Complete host specificity test plant list and associated data to assess host specificity of Archanara geminipuncta and Archanara neurica, two potential biocontrol agents for invasive Phragmites australis in North America

Introduced European genotypes of Phragmites australis are invasive and widespread in North America. Decades of management using herbicide and other means have failed to control the species and its range and populations continue to expand. Allowing continued invasion threatens native wetland biota and an endemic North American subspecies Phragmites australis americanus. The lack of conventional management to control introduced P. australis triggered research to assess host specificity of two European noctuid moths, Archanara geminipuncta and Archanara neurica. These two species are considered particularly promising potential biocontrol agents for introduced P. australis. Here we provide the complete and approved list of test plants used to assess host specificity of A. geminipuncta and A. neurica. This includes data on neonate larval acceptance and survival under no-choice conditions, and oviposition tests for all plant species tested, including for different Phragmites subspecies currently occurring in North America. We further provide temperature profiles of select cities in the temperate native European distribution of the two noctuids and those in southern US climates. We used these long-term temperature records to assess whether overwintering eggs of A. geminipuncta and A. neurica can survive under climate conditions typical for the Gulf Coast region in North America. This data article refers to “Host specificity and risk assessment of Archanarageminipuncta and Archanaraneurica, two potential biocontrol agents for invasive Phragmitesaustralis in North America Biol. Control (2018)”.


a b s t r a c t
Introduced European genotypes of Phragmites australis are invasive and widespread in North America. Decades of management using herbicide and other means have failed to control the species and its range and populations continue to expand. Allowing continued invasion threatens native wetland biota and an endemic North American subspecies Phragmites australis americanus. The lack of conventional management to control introduced P. australis triggered research to assess host specificity of two European noctuid moths, Archanara geminipuncta and Archanara neurica. These two species are considered particularly promising potential biocontrol agents for introduced P. australis. Here we provide the complete and approved list of test plants used to assess host specificity of A. geminipuncta and A. neurica. This includes data on neonate larval acceptance and survival under no-choice conditions, and oviposition tests for all plant species tested, including for different Phragmites subspecies currently occurring in North America. We further provide temperature profiles of select cities in the temperate native European distribution of the two noctuids and those Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/dib in southern US climates. We used these long-term temperature records to assess whether overwintering eggs of A. geminipuncta and A. neurica can survive under climate conditions typical for the Gulf Coast region in North America. This data article refers to "Host specificity and risk assessment of Archanara geminipuncta and Archanara neurica, two potential biocontrol agents for invasive Phragmites australis in North America Biol. Control (2018) We provide a comprehensive list and results for all test plant species used to assess host specificity (oviposition and larval development) for Archanara geminipuncta and Archanara neurica. We provide data on larval survival, which provides a full overview of the selectivity of these moth species, and hence the safety of other wetland plants.
We provide data on temperature profiles we used to assess the possibility of winter survival under different climate conditions for select locations in Europe and North America. This allows an assessment of the ability of A. geminipuncta and A. neurica to colonize regions with different climate conditions in North America based on their current distribution in Europe.

Data
The approval of specialized herbivores as biological weed control agents in North America requires extensive host specificity testing and federal (US and Canada) review [2]. Procedures to select appropriate test plant species are largely standardized and our host-plant selection was based on plant phylogeny, species of conservation or agricultural concerns, plants that are attacked by Table 1 TAG Category 1 , plant species used to determine host range of Archanara geminipuncta and Archanara neurica, potential biocontrol agents of invasive Phragmites australis, initiation of larval feeding or survival (AG indicates A geminipuncta, AN indicates A. neurica) and plant distribution according to USDA Plants Database in North America [1]. Initiation of larval feeding would result in follow-up tests [6]. (       Pontederia cordata L. any plant on which the biological control agent or its close relatives have been found or recorded to feed and/or reproduce. Our two selected potential weed biocontrol agents A. geminipuncta and A. neurica are widespread in Europe, but neither moth is found where average temperatures exceed 15°C [4]. However, in North America the host plant is currently distributed throughout most of the US, including regions of the Gulf Coast with a distinctly different climate. We tested whether southern climate conditions as they may exist in Florida or in the Mississippi River Delta would allow successful overwintering of eggs. To compare winter survival, we selected long-term winter temperature profiles of cities located within the native European range of the species and southern US locales (Table 2).

Experimental design, materials and methods
Detailed life-histories of our two experimental organisms A. geminipuncta and A. neurica are well described [5]. Eggs overwinter under leaf sheaths with tightly synchronized larval emergence just as P. australis shoots begin to elongate in spring. Larvae feed internally in the soft, nutrient-rich tissues within P. australis shoots above the growing point. Larval development takes several weeks and mature larvae pupate in lower sections of stems. Adults fly in late July and August with A. neurica occurring about two weeks before A. geminipuncta.

Larval acceptance
We conducted our experiments on larval acceptance of different test plant species in quarantine at the University of Rhode Island (URI) and in Europe at CABI in Delémont, Switzerland (CABI). We either purchased or field collected test plant species and propagated them in common gardens at either location. We obtained eggs and larvae of A. geminipuncta and A. neurica from captive colonies maintained outdoors at CABI (source of the captive colony were adults and larvae originally collected near CABI). At URI we kept eggs in an incubator (4°C) before bringing them to room temperature to stagger tests in the spring easing logistical and space constraints in quarantine, which enabled us to synchronize larval hatch and plant growth. At CABI we also tested a subset of crop and wetland plants available in Europe.
In all our different no-choice tests we used a fine paint brush to transfer neonate larvae onto cut shoots or potted test plants using 6 -15 replicates per test plant species in gauze cages or variable sizes [6]. Since these tests needed to be conducted over many years, we simultaneously assessed validity of each test using P. australis shoots or plants as controls using identical replication. Depending on testing conditions, we assessed larval survival, recording feeding marks, entrance or  Table 1 and were advanced to more detailed tests, including female oviposition choice, or larval discrimination in multiple-choice tests and results are reported elsewhere [6].

Oviposition choice experiments
In our oviposition tests we progressed from single-to multiple-choice tests using cut shoots ( Fig. 2A; [6]) or potted plants (Fig. 2C, [6]) of European or North American P. australis or native P. australis americanus in small (40 Â 40 Â 65 cm, Fig. 2 A; [6]) or large gauze cages (2 Â 2 Â 1.6 m; Fig. 2C; [6]). We conducted these tests over multiple years (2003)(2004)(2005) in accordance with availability of plant material or adult moths. We released a single A. geminipuncta or A. neurica pair in each test using cut shoots, and exchanged shoots and recorded all eggs laid every two days. We released five mated A. geminipuncta or A. neurica pairs for each test using potted plants in late July or early August as adults became available. We examined each shoot and recorded the number of eggs laid on each plant in early September.

Egg survival under southern climates
The distribution of P. australis berlandieri in the U.S. is restricted to areas south of 35°latitude, while the two moths occur only north of 35°latitude in Europe [4]. To investigate the potential of the two moths to establish in climates where P. australis berlandieri currently occurs, we set up an egg overwintering experiment with A. geminipuncta and A. neurica at URI in October 2017. We set incubator (Percival I-36LL, Percival, Perry, Iowa, USA) conditions to photoperiod and fluctuating average day and night temperatures of Fort Pierce, Florida and Basel, Switzerland [6]. We reprogrammed incubator conditions twice a week to follow seasonal changes at Fort Pierce and Basel (Table 2). We started to check for larval emergence weekly starting in February 2018, until we terminated the experiment in May 2018 [6].

No-choice, single-choice and multiple-choice oviposition tests at CABI
In no-choice oviposition tests in 2003, females accepted all plants for oviposition but laid almost three times as many eggs on European P. australis compared to native P. australis americanus (Table 3). Females retained eggs until death rather than ovipositing them on apparently less suitable hosts, including native P. australis americanus (Häfliger, unpublished data). Under these no-choice conditions, both A. donax and P. arundinacea were also accepted for oviposition with egg numbers lowest for A. donax (Table 3). In the multiple-choice oviposition test using potted plants we found similar results with no oviposition on native P. australis americanus, A. donax and P. arundinaceae (Table 3). However, the total number of eggs laid by five females was only 61, and we found 10 eggs on T. latifolia (Table 3) questioning whether tests of this type can deliver reliable results. Tests with increased replication in 2004 confirmed the high (but this year not absolute) avoidance of native P. australis americanus by A. geminipuncta. When we provided females a choice between cut shoots of European P. australis and native P. australis americanus, we found eggs exclusively on European P. australis (Table 4). When we offered females cut shoots of European P. australis, North American P. australis and native P. australis americanus, the number of eggs laid on native P. australis americanus was substantially lower than the number of eggs on introduced North American P. australis (Table 4), but only slightly lower than on European P. australis (which was the only lineage attacked in singlechoice tests in 2003; Table 3). In our field cage multiple-choice oviposition tests with potted plants, females showed strong discrimination between European P. australis and native P. australis americanus in some years, but not others (Table 4), differences we are unable to explain.
In 2005, when we focused on discrimination between the introduced and native North American Phragmites lineages, we found no difference in the number of eggs laid by A. geminipuncta on introduced P. australis vs. native P. australis americanus using cut shoots, but tests with A. neurica showed a distinct preference for the introduced lineage (Table 4). Interestingly we found the reverse pattern for the two species in our field cages in which A. geminipuncta showed a strong, although not absolute, preference for introduced P. australis over native P. australis americanus, while A. neurica females distributed their eggs evenly among different lineages (Table 4).