Introduction:
Biological control is a safe and cost-effective approach for the
landscape-wide management of weedy species (Van Driesche et al., 2010).
On numerous occasions, insects and pathogens have been identified and
promoted to reduce the environmental and economic damages of weeds in
locations from across the globe (Van Driesche, 2012). Despite concerns
about the non-target effects of biological control agents (Barratt et
al., 2010; Howarth, 2000; Simberloff, 2012), modern biological control
programs implement a system of safe-guards to reduce unwarranted damage
to non-target species – particularly when the agent is being introduced
from a novel region through a practice known as classical biological
control (Heinz et al., 2016; Messing, 2001). As a result, the use of
biological control as an alternative to other labor and chemically
intensive methods is increasingly becoming a part of both conservation
and organic management practices (Baker et al., 2020; Van Driesche et
al., 2016)
The control of invasive knotweed species, Reynoutria spp. Houtt.
(Caryophyllales: Polygonaceae), has received much attention, with
cultural, mechanical, and chemical control options all being implemented
(e.g., Delbart et al., 2012; Kadlecová et al., 2022; Martin et al.,
2020). Interest has also been directed towards harvesting knotweeds, as
the plants have unique chemical properties (Metličar & Albreht 2022;
Metličar et al. 2021) and may themselves be an important source for
biopesticides (Dara et al. 2020). However, for landscape-wide efforts
biological control is likely the most effective strategy, and as such an
international effort was established to identify and promote the natural
enemy, Aphalara itadori (Shinji) (Hemiptera: Aphalaridae), which
was observed feeding and causing damage upon wild populations ofR. japonica Houtt. on the Japanese island of Kyushu in 2004 (Shaw
et al., 2009). Prior to field releases, a laboratory reared population
of the Kyushu strain of A. itadori was then used for host-range
testing and candidate biological control reviews were conducted
(Grevstad et al., 2013; Shaw et al., 2009; Shaw et al., 2011), resulting
in the first approved biological control agent in the European Union
(Shaw et al., 2009). A second population of A. itadori , feeding
on R. sachalinensis (F. Schmidt) Nakai, was subsequently
collected in 2007 near Lake Toya on the Japanese island of Hokkaido, and
similarly brought to the laboratory for host range testing and candidate
biological control review (APHIS, 2020). Both strains were subsequently
approved for release in Europe and North America, and recently a third
strain, the Murakami strain, was identified from near the Japanese city
of Murakami and has been released in the Netherlands against R. ×bohemica Chrtek & Chrtková (Camargo et al., 2022). Review of the
Murakami strain for release in North America is currently underway.
As part of the review prior to introduction in North America, climate
suitability models for the Kyushu strain and the Hokkaido strains were
developed using the software program CLIMEX (Hearne Software, Melbourne,
Australia). These models predicted a strong climate match for both the
Kyushu and Hokkaido strains to potential release locations across North
America (Grevstad et al., 2012). However, despite this predicted climate
match, there have been no documented accounts of establishment of this
species anywhere it has been released. Note: here we use establishment
to indicate a self-sustaining population that is present in a location
for at least three consecutive years without importation or release of
additional individuals. We choose to use this more conservative
definition, though “establishment” has historically been reported in
the literature after only one year (see Van Driesche et al. 2008).
Unfortunately to date in locations where the psyllids have been released
and individuals have been observed in the field during post-release
monitoring, neither reduction in plant densities nor biomass have been
observed. We previously suspected that environmental constraints might
be limiting the success of the Kyushu strain in North America (Andersen
& Elkinton, 2022), and noted a poor climate match to the source
locality of the Kyushu strain based on North American records ofR. japonica using a different modeling approach, MaxEnt (Phillips
et al., 2006; Phillips & Dudik, 2008). The MaxEnt models based onR. japonica records did, however, predicted medium-to-high
suitability for the source localities of the Hokkaido and Murakami
strains (Andersen & Elkinton, 2022). Therefore, we were curious as to
whether climate suitability estimates conducted in MaxEnt for each
target knotweed species based on records from Europe and North America
could help provide insights into factors influencing the success ofA. itadori releases in each region and for each target species of
knotweed.
To address this, we collected public records of all three target
knotweed species from the Global Biodiversity Information Facility
(GBIF) from Europe and North America. Using the MaxEnt software
platform, we estimate climate suitability envelopes based on records
from the invasive regions of each species, and we compared the predicted
suitability of the source locality of each A. itadori strain
compared to other localities in Japan for each species of knotweed and
geographic region combination.