Aphid Biological Control in Greenhouses with a Novel Banker Plant System

Progress report for ONE20-380

Project Type: Partnership
Funds awarded in 2020: $26,229.00
Projected End Date: 07/31/2022
Grant Recipient: Cornell University
Region: Northeast
State: New York
Project Leader:
Dr. John Sanderson
Cornell University
Expand All

Project Information


Banker plants provide resources for arthropod natural enemies when target pests are scarce, enabling in-greenhouse reproduction of the natural enemies for continuous production and reducing or eliminating the costs of otherwise-required weekly natural enemy shipments. There are banker plant systems for small aphid pest species and for large aphid pest species. But they are not compatible and cannot be used in the same greenhouse simultaneously because they both use aphid species that only infest barley as the host plant. This common host plant allows the aphid species from one system to eventually displace the aphid species of the other system. This project will evaluate the use of a novel banker plant system, using pea aphids (a large aphid) on potted fava bean plants as hosts for the parasitoid Aphidius ervi. Because this novel system uses fava bean rather than barley, it would be compatible with the other barley-based banker plant system for small aphids. Experiments will be done in research greenhouses to determine optimal timing and production of the fava beans, the pea aphids, and the parasitoids. The resulting banker plant system will be evaluated in a commercial greenhouse during Spring ornamental crop production. Thus, this new banker plant system should make it possible for both A. colemani and A. ervi to co-exist, reproduce, and provide successful cost-effective biocontrol of nearly all greenhouse aphid pests. If successful, a fact sheet for growers will be prepared and used in grower presentations that will outline the details of the system.

Project Objectives:

This project seeks to develop and test a cost-effective, relatively simple system for in-greenhouse production of a parasitoid species, Aphidius ervi, that attacks larger pest aphids such as foxglove or potato aphid, using non-pest pea aphids on fava bean plants.  The project will combine research to develop the system with a commercial trial to test the system. Three research objectives will be done. The first will determine the best growth stage of fava bean plant to infest with pea aphids to balance maximum aphid production with minimal plant decline. The second will check that Aphidius colemani parasitoids, commonly used for small aphid biocontrol, do not outcompete A. ervi on the fava bean banker plants with pea aphids. The third will evaluate the pattern of parasitoid production, both in numbers and over time, when aphid-infested fava beans are exposed to the parasitoids at various growth stages. Using research results, we will trial the system in a commercial greenhouse on Spring crops. A fact sheet about the system will be prepared for growers. The successful system will give growers a way to simultaneously produce two parasitoid species, one for large aphids and another for small aphids, providing complete biocontrol of aphids.


Crop loss from cosmetic damage by greenhouse pests can occur quickly, so greenhouse flower growers apply more pesticides per cubic meter than any other commodity to protect their crops (Smith 1998). Such intense pesticide use is an unsustainable approach and leads to several pesticide-related problems. Biological control of greenhouse pest such as aphids has been shown to be an effective alternative and the popularity of biocontrol has been surging among NE greenhouse growers. However, the required weekly, inundative releases of commercial natural enemies can be expensive for many growers. Banker plants provide a cost-effective alternative. Banker plants are an open rearing system for natural enemies which provide alternate non-pest food sources for the natural enemies when target pest levels are low, so that natural enemies can reproduce in the greenhouse. This avoids the need for repeated parasitoid purchases and shipping costs.  Banker plants of potted barley infested with non-pest Rhopalosiphum padi aphids are commonly used to sustain populations of an aphid parasitoid, Aphidius colemani. But A. colemani only attacks small-sized aphids, and greenhouse crops are subject to a complex of aphid pests, including important large species such as foxglove and potato aphid.  A second parasitoid, Aphidius ervi, is an effective parasitoid of large-sized aphids, and a potted barley banker plant system for these is also available, using the non-pest aphid Sitobion avenae. However, the two banker systems are incompatible because the host aphid for one overtakes the host aphid for the other if they co-occur in the same greenhouse, because both banker plant systems use the same plant species. We propose to develop and trial a novel banker plant system for A. ervi, using potted fava bean plants infested with pea aphids. This system uses a different plant species than barley (i.e., fava bean) to avoid the non-pest aphids from interfering with each other. Pea aphids do not survive on barley, and R. padi aphids do not survive on fava beans. Thus, this new banker plant system should make it possible for both A. colemani and A. ervi to co-exist, reproduce, and provide successful cost-effective biocontrol of nearly all greenhouse aphid pests.


Click linked name(s) to expand/collapse or show everyone's info
  • Mark Yadon - Producer


Materials and methods:

The first experiment will determine the best growth stage of fava bean (Vicia faba) plant to infest with pea aphids (Acyrthosiphon pisum) that will balance maximum aphid production with minimal plant decline. We will use plants within individual bugdorm cages in a research greenhouse. A completely randomized 2x2 factorial design will be used, with all combinations of each treatment replicated 4 times. For the two treatments to measure the main effect of plant age, we will use potted fava bean plants that will be either young at no more than a week from sprouting, or mature at at least 2 weeks. For the two treatments to measure the main effect of initial pea aphid density, plants will be infested with a low treatment of only 5 aphids, or a high treatment of at least 30 aphids. Thus there will be four treatments – young plants with few aphids, young plants with abundant aphids, mature plants with few aphids, and mature plants with abundant aphids, each treatment replicated 4 times.  Plant health will be qualitatively assessed every 5 days for at least a month or until plants collapse. Aphids on 5 leaves per plant will be counted non-destructively every 5 days. For each sample date, a 2-factor ANOVA will be used to test for main effects of plant age and initial aphid density as well as possible interactions.  Results should provide a good idea of the plant age at which some, or many, aphids should be added to the plants for optimal performance.


The second experiment will check whether A. colemani will compete with A. ervi for pea aphids on the banker plants. It will also be done with plants in individual bugdorm cages in randomized design in a research greenhouse. Twenty pea aphids will be placed on each young fava bean plant and allowed to reproduce for a week.  Four female A. colemani, or A. ervi, or two of both species, will then be released into each cage to comprise three treatments. After two weeks, which would provide enough time for parasitized aphid mummies to appear, the total number of mummies will be counted on each plant, and the number of live aphids on 5 leaves will be counted.  Samples of parasitoids the emerge from mummies will be collected from each cage and identified to determine whether they are A. colemani or A. ervi, particularly in the cages win which both species were released.  A one-way ANOVA will be used to detect differences in live aphid numbers and mummy numbers among the three treatments. Very few mummies should be produced in the cages with A. colemani, and the parasitoids that emerged in the cages into which both species were released should be almost all A. ervi.


The third experiment will evaluate the pattern of parasitoid production, both in numbers and over time, when aphid-infested fava beans are exposed to the parasitoids at various growth stages. It will be conducted similarly to the first experiment. Four treatments will be used – young plants with few aphids, young plants with abundant aphids, mature plants with few aphids, and mature plants with abundant aphids, with each treatment replicated 4 times. Two female A. ervi will be released into each cage weekly for two weeks to simulate parasitoid visits in an open greenhouse. Weekly counts of all mummies and aphids on five leaves will be recorded for at least one month to measure parasitoid production from plants of each treatment. Two-way ANOVAs with repeated measures will be used for mummies per plant and for aphids per plant to detect treatment differences in main effects of plant age and initial aphid numbers as well as interactions. These results should provide a guideline for how to set up fava bean banker plants for A. ervi production in a greenhouse. 


Beginning in February, 2021, fava bean/pea aphid banker plants will be grown and placed throughout the greenhouse at Mischler’s Florist, as well as the usual barley banker plants for A. colemani. Sanderson will provide the initial pea aphids. Thus, banker plants for parasitoids of both large and small aphid pests will be simultaneously active.  Mark Yadon will scout his crops for aphid and parasitoid activity, collect mummies to give to Sanderson for identification, and grow at least one additional round of banker plants to replace old ones.  Sanderson will visit the greenhouse at least 4 times between February and May to collect the mummy samples and estimate aphid and mummy numbers on three aphid-susceptible crops which will be recommended by Mark. Mark will assess of the success of the trial, including degree of simplicity, criticisms, and possible improvements. Photographs/videos will be taken for use in presentations to grower conferences and to use on a fact sheet on the use of pea aphid/fava bean banker plants for A. ervi production.

Research results and discussion:

Results from 2020:

Due to delays caused by COVID-19 issues, our timeline for the research steps has been altered. 

We completed the first experiment and will describe and show these results below. 

In the interest of time, we have skipped the second experiment and proceeded to the third experiment which is currently underway. 

The second experiment, intended to check whether A. colemani will compete with A. ervi for pea aphids on the banker plants, has become less of a priority after a fortuitous accident occurred. Pea aphids on more than 20 fava bean plants were unintentionally exposed to an abundance of A. colemani, but no parasitized aphids resulted.  This was a surprise.  This Spring 2021, we eventually plan to conduct the second experiment to test the susceptibility of pea aphids to A. colemani experimentally, to confirm the anecdotal episode.

Our grower partner is gearing up for using the banker plants for his Spring crops and will be using our results from the first experiment to infest his fava bean plants at the optimum plant age.  We are seeking COVID-related permission from Cornell to travel to his greenhouse to take monthly assessments of the success of the banker plants and foxglove aphid biocontrol in his crops.

Here are results from the first experiment: to determine the best growth stage of fava bean plant to infest with pea aphids:

Plants that were infested with a high number of pea aphids resulted in more aphids than plants that were infested with a low level of aphids, and the plants were able to sustain this level for at least 3 weeks.  Among plants that were infested with the high number of aphids, more aphids were produced on plants that were infested 14 days after emergence, compared with plants infested at emergence or a week after emergence.  The 14-day-old plants generated more aphids and were still healthy after 3 weeks.  Thus, for maximum aphid production for eventual abundant wasp production, growers should grow the fava bean plants for 2 weeks before infesting with aphids.


Results for 2021:

Experiment/Objective 1: Other than confirming our 2020 results, no further work was done toward Objective 1 in 2021.

Experiment/Objective 2: For the second experiment, intended to check whether A. colemani will compete with A. ervi for pea aphids on the banker plants, our 2021 results were generally consistent with the accidental exposure we reported for 2020.  Though identification of some wasp specimens was not possible due to poor specimen condition, we found no A. colemani from pea aphids when the aphids were exposed to A. colemani.  Moreover, samples of pea aphids collected during the 2021 Spring crop cycle at the grower partner’s greenhouse revealed no A. colemani.  Thus, it appears that the pea aphid/fava bean banker plants for A. ervi production are compatible with the barley banker plants for A. colemani production. 

Experiment/Objective 3:

(Modified third objective: Aphid disappearance from banker plants)

In our experiments, as well as from the experiences of some growers who have contacted us, for unknown reasons the pea aphids on a fair number of the banker plants disappear after the wasps are released, and these plants obviously produce no wasps.  Yet an abundance of wasps may be produced on other fava bean banker plants concurrently in the same greenhouse.  We sought to investigate this phenomenon.

First, we, and our grower partners, have consistently observed that pea aphids will readily fall of fava bean plants if the plant is even slightly disturbed. Thus to successfully infest the plants with pea aphids, the fava bean plants must first be situated in their final location in the greenhouse and remain unmoved.  Then aphids can be added to them. But any slight jostling or movement of the plants will cause most or all of the pea aphids to drop off the plant. This dropping behavior due to jostling could be responsible for some of the disappearance of pea aphids from the banker plants.

Second, we also carefully observed pea aphid behavior in reaction to the presence of A. ervi wasps. Individual wasps were released onto aphid infested plants to observe the interaction between the foraging wasp and the pea aphids.  In all cases, as soon as the wasp was detected, the aphids either fell off the leaves or began quickly walking down the plant stems to exit the plant.  However, twice we have set up fava bean banker plants on an open greenhouse, successfully infested these with pea aphids, allowed the aphid population to establish and begin to grow for one week, and then released A. ervi wasps.  Many of the plants subsequently lost aphids and had few or no mummies after two weeks, though always some plants maintained an abundance of aphids and produced well over 60 mummies. We still do not understand this phenomenon of aphid disappearance on some banker plants but not others. 

Our partner grower has apparently avoided this problem by adding pea aphids from his aphid stock culture to his banker plants weekly.  Apparently his wasp production has remained relatively consistent by this weekly addition of aphids. 

Using plants that maintained their aphids in the presence of wasps, we have determined the best time to release the wasps after the plants have been infested with aphids. Using caged fava bean plants, A. ervi mummy production was greatest when wasps were released one week after initial aphid infestation, using the higher initial aphid infestation level noted from the first experiment in 2020. This resulted in at least 100 mummies per plant after two weeks, with likely additional aphids that would eventually become mummified but were not mummified at the end of two weeks.

Partner grower experience in 2021:

Our grower partner used fava bean banker plants during the Spring 2021 growing season and monitored the general numbers of mummies on his banker plants and on his cash crop. He collected and sent us samples of mummies from his banker plants and from his crop for wasp identification. (See Experiment/Objective 2 results). He was satisfied with the degree of foxglove aphid control that the banker plants provided, though in 2022 he plans to start his banker plant releases sooner to better time the wasp activity with the stage of his seedling production of perennial crop species.

Research conclusions:

2021: Some work is still ongoing. Current results are described above; concise description of conclusions will be forthcoming.

Participation Summary
1 Farmer participating in research

Education & Outreach Activities and Participation Summary

Participation Summary:

Education/outreach description:

By the end of the project, a fact sheet will be produced to provide growers with details of the use of pea aphid/fava bean banker plants. We may eventually also produce a video of the procedure. The one-year timeline of this project may not provide time to prepare a video. The fact sheet and video will be available online at the New York State IPM Program website. We will also send the fact sheet to the network of extension educators in New York and beyond (e.g., Leanne Pundt at UConn). Also, though beyond the timeline of this project, during the winter greenhouse conferences in New York State in 2022, usually held on Long Island, Albany, Buffalo, and the Hudson Valley, the results of this project will be presented by Sanderson.  Typically, these conferences draw about 250 growers. Furthermore, as soon as results from the project are final, Sanderson will look forward to sharing them at any greenhouse pest management conference in which he participates, such as the annual Greenhouse IPM In-Depth workshop at Cornell in July 2021. He will also present them in a research symposium for the Entomological Society of America annual meeting, to an audience of entomologists, including many extension entomologists.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.