Hoddle. M.S. 2023. A new paradigm: proactive biological control of invasive insect pests. BioControl https://doi.org/10.1007/s10526-023-10206-5

Kenis et al. 2022. Horizon scanning for prioritizing invasive alien species with potential to threaten agriculture and biodiversity in Ghana. Neobiota 71: 129-148 (2022) doi: 10.3897

Martinou, A.F., J. Demetirou, I. Angelidou, N. Kassinis, A. Melifronidou, J.M. Peyton, H.E. Roy, A.N.G. Kirschel. 2026. Multiple introductiions of invasive alien species on a Mediterranean Island predicted by horizon scanning. Biological Invasions (2026) 28:41 https://doi.org/10.1007/s10530-025-03729-8

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
Or
https://fadingforests.org/

Tree-Killing Pests = Existential Threats to U.S. National parks

black bears in a whitebark pine (*Pinus albicaulis*) in Yellowstone National Park; tree species is vulnerable to white pine blister rust. Public image

America’s national parks protect some of Earth’s most unique and valued species, ecosystems, geologic features, and cultural sites. These values are under threat from multiple interacting climatic changes. Over the last 100 years, national park units have experienced a disproportionate degree of warming and precipitation change relative to the United States in general. These changes are projected to continue.

The types of change are not limited to temperature and precipitation. These alterations bring multiple cascading impacts such as extreme weather events, forest insect outbreaks, more frequent and severe wildfires, and other novel disturbance regimes. Furthermore, the new events occur both individually and simultaneously. Michalak et al. (2026) fear that these disturbances and stressors might trigger irreversible ecological transformations in our national parks. The authors hope to prompt park managers to evaluate park-specific threats and plan how to respond.

Michalak et al. (2026) analyzed threats from the multiple interacting forces to determine which parks are greatest at risk. They limited their analysis to 259 parks in the continental states (including Alaska) and to parks recognized by the agency as possessing natural resource values. Some historic or cultural sites are included; I am somewhat confused about the criteria applied. I regret that they lacked sufficient data to include parks on the Hawaiian and Caribbean islands.

Hawaiian birds threatened by avian malaria; picture via Flickr

Their analysis defined potentially transformative impacts as heightened risk of fire, drought, sea-level rise, and forest insects and pathogens (not limited to non-native species). An example of such impacts is a prediction that a significant proportion of the park’s area would be inundated during storm surge.

Michalak et al. (2026) identified 174 parks (67% of the units analyzed) as most exposed to one or more of these potentially transformative impacts. The number of parks facing cumulative vulnerability across multiple dimensions was highest in the Midwest and East. Their peril is due to high physical exposure to the transformational change, exacerbation of existing stressors, and high surrounding land-use intensity. Parks in the West were partially protected by less intense human land-use and the varied topography, which might provide climate refugia. However, those western parks tended to be the most exposed to multiple transformative impacts (as defined above).

At the national level (excluding the islands), 28% of the parks have a high fire hazard now; this rises to 38% of parks by an unspecified future time. They provided no estimate of the proportion of parks facing a risk in the future from the other factors. Current levels of risk are 25% at risk to summer drought; 36% (92 parks) at risk to forest pests; and 11% to sea-level rise. Again, across all parks analyzed, 174 – or 67% of the total – face one or more of these threats.

The authors conclude that the 60-old goal of conserving National parks as a “vignette of primitive America” – as stated by Leopold et al. (1963) – is no longer possible. Instead, park managers should now seek to steward resources “for continuous change that is not yet fully understood” as advocated by Colwell et al. (2014).

Michalak et al. (2026) found that the National parks are not prepared. Only 10% have had park-specific assessments; 37% had no assessment at any level. For individual National parks, likelihood of climate impacts and potential transformational changes remains uncertain. Determining where more in-depth, park- specific assessments are warranted is essential for allocating resources.

Michalak et al. (2026) define climate change vulnerability as the combined effects of exposure, sensitivity, and adaptive capacity.Exposure is the intensity of changes a location might experience. This includes changes in the climate itself (e.g., temperature or precipitation) plus changes in climate-exacerbated disturbances (e.g., fire, drought, and sea-level rise). Sensitivity is the extent to which a location or resource is affected – or existing stressors are amplified – by the changing climate, which can be either adversely or beneficially. For example, imperiled species might be further threatened if new conditions are more conducive to bioinvasion. Adaptive capacity is the ability of a system to adapt to the climate change impacts. For example, does human development impede species’ dispersal to new regions that support more suitable climate regimes. I appreciate that the authors note the importance of ensuring continuation of evolutionary processes.

A Subset of Threats: Invasive Species and Forest Pests

According to Michalak et al. (2026), National parks with the highest cumulative vulnerability scores were in the Midwest, Washington, DC, and along the Gulf Coast. The threats were high levels of human development, poor air quality, high proportions of non-native species, and low environmental diversity.

mountain pine beetle in Rocky Mountain National Park; photo by Bchemicoff via Wikimedia

National parks that scored high for forest pest risks are concentrated in the mountainous West and Northeast. While Michalak et al. (2026) do not say so, I assume this refers to widespread mortality of pines due to outbreaks of the native mountain pine beetle (Dendroctonus ponderosae). Thirteen parks in the West scored high for a “trifecta” of fire, drought, and forest pests. The consequences for these parks’ natural resources might be rapid, dramatic, and irreversible transformation of ecosystems. Michalak et al. (2026) mention specifically Rocky Mountain and Yellowstone National parks. Other parks facing a threat from forest insects or pathogens include all the crown jewels of the West: Grand Teton National Park, Crater Lake National Park, Glacier National Park, Great Basin National Park, Kings Canyon-Sequoia National Park, Yosemite National Park, and Mount Rushmore National Memorial.

limber pine (*Pinus flexilis*) at Haiyaha Lake, Rocky Mountain National Park. Species is vulnerable to white pine blister rust. Photo by F.T. Campbell

Another example is Mojave National Preserve, which has experienced increased fire risk linked to the presence of invasive annual grasses.

I know that in the Northeast, more than a dozen species of introduced insects and pathogens threaten forest resources in the parks, including hemlock woolly adelgid, emerald ash borer, spongy moth, and – most recently – beech leaf disease. Parks mentioned in supplementary material provided by Michalak etal. (2026) include Delaware Water Gap National Recreation Area, New River Gorge National River, Harpers’ Ferry National Historical Park, and the homes of Eleanor and Franklyn Roosevelt. See blog 356a and underlying article by Miller et al. (2023).

mature Fraser fir killed by balsam woolly adelgid in Great Smoky Mountains; photo by F.T. Campbell

Many other National parks in the East and Midwest also are reported to be impacted by introduced forest pests, among them Great Smoky Mountains National Park, Blue Ridge Parkway, Shenandoah National Park, Appalachian National Scenic Trail, Prince William Forest Park, Cumberland Gap National Historical Park, Gauley River National Recreation Area, Mammoth Cave National Park, Ozark National Scenic Riverways, Pictured Rocks National Lakeshore, Sleeping Bear Dunes, St Croix National Scenic Riverway, and Big Thicket National Preserve.

There are some odd omissions. The supplementary data list the Chesapeake and Ohio Canal National Historical Park as facing a threat from tree pests, but does not so list Rock Creek Park. The two parks are a few miles apart and share the same invasive forest pests! The supplementary data do not mention Gettysburg National Military Park, although Miller et al. (2023) say that more than half of the seedlings and a quarter of the saplings in the park are ashes. These trees are likely to be killed by the emerald ash borer. Perhaps the explanation is that canopy trees threatened by pests in these parks do not occupy more than 80% of the parks’ cover.

I appreciate the effort to compile a nationwide analysis of threats to our national treasures. By focusing on one of those threats, I do not intend to downplay the others. Specific to climate changes, the Trump Administration has told the National Park Service to remove educational signs describing the impact of climate change on, for example, the glaciers at Glacier National Park. An earlier Executive Order https://climate.law.columbia.edu/content/trump-issues-executive-order-climate-change-0 reversed President Obama’s 2015 memorandum that required Interior and other departments to “avoid and then minimize harmful effects to land, water, wildlife, and other ecological resources (natural resources) caused by land- or water-disturbing activities, and to ensure that any remaining harmful effects are effectively addressed, consistent with existing mission and legal authorities.” In February 2026, the Environmental Protection Agency revoked the “endangerment finding” for greenhouse gases, which is the foundation for all regulations governing emissions of those substances. Clearly we cannot hope for federal efforts to address these threats to the National parks during this Administration’s tenure.

I hope, nevertheless, that this study gets wide attention and stimulates renewed campaigns to counter all threats to our natural heritage.

shrunken glacier in Glacier National Park; photo by F.T. Campbell

SOURCES

Colwell, R. S. Avery, J. Berger, G.E. Davis, H. Hamilton, T. Lovejoy, S. Malcom, A. McMullen, M. Novacek, R.J. Roberts, R. Tapia, and G. Machlis. Revisiting Leopold: Resource Stewardship in the National Parks. Parks 2014 Volume 20.2

Leopold, A. et al. 1963. Wildlife Management in the National Parks. available here: chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://static-gcs.edit.site/users-files/30eb6df2212095e14d89a611f0f8f0f1/leopold-report-wildlife_management_in_the_national_park-1963.pdf?dl=1

Michalak, J.L., C.E. Littlefield, J.E. Gross, T.G. Mozelewski, J.J. Lawler. 2026. Relative Vulnerability of US National Parks to Cumulative and Transformational Climate Impacts. Conservation Letters, 2026 Vol 19, Issue 1; 19:e70020

Miller, K.M., S.J. Perles, J.P. Schmit, E.R. Matthews, M.R. Marshall. 2023. Overabundant deer and invasive plants drive widespread regeneration debt in eastern United States national parks. Ecological Applications. 2023; 33:e2837. https://onlinelibrary.wiley.com/r/eap

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

https://fadingforests.org

Plant invasions grow everywhere

invasion of Chinese privet (*Ligustrum sinense*)

A decade ago I posted a blog reporting that 39% of forests surveyed under the Forest Inventory and Analysis (FIA) system were invaded by one or more invasive plants (Oswald et al. 2015). By regions, Hawai`i had the highest invasion intensity – 70%. The second highest density was in the eastern forests – 46%. Forests in the West ranked third, with 11% of plots containing at least one of the monitored invasive plant species. Finally, forests in Alaska and the Intermountain regions both had 6% of plots invaded.

I rejoice that US Forest Service scientists have continued to analyze their data on plant invasions. Analysis of the most recent data shows alarming increases in invasions everywhere since 2015. However, the scientists could not determine a nation-wide percentage because many areas in the West had not yet been surveyed anew. They did determine that the number of inventory plots containing invasive plant species rose in 58.9% of surveyed counties. Furthermore, in 73.2% of the counties the plots experienced an increase in species richness of invading plant species. While increases were observed in all regions, they were greater in the East than in the West — and in the USFS Southern region compared to the Northern region. Specifically, the proportion of forest plots in the East (USFS Southern and Northern regions) invaded has risen from 46% to 52.8%. In the Rocky Mountains they rose from 6% to 11%. In Hawai`i plots having invasive plants grew from 70% to 83.2%. Surveys in the Pacific Coast states have not yet been completed so this region is not included in the analysis (Potter et al. 2026). It is not clear to me how the current boundaries of the western regions – which are based on Bailey’s ecosystem boundaries relate to the 2015 boundaries, which were based on USFS official regions. Hawai`i is clearly the same.

Porter et al. (2026) concluded that in the forests of the East plant invasions are so extensive that elimination of their impacts is practically impossible. Their spread to new areas is unhindered now and, I would add, is likely to remain so without heroic counter measures.

Forests in the East have a greater mean richness of invasive plant species than do western forests. In particular, there is a profusion of shrubs and vines as well as trees. The West has a greater diversity of invasive forbs. The diversity of invasive grasses is high in both regions.

kudzu (*Pueraria montana*) spreading from edge into forest in Virginia; photo by F.T. Campbell

Potter at al. (2026) worry that the apparently lower level of plant invasions in the West might be an artifact of a higher proportion of plant species being at an earlier stage of invasion. That is, the species have not yet established sufficiently widely to be classified as invasive.

thicket of guava (*Psidium cattleianum* ) replacing `ohi`a killed by ROD; Hawai`i Island; photo by F.T. Campbell

Of course, the situation in Hawai`i is much worse. Another, more detailed, discussion of invasive plant species in Hawai`i pointed out that relying on data reflecting canopy-level trees obscures the real picture. While “only” 29% of large trees across the Islands are non-native, about two-thirds of saplings and seedlings are. Potter et al. (2023) expected that plant succession will result in non-native tree species taking over the canopy. This likelihood exists regardless of the impact of rapid ‘ohi’a death since ‘ohi’a lehua (Metrosideros polymorpha) is not reproducing even when seed sources are plentiful and people remove invasive forbs and grasses Potter et al. (2023).

The nation-wide analysis of Potter et al. (2026) does not include forests on U.S. Caribbean islands, i.e., Puerto Rico and the Virgin Islands. See here for a description of this situation. In summary, 33 of 57 (58%) of non-native tree species tallied by FIA surveyors are actual or potential high-impact bioinvaders. Furthermore, 21 (38%) of the non-native species occurred on at least 2% of the FIA plots – far above the seven species fitting this description in the continental U.S.

As these sources, and those with a broader perspective, demonstrate that we should not ignore invasions of our forests by non-native plants. These species erode forest productivity and provision of the full range of ecosystem services, hinder shifting (?) forest uses, and degrade biodiversity and habitat.

These invasions also impose extensive financial costs from lost or damaged resources (Potter et al. ( 2022). Potter et al. (2026) note that these negative outcomes depend on interactions between the traits of the non-native plants and the biomes being invaded. These impacts are greatly exacerbated in Hawai`i because more than 95% of native species on the Islands are endemic. This includes 67% of the large trees still present in the forests. As Potter et al. (2023) point out, extirpation of any of these species is a global loss.

ʻōhiʻa lehua (*Metrosideros polymorpha*); photo by F.T. Campbell

Data issues

Potter et al. (2026) note that in the Northern region only about 20% of plots were surveyed for invasive plants. They state that these difference in sampling intensity does not affect statistical analyses across broad scales.

The regional lists of invasive plants were developed by experts. They include those species thought at the time to be most damaging. Of course, there are other non-native plant species that might be present – and some might prove to be invasive over time (Potter et al. 2026). I have been unable to determine whether the regional lists are updated periodically. Because of this structure of the FIA system, these surveys can assess only spread of already-established species. It is not suitable for early detection of new species entering the forest.

For all these reasons, the analyses in Porter et al. (2026) probably underestimate the total abundance of non-native plant species in U.S. forests. Indeed, the time lag between introduction or even identification of invasive species and their eventual ecological and economic impact obscures their full impact. This ever-increasing invasion debt probably contributes to decisions not to implement effective countermeasures.

Recommendations

How do we set priorities for responding to nearly unmanageable situations? We sharpen our focus on the most damaging pathways of introduction, the most vulnerable regions, and the most at-risk species.

The high-risk pathways are imports of plants for planting and wood – including but not limited to crates, pallets, and other forms of packaging.

Vulnerable regions start with the Hawaiian Islands, Puerto Rico, and the Virgin Islands; and include many biodiversity-rich areas on the continent. We should enhance monitoring of these vulnerable regions by federal, state, and tribal agencies, conservation organizations, citizen scientists, and others. Surveys must report all non-native plant present, not just those already known to be invasive. These data will improve detection of new species and better inform us about factors affecting species’ spread.

Also, I support Potter et al.’s (2026) emphasis on the wildland-urban interface as an area of high human-environment conflict.These include, but are not limited to, plant invasions. The authors point out that we need new policy, management, and scientific tools to address threats in these vulnerable and too-often ignored social and ecological zones.

This increase in available information must be paired with management of the factors that facilitate invasion. Some of these are associated with ecosystems. But the key target must be plant species being brought into the region by people for various purposes. This is often for ornamental horticulture.

lesser celandine (*Ficaria verna*) dominating herb layer in a Virginia forest; photo by F.T. Campbell

We must ask state legislatures and Congress to empower regulatory agencies – e.g., their state departments of agriculture and USDA’s Animal and Plant Health Inspection Service – to be far more more assertive and pro-active. For example, they must give higher priority to the full range of ecological and economic impacts of invading plants, not just damage to agriculture.

Evans et al. (2024) urged prioritizing for state regulation those species in the ornamental trade that are projected to remain or become abundant under evolving climate conditions. Beaury et al. (2023) called for regulating the nursery trade at the national level – reflecting the scope of sales.

SOURCES

Beaury, E.M., J.M. Allen, A.E. Evans, M.E. Fertakos, W.G. Pfadenhauer, B.A. Bradley. 2023. Horticulture could facilitate invasive plant range infilling and range expansion with climate change. BioScience 2023 0 1-8 https://doi.org/10.1093/biosci/biad069

Evans, A.E., C.S. Jarnevich, E.M. Beaury, P.S. Engelstad, N.B. Teich, J.M. LaRoe, B.A. Bradley. 2024. Shifting hotspots: Climate change projected to drive contractions and expansions of invasive plant abundance habitats. Diversity and Distributions 2024;30:4154

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

Potter, K.M., B.V. Iannone III, K.H. Riitters, Q. Guo, K. Pandit, C.M. Oswalt. 2026. US Forests are Increasingly Invaded by Problematic Non-Native Plants. Forest Ecology and Management 599 (2026) 123281

Potter K.M., K.H. Riitters, and Q Guo. 2022. Non-native tree regeneration indicates regional and national risks from current invasions. Frontiers in Forests & Global Change Front. For. Glob. Change 5:966407. doi: 10.3389/ffgc.2022.966407

Potter, K.M., K.H. Riitters, B.V. Iannone, III, Q. Guo and S. Fei. 2024. Forest plant invasions in eastern US: evidence of invasion debt in the wildland‑urban interface. Landsc Ecol (2024) 39:207 https://doi.org/10.1007/s10980-024-01985-y

Posted by Faith Campbell

https://fadingforests.org

Invasive plants threaten integrity of eastern U.S. forests

garlic mustard (*Alliaria petiolata*); photo by Katja Schulz via Wikimedia

I welcome a recent series of studies documenting the extent of plant invasions in forests of the eastern United States and the socio-economic conditions that contribute to a state of affairs increasingly recognized as a crisis. I wish, however, that the authors had devoted more attention to the role of deliberate planting of non-native species and the resulting propagule pressure.

I summarize here findings of two studies written by largely the same scientists and relying on the same underlying data: surveys of forest plots conducted under the Forest Inventory and Analysis (FIA) program. In this blog, if focus on the extent of invasive plant presence in the forests of the eastern United States. In an accompanying blog I will summarize the status of plant invasions in forests nation-wide.

As I have noted in earlier blogs, link a decade ago one or more invasive plant species had already invaded 46% of FIA plots in the eastern U.S. (Oswald et al. 2015). This situation has worsened. Updated data show that 52.8% of these plots contain invasive plants. In the USFS Southern Region, invasive plants have been documented on 55.3 million ha. In the Northern Region, they are found on 36.9 million ha. (Only ~20% of FIA plots in the Northern Region were surveyed for invasive plants.) In some counties of the 37 states constituting these two USFS regions, 80% of inventoried forest plots contain invasive plants. Areas with lower levels of invasion are found in parts of New England, the Great Lakes states, southern Appalachians, southeastern coastal plain, and western Texas and Oklahoma (Potter et al. 2026). Spread of these bioinvaders is largely unchecked – either throughout the East or “just” in the South. In any case, the extent and intensity of these invasions are so great that their complete removal – or elimination of their impacts – is “practically impossible” (Potter et al., 2024; Potter et al. 2026). [It is not clear whether the scientists mean “nearly” or “in practice”. Or that this difference is important.]

[In comparison, in the West less than 30% of FIA plots are invaded, on average. In Hawai`i, more than70% are (Potter et al. 2026).]

The scientists analyzing the FIA data warn that the extent and impact of plant invasions in eastern forests is undoubtedly worse than these data indicate. The records include only some of the non-native plant species present — those considered to be the worst invaders at the time regional lists were compiled – apparently in the first years of the 21^st Century (Potter et al. 2026).

Japanese honeysuckle (*Lonicera japonica*) photo by Chuck Bargeron

The scientists emphasize the role of disturbance in promoting plant invasions. They cite various studies as well as the FIA data to document that forest edges facilitate non-native plant establishment and spread into forests. They stress various aspects of suburban development, including roads and other transportation corridors. It follows that invasion rates are highest in the “wildland-urban interface (WUI).” [The wildlife-urban interface is the zone of transition between unoccupied land and human development; the zone where structures meet or intermix with undeveloped land and its vegetation.] They worry that the WUI is growing faster than any other land use type in the country – and especially rapidly in the East. As a result, the scientists expect more and worse invasions in the future (Potter et al., 2024 and Potter et al. 2026).

I appreciate that they highlight the uniqueness of WUI ecosystems. Housing development in the WUI has numerous effects on natural ecosystems, including habitat modification and fragmentation followed by diffusion of the direct and indirect effects of anthropogenic activities into neighboring ecosystems at different scales. As regards specifically non-native plants, this transmission occurs through a combination of (1) human-driven disturbances to native ecosystems that promote plant invasion and (2) providing a source of non-native plant propagules in their yards and gardens. These plants can then spread into and establish in nearby ecosystems (in this case, forests). [I note that tree-killing arthropods and pathogens also can be introduced in the WUI.] (Scroll below “Archives” to “Categories”, click on “forest pests” and “wood packaging”.)

They also found that plant invasions are more strongly related to older, than more recent, land-cover changes. Survey plots that have been located in the WUI since 1990 or earlier had on average 2.6% more invasive plant cover and 0.33 more invasive plant species than those that were classified as being in the WUI in 2000 or 2010. Their explanation is that the WUI forests experienced decreased spatial integrity, increased forest-developed area edges, and falling proportions of forest in the surrounding landscapes. In addition, the human population in the vicinity might have grown. All these factors that would increase forest fragmentation and the plots’ susceptibility to invasion.

The other side of the coin is propagule pressure. Both Potter et al (2024) and Potter et al. (2026) note that the flora of residential landscapes – rural as well as suburban – is typically dominated by non-native plant species. Still, I think these studies downplay the impact of this ubiquity of non-native plants in all anthropogenic landscapes.

In discussing the higher invasion rates found in survey plots located in WUIs dating from the 1990s they made no mention of human activities that promote plant invasions. There are several. Plants growing in those older yards had one or two more decades to flower – and for their fruits and seeds to be transported into the forest by birds, wind, or water. Residents might have decided to beautify their neighborhood by planting shrubs or flowers in the woods. Maybe they succumbed to the temptation to dump yard waste in the woods – thinking it would be absorbed by “nature”. Since plant invasions take time to unfold, these additional years of human-mediated exposure are highly relevant. Another factor is that people who choose to live in wooded surroundings probably choose horticultural plants that thrive under such conditions – exactly those best able to establish beyond the property line.

Another opportunity to discuss these factors came from the discovery that plant invasion rates are higher in association with “interface” rather than “intermix” WUI forests. [“WUI interface forests” are those where settled areas abut wildlands. In “WUI intermix forests” the structures are scattered.] They speculate about reasons. Potter, et al. (2024) mention that invasions originating from older housing developments have had more time to establish (or at least to be detected) given the well-known lag associated with plant invasions.

I wish they had focused more on the probable difference in suburban development across time. While I was growing up in expanding suburbs in the 1950s, I observed that the earlier housing developments were either built on land that had been cleared to support agriculture or the builders cleared the forest to make construction easier and cheaper. More recently, wealthier buyers have sought residences on more wooded sites – so creating an “intermix” WUI. Potter et al. (2024) speculate that locations in the “interface” WUI are closer to high-density urbanization so have higher exposure to non-native plants. They do not discuss whether the “interface” WUIs are older, thus giving associated plantings longer years to proceed through the stages of bioinvasion.

burning bush (*Euonymus alatus*) invading a forest in Virginia; photo by F.T. Campbell

The Role of Deliberate Planting?

I recognize that these authors analyzed mountains of data. However, I wish they had incorporated the findings of numerous scientists who have analyzed the role of deliberate planting – especially ornamental horticulture – in facilitating introduction and spread of invasive plants. (Scroll below “Archives” to “Categories” and click on “invasive plants”. Also See Reichard and White 2001 and Mack 2000).

As I hope USFS scientists are aware, recent studies confirm the continuing role of ornamental horticulture in plant invasions. Kinlock et al. (2025) blog 440 found that more than 1,600 plant species sold by nursery and seed catalogs over 200 years had “naturalized” somewhere in the continental 48 states. They do not discuss what proportion of these species are truly damaging invaders. Fertakos and Bradley (2024) found that species were likely to establish if they were introduced to as few as eight locations. Beaury et al. (2024) found that half of 89 plant species recognized as invasive are sold in the same locations where they are invasive. Another 25 species are sold by one or more nurseries located in an area that is currently unsuitable for those species, but that will become more suitable for invasion as temperatures warm.

Potter et al. (2026) acknowledge that the ornamental plant trade is likely to continue introducing new plant species into U.S. forests. However, they recommend only updating the lists of invasive plants to be included in future surveys. Apparently these lists have not been updated since 2004.

Potter et al. (2024) go farther, urging efforts to encourage homeowners to plant more native and environmentally friendly private landscapes. They note that such advocacy is complicated by the fact that non-native – even invasive – species provide valued ecosystem and cultural services.

I add that the nursery industry and their customers enjoy enormous lobbying clout.

Many associations – native plant societies, regional or state invasive plant councils, etc. – are pursuing this approach. To research these efforts, visit the websites for the state native plant societies and the Southeast Exotic Pest Plant Council, Mid-Atlantic Invasive Plant Council, and Midwest Invasive Plant Network. These voluntary efforts have yielded some success. But they have not resulted in adequate protection for our ecosystems. Dr. Douglas Tallamy points out that even non-invasive, non-native plants disrupt food webs.

The insufficient attention to the role of the plant trade in articles intended to be comprehensive has crucially important impacts. As both Potter, et al. (2024) and Potter et al. (2026) affirm, determining which factors are most important in facilitating plant invasions of eastern American forests is the necessary foundation for identifying and implementing the most efficient and effective counter measures.

These scientists are employees of the U.S. Department of Agriculture. If departmental leadership interpret their studies as justifying inaction on regulating plant sales, USDA’s regulatory agencies will not respond. And we will continue failing to curtail introduction and spread of damaging plant invasions.

I agree with the authors on the need for enhanced monitoring and management of WUI zones in the East to detect new species or new locations of invasion and the need to develop better tools for these purposes. However, I ask all stakeholders to follow Evans et al. (2024), who urge prioritizing for state regulation those species in the ornamental trade that are projected to remain or become abundant under evolving climate conditions. Or, more aggressively, follow Beaury et al. (2023)’s call for regulating the nursery trade in a manner consistent with the scope of the horticultural trade at the national level. That would require legislation, since the Federal Noxious Weed Act does not currently address long-established, widespread species. Beaury et al. (2023) also note that existing state restrictions are outdated, tend to include only a few weeds that plague agriculture rather than those that invade natural systems, and are irregularly enforced.

orchids in Everglades National Park; photo by F.T. Campbell

I conclude by agreeing with the scientists that managing the disturbance component of plant invasions points to protecting particularly forests of high conservation value. They suggest adoption of land-use planning rules aimed at this goal. However, as they point out, such action will be extremely unlikely given the magnitude of predicted land-use changes in the country and powerful demographic factors driving them. I would add other barriers: the lobbying clout of the real estate industry and homeowners plus the local nature of zoning decisions.

SOURCES

Fertakos, M.E. and B.A. Bradley. 2024. Propagule pressure from historic U.S. plant sales explains establishment but not invasion. Ecology Letters 2024;27:e14494 doi: 10.1111/ele.14494

Kinlock, N.L., D.W. Adams, W. Dawson, F. Essl, J. Kartesz, H. Kreft, M. Nishino, Jan Pergl, P. Pyšek, P. Weigelt and M. van Kleunen. 2025. Naturalization of ornamental plants in the United States depends on cultivation and historical land cover context. Ecography 2025: e07748 doi: 10.1002/ecog.07748

Oswalt, C.M., S. Fei, Q. Guo, B.V. Iannone III, S.N. Oswalt, B.C. Pijanowski, K.M. Potter. 2016. A subcontinental view of forest plant invasions. NeoBiota. 24:49-54 http://www.srs.fs.usda.gov/pubs/48489

Potter, K.M., K.H. Riitters, B.V. Iannone III, Q. Guo and S. Fei. 2024. Forest plant invasions in the eastern United States: evidence of invasion debt in the wildland‑urban interface. Landsc Ecol (2024) 39:207 https://doi.org/10.1007/s10980-024-01985-y

Posted by Faith Campbell

https://fadingforests.org

Bird nesting habitats – why no mention of invasive species or deer?

ovenbird (*Seiurus aurocapilla*); photo by Rhododentrities via Wikimedia

Studies of forest ecosystems in eastern North America that claim to be comprehensive still too often make no reference to invasive species – pests, earthworms, or plants. I try here to bridge these gaps.

Akresh et al. (2023) conducted a meta-analysis of bird species’ use of forests as nesting habitat. They applied the Partners-in-Flight to evaluate the community-wide bird conservation values of unmanaged forests compared to various levels of tree removal by harvest. Because of the decline of many bird species that prefer shrubland or early-successional stands, their process gave highest ranks to management approaches that retained 40%–70% of the canopy trees.

Their study notes that habitats for shrubland birds comprise only about 6% of forests in the eastern U.S. They don’t provide data for southeastern Canada. But hasn’t this scarcity of open upland, non-wetland, habitats in this region been true for thousands of years?

The type of forest that undoubtedly has shrunk significantly in recent centuries is “virgin” (or old-growth or late-seral) forests. As Akresh et al. (2023) report, contemporary closed-canopy forests in eastern North America are predominantly structurally homogeneous, mid-seral, even-aged, stands that have regenerated on land previously cleared for either agriculture or timber. These forests are much younger from a forest developmental perspective than precolonial forests; they lack the latter’s range of tree fall gap sizes and multiple age-classes. The tiny fraction of eastern forests that are in the late-seral stage might have higher species richness and conservation value for birds, but since they are usually not under management, Akresh et al. (2023) did not include that question in their analysis.

Akresh et al. (2023) list the bird species whose density appears to be closely linked to various tree canopy densities. For example, ovenbirds and brown creepers promptly decline in abundance in response to any amount of tree harvesting. Two other species — wood thrush and cerulean warbler — have declined steeply range-wide in recent decades. Nesting densities of three of these four species (excluding the warbler) are significantly higher in areas harvested in ways that retain a greater percentage of trees. Densities of another five bird species (Acadian flycatcher, hermit thrush, black-throated green warbler, and red-breasted nuthatch) are also higher in areas with a greater proportion of trees retained.

Another nine species had a more complex relationship with tree densities but still had lower densities in stands with low tree retention. These were blue-gray gnatcatcher, blue-headed vireo, blackburnian warbler, black-throated blue warbler, eastern wood-pewee, least flycatcher, red-eyed vireo, scarlet tanager, and yellow-bellied sapsucker. They found little relationship between bird density and tree retention for five putative mature-forest species (American redstart, great-crested flycatcher, hooded warbler, veery, and yellow-rumped warbler).

scarlet tanager (*Piranga olivacea*); photographed in scrub at Edwin B. Forsythe (Brigantine) NWR by F.T. Campbell

Akresh et al. (2023) claim that silviculture approaches can be used to restore aspects of the structural and compositional conditions found in old-growth forests to second-growth systems, providing a potential pathway for rapidly increasing the conservation value of these areas for bird species. They advocate reducing canopies moderately via variable retention harvests, shelterwood establishment harvests, and irregular shelterwood systems. This strategy can increase understory vegetation density, which they assert can then increase foraging and nesting opportunities for both many mature-forest bird species and many shrubland birds.

I am skeptical; it is much easier to create openings in the canopy than to “create” large trees supporting cavities and associated fauna and flora utilized by some bird species. The authors do advise managers that late-seral, unharvested stands can provide important habitat for old-growth-dependent taxa and any intensive forestry should also take into account other factors.

old-growth hemlock stand in Cook Forest State Forest, Pennsylvania; photo by F.T. Campbell

In addition, often the understory vegetation that responds to the more open environment will be invasive non-native plants. Already about half of eastern U.S. forests have been invaded by non-native plants (Oswalt et al. 2016; Kurtz 2023). Many of these are shrubs: honeysuckles, privets, roses, buckthorn. Management of these plants is difficult – especially when opening the canopy to allow light to reach the forest floor. (at www.nivemnic.us, scroll down to “categories”, click on “invasive plants”.) So the question arises, do the non-native plant species adequately substitute for native shrubs in providing resources needed by those birds?

Maybe. Gleditsch and Carlo (2014) found that a shrub layer dominated by non-native honeysuckle shrubs (Lonicera species) does support nesting populations of several common species, especially catbird (Dumetella carolinensis), American robin (Turdus migratorius ), and northern cardinal (Cardinalis cardinalis). However, they did not consider the species of concern to Akresh et al. (2023) – the rare species that prefer open-canopy, early-successional communities. So they do not inform us whether these high-priority species can utilize shrublands dominated by non-native species. Gleditsch and Carlo (2014) apparently did not find nests of several species considered to be associated with mature forests. So, again, these forests’ value for conservation remains unclear. Gleditsch and Carlo (2014) do counter earlier fears that these non-native shrubs are “traps” for nesting passerine birds. (The concern was that the plants’ structure facilitated nest raiding by predators.) They say, instead, that these plants’ effects are species-specific, context-dependent, and often a mix of both positive and negative outcomes.

invasive shrub honeysuckle; photo by Kevin Casper via public.domain.pictures.net

Akresh et al. (2023) also do not address the impact of browsing by super-abundant deer. Others (at www.nivemnic.us, scroll down to “categories”, click on “deer”.) have demonstrated that interactions of deer predation with invasive plants is especially damaging to native flora. Considering forests from Virginia to Maine, Miller et al. (2023) advise opening the canopy or subcanopy of forests to promote tree regeneration where deer and invasive shrubs overlap only where deer are controlled.

I have seen no recent analyses of the impact of widespread pest-caused tree mortality beyond some early efforts focused on eastern hemlocks and on high-altitude whitebark pines.

SOURCES

Akresh, M.E., D.I. King, S.L. McInvale, J.L. Larkin, and A.W. D’Amato. 2023. “Effects of Forest Management on the Conservation of Bird Communities in E North America: A Meta-Analysis.” Ecosphere 14(1):e4315. https://doi.org/10.1002/ecs2.4315

Gleditsch, J.M. and T.A. Carlo. 2014. Living with Aliens: Effects of Invasive Honeysuckles on Avian Nesting. PLOS One. September 2014. Volume Nine Issue Nine. E107120

Posted by Faith Campbell

https://fadingforests.org

It Zigs, We Zag: Following the Invasion of Elm Zigzag Sawfly in North America

elm zigzag sawfly larvae feeding on an elm leaf; photo by Delaney Serpan

As one of the newest – and most unique – invasive insects, elm zigzag sawfly (EZS; Aproceros leucopoda) has been making headlines across the eastern U.S. and Canada since 2020. The defoliating pest was first confirmed in North America in Québec, Canada and has since spread rapidly across many states and provinces. As its name suggests, EZS larvae feed primarily on elm in a distinctive zigzag pattern. Moving inwards from the leaf edge, the larvae can eventually consume nearly the entire leaf, leaving nothing but the midrib and a few lateral veins behind. Defoliation from EZS can range from nearly undetectable to 100% canopy defoliation of a mature tree.

(See earlier blog here.)

adult zigzag sawfly; photo by Delaney Serpan

Elm zigzag sawfly biology

EZS is a multivoltine insect, meaning it can have multiple generations in a single growing season. In Europe, where EZS has been invasive since 2003, 1 to 4 generations are common though up to 6 generations have been recorded. In the U.S., many regions document up to 5 generations per year.

In the early spring, EZS emerges from the soil where it has overwintered. They reproduce parthenogenetically- a form of asexual reproduction- allowing them to lay eggs immediately following adult emergence. Each individual is able to lay up to 49 eggs, drastically increasing EZS reproductive potential. Once the eggs hatch, the larvae begin feeding on the foliage until they are ready to pupate. At that point, the larva may build a summer cocoon, attached to a nearby object such as a branch or fence post. Four to 7 days later, an adult will emerge. The entire life cycle only takes 3 to 6 weeks. Alternatively, the larva could drop to the soil beneath the tree’s canopy where it will build its winter cocoon and overwinter, waiting to repeat the cycle the following spring. A small portion of each generation create overwintering cocoons.

EZS summer cocoons attached to the underside of a leaf with evidence of larval feeding; photo by Delaney Serpan

Where is EZS now?

As of the end of 2025, EZS can be found in 15 states and 4 provinces as far west as Minnesota and Manitoba and as far south as North Carolina and Tennessee.

map of states/provinces with official EZS detections;

Invasion pathways in North America are currently unknown; however, EZS has been documented attaching its summer cocoons to truck wheel wells and other objects which may be moved. The subsequent movement of these objects can potentially contribute to EZS spread. It has also been suggested that infested elm nursery stock or potted soil of any plants could be a potential pathway for EZS, but more research is needed to fully understand this.

EZS cocoons on truck – under side mirror & wheel well; photos by Jared Beach, adapted from Oten et al. 2025

How does EZS affect the trees?

Defoliating pests typically decrease the aesthetic value of trees but leave the host largely unharmed. Across Europe and its native range of eastern Asia, EZS defoliation is relatively minimal, with the occasional severe outbreak resulting in total defoliation of a tree. Resulting branch dieback is even more uncommon.

When EZS was first found in North America, particularly North Carolina and Virginia, there were initial concerns about the implications of a warmer climate accelerating development. Like most insects, EZS development is related to temperature; a warmer climate allows for faster insect development. It was hypothesized that a longer growing season could allow for faster population growth and potentially more damage to host trees. At this point, it is still unclear if this will consistently occur in the southern extent of the range. In North Carolina, reported damage has varied widely since it was found there in 2023. Some trees have been 75% defoliated or more multiple years in a row and are exhibiting upwards of 20% branch dieback after just 3 years. However, trees with less than 10% defoliation and no branch dieback have also been recorded.

Since its first detection in North America, researchers have been working to better understand how this pest will affect stakeholders. They’ve been conducting research on the phenology and voltinism of EZS, exploring novel host associations, and evaluating management techniques. Here’s what they’ve learned so far.

A severely defoliated American elm in Surry County, N.C. Photo by Delaney Serpan

First, the bad news.

Elm zigzag sawfly has recently been found to feed on Japanese zelkova (Zelkova serrata), another common ornamental planting within the Ulmaceae family. However, it is important to note that Japanese zelkova is likely not a preferred host. It is suggested that while EZS can complete its life cycle on Japanese zelkova, it will do so only when no other suitable host is present. Researchers are continuing to explore this novel host association.

But help is on the way! There are management recommendations to control elm zigzag sawfly.

Research conducted at North Carolina State University has determined that soil injections of imidacloprid or dinotefuran at label rate are effective methods to significantly reduce larval populations on infested trees. Both active ingredients are easily accessible to landowners and can provide at least one year of protection against EZS. There is ongoing research to explore more treatment options, including augmentative biocontrol.

What can you do about elm zigzag sawfly?

If you are in an EZS-infested region, check vehicles or outdoor items before moving them.

And if you find EZS, report it! To best manage and prevent the spread of EZS, forest health professionals need to know where it is. Elm zigzag sawfly is the only insect that feeds in the unique zigzag pattern on elm trees. If you see the diagnostic feeding pattern, take a picture of it and contact your county’s local Extension agent or state forestry agency to report it.

Invited blog posted by Delaney Serpan

Delaney Serpan is a second-year Ph.D. student in the Forest Health Lab at NC State University, where she studies elm zigzag sawfly biology and management. She first began working with elm zigzag sawfly as an undergraduate researcher shortly after it was detected in North Carolina for the first time. Working with a novel invasive species on the leading edge of its invasion has been incredibly rewarding. Her work aims to provide accessible management techniques to stakeholders while also protecting elms, an already imperiled species, from further damage.

CISP welcomes 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.

https://fadingforests.org

New thinking on how non-native plants invade forests

It used to be thought that closed-canopy forests are resistant to bioinvasion because of the low light availability and relatively infrequent disturbance. Yet many are badly invaded! (On this site, scroll down past the Archives, choose “invasive plants” category.)

Nor is it just temperate forests in North America. Subtropical and tropical forests have also been invaded, as have the temperate forests of South America and, to a lesser extent, Europe. Temperate forests in Asia are less invaded; boreal forests very little (Fridley et al. 2025; see full citation at the end of this blog).

Fridley et al. (2025) have proposed a conceptual model to explain how this happens: “superinvaders” – a special class of woody plants – that achieve competitive dominance across a wide range of forest conditions. The “superinvaders” pose especially grave threats to native biodiversity because they use life-history strategies unlike those of early successional native species.

The result is that existing invasion and succession theories poorly predict which forests are most invasible and by which species. This failure undermines pest risk analyses and early detection.

Fridley et al. have raised lots of interesting ideas – some of which cannot yet be demonstrated by observations.

Temperate forests of North and South America are increasingly dominated by non-native deciduous and semi-evergreen shrubs and trees that combine fast growth rate in high light and high survivorship in forest interiors. These traits enable them to outcompete the native species. Many invaders also produce many more seeds than co-occurring native species. Similar traits are found in the successful non-native plants in subtropical and tropical forests.

Fridley et al. stress that shade tolerance alone does not endow the invasive plants with sufficiently large advantages in their competition with native species. The forest “superinvader” phenotype must combine this ability to persist in shade with high maximum growth rate and high fecundity when conditions become favorable for reproduction.

They explain the invader’s competitive advantage as the result of their experiencing relatively fewer carbon costs because of enemy release in the novel environment, recent environmental changes that alleviate some stress formerly present in the novel environment, or phylogenetic constraints on the local flora that limit natives’ resource-use efficiency. The non-native plant species enjoy this advantage regardless of whether they also possess other competitive mechanisms, e.g., production of allelopathic compounds, denser growth or shading, greater apparent quantum yield. However, Fridley et al. concede that they lack sufficient evidence to incorporate these other competitive mechanisms into their model.

Since any reduction in carbon costs will enhance both shade tolerance and growth rate when light levels are high, these “superinvaders” can outcompete native species in either situation.

To support these concepts, Fridley et al. note that increased abundance of invaders following disturbance is more pronounced in forests than other habitats. They suggest this is because of the much greater magnitude of change in light levels in forests than in open habitats such as grasslands.

They propose that an analogous situation applies to the presence or absence of mutualist microbial associations, although existing studies are insufficient to reach conclusions about the role of carbon allocation to mycorrhizae in the “superinvader” phenotype. The extent to which these forest invasions alter ecosystem-level carbon dynamics, especially soil processes and litter decomposition is also largely unknown.

Fridley et al. emphasize the role of carbon costs in driving both growth rate and whole-plant light compensation point. This point is defined as the light level at which carbon gain through photosynthesis balances carbon losses from tissue respiration (maintenance and growth) and turnover (shedding and loss from disturbance and herbivory).

To survive in low-light conditions, plants must minimize tissue respiration and turnover. The traits that enable those behaviors have been thought to prevent rapid growth and competitive dominance in high-light conditions. However, the “superinvaders” defy this trade-off by growing faster than most co-occurring native species when light is abundant. Fridley et al. say this is because the plants’ reduced carbon costs enhance competitive advantage in both shade and adequate light conditions.

Fridley et al. name several reasons why a native plant’s carbon costs might exceed those of an introduced species. First on the list is either herbivory or investment in defensive traits. Native plants might be challenged by rising abundance or consumption rates of native or introduced herbivores, such as deer or seed predators, that avoid the introduced species.

A second factor is that the non-native species expends fewer resources to sustain adaptations that confer resistance to other stresses, such as drought or freezing. If a long-standing stress is weakened by global change processes (e.g., atmospheric CO₂ levels, growing season duration, precipitation levels and seasonality, suppression of fire, atmospheric nitrogen deposition), a non-native plant that lacks defenses against that now-weakened stress will have a lower carbon cost and therefore an advantage. In some cases, the non-native species might benefit directly from these changes, e.g., droughts.

In some regions phylogenetic constraints have limited evolution of adaptive solutions to various biotic and abiotic stresses. This is most obvious on tropical oceanic islands. Fridley et al. report that native trees in Hawaiian montane rain forests are less energy-efficient conducting photosynthesis than are the invaders. However, this phenomenon also occurs on continents. Two continents’ floras might experience different climatic histories even when at they are at similar latitudes. For example, Eurasian woody species leaf out earlier and senesce later than North American trees and shrubs – possibly as a result of more predictable spring and autumnal climate across Eurasia. They name as one example Norwegian maple (Acer platanoides) in North America.

The future is uncertain

Fridley et al. consider enemy release to be a key factor in these shrub invasions of closed-canopy forests. Therefore, if enemy release decays over time because the introduced plant species accumulate pests, or the forest environment shifts to favor more stress-tolerant phenotypes of some native species, then the dominance of superinvaders will decline. If, on the other hand, resource enrichment continues, e.g., nitrogen deposition and elevated CO₂, the impacts of woody invaders – present or newly introduced – might continue to rise. The likelihood that additional introductions of more resource-efficient species will continue to damage floras of oceanic islands.

Implications for risk assessments and management

Fridley et al. warn that habitat-matching criteria might be unreliable predictors of forest invasiveness. Among several examples of species that are invasive in interior forest systems in a novel region that do not exhibit this trait in their native range is red oak (Quercus rubra). It is locally dominant in both natural and managed forests in central Europe while in North America, red oak struggles to regenerate in closed-canopy forests. They suggest that Q. rubra in Europe has escaped seedling pathogens present in its native range in North America.

red oak sapling in swampy forest in Virginia; photo by F.T. Campbell

Fridley et al. call for research on traits they have identified as important but that are rarely measured in invasion studies. These include rates of tissue loss and respiratory processes above- and below- ground, plant carbon allocation to tissues and processes, and the whole-plant light compensation points of native and invasive plant species.

The Fridley et al. hypothesis has been supported explicitly by Kinlock et al. (2025). This article says that consistent findings have been reported by earlier small-scale studies in U.S. forests.

I ask for your input on how well the Fridley et al. hypothesis explains shrub and tree invasions in American forests – including those on tropical islands! Is it helpful? Is APHIS incorporating these ideas into plant risk assessments? –

Fridley et al. take pains to reiterate the long-accepted importance of ornamental horticulture in explaining invasive plants’ entry and establishment. They do so in the context of concurring that ruderal traits are not universally advantageous; traits’ benefits depend on the landscape into which the species was being introduced.

SOURCES

Fridley, J.D., P.J. Bellingham, D. Closset-Kopp, C.C. Daehler, M.S. Dechoum, P.H. Martin, H.T. Murphy, J. Rojas- Sandoval, D. Tng. 2025. A general hypothesis of forest invasions by woody plants based on whole-plant carbon economics.

Posted by Faith Campbell

https://fadingforests.org

Urban centers as plant invasion “hotspots” – do global data fail to focus on most important?

*Leucanthemum vulgare* (ox-eye daisy); ranked by EICAT as “major impact”; photo via picryl

Because urban centers are “hotspots” of species introductions and reservoirs supporting their spread into areas less altered by human activity, a global group of scientists (Richardson et al. 2025) sought to determine whether the same plant species naturalize in urban areas around the world and – if so – where most of those plant species originate.

They chose to pursue this question because urban areas share many interacting environmental and biotic features that they thought might partially overcome the distinct biomes of the continents. These shared features include the prominence of impervious surfaces; increased habitat heterogeneity; eutrophication; fragmentation of any remaining semi-natural habitats; complex human-influenced disturbance regimes; diverse opportunities for dispersal; novel biotic assemblages and interactions; and human facilitation of non-native species’ colonization and local species’ extinction. In addition to the similarities of the receiving ecosystems, these commonalities are facilitated by shared introduction pathways – although Richardson et al. to not pursue this aspect.

The scientists consulted global invasive plant databases to compile a list of 7,792 plant species recorded as naturalized in one or more of 553 urban centers on all six continents (all except Antarctica). Just over 300 species (4%) were reported on all six continents. They call them the “omnipresent” taxa. Further refinement resulted in a list of 96 species that are particularly widespread, defined as being present in more than half of the urban centers of Oceania, North and South America, and Europe. These 96 species are present in a lower proportion of cities in Asia and Africa. Richardson et al. proposed that these species be folded into a new ecological category, the “urban florome”.

I wonder whether this set of species tells us more about biases in the data than the actual “urban florome”. First, 87% of the 96 “most widespread” species (n= 84) are annual or perennial herbs. Only seven tree, six vine, and six shrub or subshrub species were included among the 96 species. In other words, global lists of invasive species are heavily slanted toward species that thrive in disturbance. Is this surprising? As another study (Kinlock et al. 2025) notes, disturbance is ubiquitous!

Second, only a third of the “urban florome” species have been formally evaluated using the Environmental Impact Classification for Alien Taxa (EICAT) system. Of these 32 species, only six were categorized as having a “major” or “massive” impact. Richardson et al. (2025) conclude that many of the species on the most widespread list are human commensals that have few or negligible known impacts.

Still, this finding might underestimate their impacts. First, as noted, two-thirds have not been evaluated. Second, impacts important in urban systems might not be those that increase a species’ rank based on impacts to natural systems (Richardson et al.). Those with substantial nuisance value in the urban setting still require management. Of course, some of the species have severe impacts in both natural and urban ecosystems. For example, Ailanthus altissima causes major infrastructural damage and pollen allergies, while Robinia pseudoacacia alters soil fertility. Both reduce species richness.

I note that these examples are both trees – which constitute only 7 of the 96 species. Fridley et al. report that trees and shrubs have severe impacts in closed forest systems. I suggest that since many of the urban areas in temperate, subtropical, and tropical regions are probably located in formerly forested areas, remnant (semi-)natural stands and even recreational parks have probably been invaded by these high-impact species. Surely that is more important – at least as regards the level/intensity of the non-native plant species’ impact on biodiversity – than the annual weeds growing along highway verges.

Richardson et al. fear that many cities also have substantial invasion debt. The note specifically that due to the heat island effect, species that can now survive only in cities are likely to spread into surrounding rural and natural areas as temps increase. Thus, these species amplify the urban source effect of plant invasions.

Generalities

Richardson et al. call attention to certain parts of the world acting as ‘factories’ for the evolution of plant species that are well equipped to become invasive when intro to new regions. They name Australian woody flora — although only one species, Melia azedarach, is included among the 96 most widespread species. They also name African grasses and Europe (no taxa specified).

Richardson et al. say that while non-native species in urban areas have usually been described as “passengers” taking advantage of anthropomorphic environmental change, bioinvasions are increasingly recognized as drivers of secondary changes that alter the capacity of these ecosystems to deliver key ecosystem services, or even create disservices. These modifications occur in urban as well as more natural environments.

Regional Differences

Richardson et al. developed lists of the most widespread naturalized urban species for each continent (‘continental lists’). Eighty-seven percent of the 96 “most widespread” species are present in cities of North America, 80% in cities of Oceania, and 34% in European cities. Only 17% of the “widespread” species are present in cities of South America, 13% in cities of Africa or Asia.

While there is considerable overlap regarding species found on several continents, Europe’s urban florome differed significantly from those of the other continents.

The principal source region for these naturalizing species was temperate Asia (145 records); followed by Europe (128 records) and Africa (121 records). Lower numbers came from tropical Asia (95 records); South America (54) records; North America (53 records); and Oceania (8 records). Europe has received 50% of its widespread urban invasive species equally from temperate Asia and North America. Africa has received 75% of its widespread urban species from the two Americas equally.

According to these data, Oceania has been a significant contributor only to South America. I am surprised given the publicized problems caused by Australian Acacia and Hakea in South Africa. I guess these trees are more invasive in the vicinity of urban areas rather than in the cities themselves.

Richardson et al. note a highly skewed relationship between North and South America: while 15.4% of species naturalized in South American cities come from North America, only 2.7% of naturalized species in North American cities are from South America.

*Lepidium didymum* – brassica from South America introduced widely, including throughout California; photo by Miguel A.C. via Pl@ntnet

Richardson et al. found a distinct division between the “Old” and “New” Worlds (defined by whether the soil was historically cultivated by plough vs. hoe). The latter has more naturalized species (9,905 taxa vs 7,923 taxa), although the “Old World” covers a larger area. Citing di Castri (1989), they suggest that the much longer history of intense human-mediated disturbances in Europe might have allowed its flora to adapt to coexist w/ humans. I wonder, however, whether it is just too difficult to distinguish introductions that occurred millennia ago.

Richardson et al. also found an “echo” from European colonization — strengthened by activities of acclimatization societies. The result is that the continents with longer histories of European colonization, i.e., South and North America and Oceania, have more widespread naturalized plant species than do Africa and Asia.

SOURCES

Kinlock, N.L., D.W. Adams, W. Dawson, F. Essl, J. Kartesz, H. Kreft, M. Nishino, Jan Pergl, P. Pyšek, P. Weigelt and M. van Kleunen. Naturalization of ornamental plants in the United States depends on cultivation and historical land cover context. Ecography 2025: e07748 doi: 10.1002/ecog.077

Richardson, D.M., L.B. Trotta, M.F.J. Aronson, B. Baiser, M.W. Cadotte, M. Carboni, L. Celesti-Grapow, S. Knapp, I. Kühn, A.C. Lacerda de Matos, Z. Lososová, D. Li, F.A. Montaño-Centellas, L.J. Potgieter, R.D. Zenni, P. Pyšek. 2025. Here, There and Everywhere: Widespread Alien Plants in the World’s Urban Ecosystems. Global Ecology and Biogeography, 2025; 34:e70159 https://doi.org/10.1111/geb.70159

Posted by Faith Campbell

https://fadingforests.org

The invasive risk of Eucalypts

*Eucalyptus grandis* (in Australia); photo by Poyt448 Peter Woodard via Wikimedia

Deus et al. 2025 (full citation at the end of this blog) have published a review of current knowledge on the invasiveness of trees in the Eucalyptus genus. They report that eucalypt plantations cover more than 30 million ha globally; they could not determine the actual extent more precisely. The area is expanding at an estimated 4% per year. Eucalypts are so popular as timber trees because of their fast growth, ease of management, wood quality and environmental tolerance.

Until recently, trees in the Eucalyptus genus were thought to pose a low invasion risk. This was because these trees have limited seed dispersal, high juvenile mortality, and were expected to lack compatible ectomycorrhizal fungi in novel environments. However, several risk assessments and reports of ongoing invasions in some locations have raised questions. So Deus et al. undertook a literature survey to try to resolve the issue.

One of the risk assessments concerns the United States; see Gordon et al. (2012). This study – completed a dozen years before Deus et al. undertook their literature survey – cited several other sources documenting harmful invasiveness of nearly a dozen species, including Eucalyptus globulus, E. camaldulensis, E. grandis, and E. tereticornis.

Deus et al. found that the limitations listed above actually can be overcome, so they do not prevent invasions:

seeds can disperse farther than 100 meters from parent plants;
high recruitment densities can compensate for the high juvenile mortality; and
ectomychorrhizal fungi can be found in the root systems of introduced eucalypt plants.

In fact, several Eucalyptus species meet criteria defining invasiveness in the Australian Weed Risk Assessment system. Still, Deus et al. found that existing studies cover too few plantations and species to allow an in-depth comprehensive understanding of eucalypts’ invasion ecology.

One reason that eucalypt trees’ invasiveness remains unresolved is that the countries which have established most large Eucalyptus plantations (Brazil, India and China) have not conducted many studies. Instead, most studies have been done in Iberia and South Africa, which together host less than six percent by area of the world’s estate of eucalypt plantations.

Deus et al. say that several possible reasons have been proposed to explain why Eucalyptus is considered to pose an invasion risk by scientists in Iberia and South Africa, but not in Brazil.

The few studies in Brazil were conducted in intensively managed plantations with very short rotations, which are probably less prone to invasion than plantations managed at low intensity levels.
The Brazilian plantations were established 40 to 50 years ago, whereas those in Iberia were introduced ~ 200 years ago.
Iberia experiences recurrent forest fires.
In Brazil, leaf eating ants attack the trees; this might reduce trees’ vigor.
In Brazil, native forests dominate the environs.

Deus et al. say that these hypotheses have never been tested.

Since studies have been conducted in only a few countries, they have evaluated only a few of the species used in plantations. At least 372eucalypt species have been introduced outside their native range; nine species are planted widely. Yet most of the studies reviewed by Deus et al. covered just two species, Eucalyptus globulus (46% of the studies), and E. camaldulensis (33% of the studies). Still, these two widely cultivated species received the highest invasiveness ranking of all species reviewed (65 and 72, respectively). According to Deus et al., these scores are higher than the average score for 32 species of Acacia – a genus considered to be one of the most invasive tree genera in the world.

Other, potentially invasive species, have not received adequate attention. Deus et al. note that E. tereticornis, which is widely planted in China, India and other regions of Southern Asia, has an invasiveness score of 66, placing it second highest in the evaluation. However, only 12 of 140 articles analyzed by Deus et al. addressed this species.

These eucalypts’ high scores result from their potential to hybridize, to naturalize outside their natural habitat, and from high flammability. Other contributing factors are high seed production and ability to resprout after cutting or fire.

The analysis determined that the major drivers for Eucalyptus invasions are soil disturbance, availability of moisture (essential for seedling establishment), and fire. Recruitment density increases with harvesting and tree age; it decreases when the understory is managed. This partially explains why the abandonment of plantations might promote invasions by eucalypts.

Deus et al. fear that there might be a large “invasion debt” in the regions where few studies have been conducted. Assessments for California and Iberian Peninsula indicate that the best areas for cultivation – under either current conditions or expected new environments linked to climate change – are also those most prone to invasion. A further complication is that in some regions it might be difficult to distinguish plants escaping from small plantations from the plantations themselves. They suggest ways to overcome this difficulty: 1) surveys of recruitment along roadside, where trees would not have been planted; 2) genetic analysis of seedlings and possible parents

Another weakness is that that none of the studies considers changes in fire regime, which probably increases the areas prone to invasion.

Deus et al. think it is unlikely that eucalypt invasions will turn out to be as damaging as those of acacias or pines, but that further invasions involving more species and more regions are very likely.

Deus et al. call for considering eucalypt species’ potential invasiveness when developing strategies for the sustainable management of these plantations, including how to manage those that are no longer economically viable.

Status in the United States

The risk in the United States was evaluated by Gordon et al. in 2012. At the time, there were proposals to plant 5,000 to 10,000 ha/year in the Southeast over the next decade.

Gordon et al. adapted the Australian weed risk assessment system to evaluate 38 Eucalyptus taxa then being tested and cultivated in U.S. for pulp, biofuel, and other purposes. Their analysis concluded that 15 of these taxa posed a low risk; 14 taxa posed a high risk; and 9 taxa could not be ranked without further information. The four taxa cultivated most extensively – E. globulus, E. camaldulensis, E. grandis, and E. tereticornis – all had high risk outcomes, as did several other taxa. Gordon et al. thought that these differences reflected both new data and differences in how the assessors reacted to insufficient data.

Gordon et al. warned that novel genotypes with unknown invasiveness were being propagated in the search for increased cold tolerance. This meant that the taxa they had assessed might not indicate of the actual long-term invasion risks associated from this genus. A major source of uncertainty is the long lag time in appearance of evidence of a tree species’ invasiveness. Only one study (as of 2012) had quantified lag time for introduced tree species; it found an average of 170 years from the time introduction to identification of the taxon as invasive. Propagule pressure also influences the lag time and the probability of invasion.

Since the bulk of expanded cultivation was expected to be in the southeast, Gordon et al. recommended that a regional assessment be conducted to more precisely specify the effects of possible differences in phenology, age at reproductive maturity, seed viability, and cold tolerance.

Gordon et al. suggested several actions to reduce the invasion risk. First, selection and breeding strategies could aim to minimize relevant traits – especially eliminating seed production. Second, plantations could be so managed by avoiding cultivation near waterways, harvesting stems before seeds can mature, and restricting the extent of cultivation of any one taxon. More broadly, a fund could be established to cover control costs; growers would contribute the money.

What has happened in the dozen years since the analysis was published? My Google search led to publications from 2013 and earlier. I hope this indicates that no one has funded major expansions. Dr. Gordon reports that most Eucalyptus pulp is imported. ArborGen continues to breed Eucalyptus in Brazil – as I noted earlier, scientists there are not pursuing studies of possible invasiveness of eucalypts.

Still, the regional risk assessment has not been conducted. Worse, Dr. Gordon reports that the Florida Department of Agriculture and Consumer Services has exempted several species [E. amplifolia, E. benthamii, E. dorrigoensis, E. dunnii, E. grandis, E. gunni, E. nitens, E. smithii, and E. urograndis (E. grandis E. urophylla)] from a requirement that growers obtain Non-Native Species Planting Permits. So if the market does take off, there will be no regulation by the state.

At the end of December 2025, Dr. Gordon received information from Florida Division of Plant Industry that no one has applied for a permit to grow Eucalyptus in the state other than under USDA research auspices. So my worst fears have not (yet) come to pass.

I note that in 2022, Potter, Riitters, & Guo ranked Eucalyptus grandis & E. globulus as potentially highly invasive. Their criterion was that at least 75% of stems detected by USFS Forest Inventory and Analysis (FIA) surveys were saplings or seedlings.

SOURCES

Deus, E., D.M. Richardson, F.X. Catry, F.C. Rego, J. Gaspar, M. Nereu, M. Larcombe, B. Potts, J.S. Silva. 2025. Invasion ecology of eucalypts: a review. Biol. Invasions (2025) 27:239 https://doi.org/10/1007/s10530-025-03695-1

Gordon, D.R.,S.L. Flory,.L. Cooper, and S.K. Morris. 2012. Assessing the Invasion Risk of Eucalyptus in the United States Using the Australian Weed Risk Assessment. International Journal of Forestry Research Volume 2012, Article ID 203768, 7 pages doi:10.1155/2012/203768

Potter K.M., Riitters, K.H. & Guo, Q. 2022. NIS tree regeneration indicates regional & national risks from current invasions. Frontiers in Forests & Global Change

doi: 10.3389/ffgc.2022.966407

Posted by Faith Campbell

https://fadingforests.org

Welcome high-level attention to bioinvasions – although key issues remain unresolved

Japanese knotweed (*Reynoutria japonica*) – one of the worst invaders around globe. Photo by Will Parson, Chesapeake Bay Program via Flickr

On 23 October, Science published a five-page, data-packed analysis of bioinvasion impacts on terrestrial ecosystems!!!

Thakur, Gu, van Kleunen, and Zhou (full citation at end of this blog) analyzed 775 studies with the goal of improving understanding of factors contributing to invasions’ impacts – as distinct from “invasibility” (ability to establish). This knowledge is essential to assessing the risk posed by introduced species and setting priorities for management. They analyzed five ecological contexts—diversity of native species and introduced species in the recipient systems, latitude, invader residence time, and invader traits.

They concluded that ecological factors commonly used to explain invasion success do not consistently translate into strong predictors of invasion impacts. Impacts vary in response to the context of the invasion.

[In January 2026, the authors announced changes in details of the article due to some errors in the database and their understanding of it. (Science 8 Jan 2026 Vol. 391 Issue 6781) They conclude that the corrected analysis did not alter the trends described or the overall conclusions.]

limber pine (*Pinus flexilis*) – one of the species killed by *Cronartium ribicoli*; photo by F.T. Campbell

Among the studies available for analysis, reports on plants dominated: 605 focused on plant invasions, 114 on animal invasions, and only 56 on microbial invasions. Among the animals were one study of Adelges tsugae (hemlock woolly adelgid), two studies of Agrilus planipennis (emerald ash borer) and one study each of Lymantria dispar (spongy moth), and Ips pini (North American pine engraver). Studies also addressed earthworms, ants, rats, and feral hogs. Microorganisms included Cronartium ribicoli (white pine blister rust) and several Phytophthora species, including P. agathidicida (kauri dieback), P. alni (affects alders), and P. ramorum (sudden oak death).

Thakur et al. note the skewed taxonomic coverage and say that the low number and narrow taxonomic/ecological variety in the animals and microorganisms probably limit their ability to reach robust conclusions about the impacts of such invasions.

The most consistent negative impact they found is reductions in native plant diversity. While this is not surprising given the studies analyzed, I think it is still important since it counters the widespread sense that plant invasions are somehow less deserving of a robust response.

The authors also detected some broader ecosystem impacts of plant invasions. Plant invasions increased soil organic carbon; soil nitrogen (ammonium and nitrate), and available phosphorus; soil moisture, litter biomass; and emissions of carbon dioxide (CO₂), nitrous oxide (N₂O), and methane (CH₄). The changes in biogeochemical properties might reinforce impacts on native plant communities. The reported increase in greenhouse gas emissions might reflect a bias in the studies so Thakur et al. call for more research to solidify this finding.

High native plant species richness had only a weak overall effect on ecosystem-level impacts. While plant invasions often resulted in higher overall plant species richness, when considering only native community responses, the gain in species numbers did not necessarily indicate conservation benefits. Native plants’ biomass increased after invasion. This might reflect short-term increases in productivity in response to altered resource conditions or structural facilitation, rather than a long-term reversal of competitive exclusion. Finally, the longer the invasive [plant] species had been present, the greater the negative effects on native diversity. However, soil abiotic property impacts weakened over time. In fact, the initial increase in soil organic carbon and total nitrogen disappeared after 6 to 10 years. This development might reflect fertilization of ecosystems by long-established nitrogen-fixing invaders such as non-native legumes.

Traits of non-native plant species related to growth and resource acquisition were overall weak predictors of ecosystem impacts. Thakur et al. consider that this finding reflects the relatively narrow range of specific leaf area exhibited by the plant species studied most commonly.

Consequently, Thakur et al. urge managers to focus on containment and impact mitigation, and to prioritize persistent losses of native plant diversity. When considering abiotic responses that might lessen over time, managers should apply “adaptive monitoring” (which is not defined).

Thakur et al. had greater difficulty determining the impacts of animal and microorganism invasions because of the smaller number of studies. They could not determine the effect of native species richness. The observed decline in soil organic carbon they thought was attributable to the large proportion of studies (9 out of 114) that focused on introduced earthworms. Earthworms reduce organic matter by consuming litter. Mammals were also found to reduce soil organic carbon. Introduced insects had no significant ecosystem effects on soil organic carbon. Non-native animals also increased soil emissions of carbon dioxide and nitrous oxide. The microorganisms included in reviewed studies decreased soil ammonium and increased nitrate, consistent with elevated nitrification. While data on body size of invasive animals were sparse, the authors could determine that larger-bodied species tended to increase soil nitrate while reducing effects on total soil N.

Applying the Results

Thakur et al. report that residence time outperformed other factors as a predictor of invasion impacts. The authors regret the scarcity of long-term studies, especially in the Global South, that could increase our understanding of whether these impacts persist or shift under sustained invasion pressure.

How can scientists apply this information in risk assessments evaluating not-yet introduced species or in deciding what is the appropriate intensity of immediate response to newly detected incursions. Should they give greater weight to others’ studies that focus on long-established invasions by the species in question? Otherwise, this finding seems to largely duplicate the long-established “invasion curve”.

I hope scientists will note that observational studies generally showed stronger impacts than experimental ones, particularly in the case of plant invasions. Perhaps this is true because observational studies better incorporate environmental heterogeneity and longer time spans.

Agrostis stolonifera – one of the plants invading on Prince Edward Island, an Antarctic region island under South African jurisdiction. Photo by Stefan Iefnaer via Wikimedia

Thakur et al. note that one factor they analyzed, “latitude”, incorporates several ecological and anthropogenic components relevant to invasion impacts. One element is the greater native bioidiversity in warmer, lower-latitude, regions. According to the “biotic resistance” hypothesis, greater diversity might make these systems more resistant to bioinvasion. However, the situation is complicated by the fact that temperate regions have also often experienced longstanding and intensive land-use modifications — which are believed to facilitate invasive species establishment and spread. I regret that the authors make no attempt to separate the effects of factors that are anthropogenic from those arising from immutable conditions, e.g., latitude, topography, weather patterns, etc.

Thakur et al. call for more studies that cover a wider geographic range. In addition, the studies should include more experimental designs and explore the relationship between invaders’ traits and impacts — especially regarding animals and microbes.

SOURCE

Thakur, M.P., Z. Gu, M. van Kleunen, X. Zhou. 2025. Invasion impacts in terrestrial ecosystems: Global patterns and predictors. Science 23 October 2025

Posted by Faith Campbell

https://fadingforests.org