Natural pest management in New York high tunnel and greenhouse vegetables

Final Report for LNE07-262

Project Type: Research and Education
Funds awarded in 2007: $108,905.00
Projected End Date: 12/31/2010
Region: Northeast
State: New York
Project Leader:
Judson Reid
Cornell Vegetable Program
Expand All

Project Information


A three year project by the Cornell Vegetable Program researched and promoted natural pest management in greenhouse and high tunnel vegetables via 23 on-farm trials, 4 on-farm workshops, 7 formal winter meetings, 20 articles and 1 webinar. These efforts lead the project to exceed the performance target of 15 growers adopting biological control of pests and diseases in greenhouse/high tunnels, with 20 implementing. Trial data and grower input indicated that while difficult to quantify economically in true field settings, natural pest management can be the difference between a crop and a failure and for many organic growers is their only option.


High tunnel and greenhouse vegetable production (protected culture) offer high returns and season-long market capture for Northeast vegetable farmers. Alternatives to chemical pest control were researched and promoted by this project. Biological methods such as predatory insects, resistant varieties and microbial fungicides are needed for several reasons:

• Persistence, degradation and proper rates of field pesticides in protected culture have not been well researched or documented.

• A growing number of small operations that depend on family labor are using greenhouse and high tunnel technology. This means children are often working within the structures and have a much lower chemical exposure tolerance than adults.

• Unless biological methods are employed, the unique set of pests and diseases in these settings would require small-scale growers to apply highly specific, unfamiliar chemical pesticides with high cost for relatively low acreage.

Performance Target:

Of the 80 farmers who attend demonstration farms and educational meetings as part of this project, 15 will adopt biological control of pests and diseases in greenhouse/high tunnel vegetable production within 3 years. These controls could include predatory insects, resistant varieties and microbial fungicides.


Click linked name(s) to expand/collapse or show everyone's info
  • Kathryn Klotzbach


Materials and methods:

In each of three years, grower-cooperators were recruited to host on-farm trials. In these trials pests were identified, then plots established and pest populations were monitored over time, before and after the use of natural controls, by a Cornell Vegetable Program technician. The controls used were nearly always natural enemies (beneficial insects and mites ordered from commercial distributors). Pest population data was entered into spreadsheets and mean population (or infection level) was calculated. Size of plot, number of plants per plot and number of leaves per plant varied by site. Details on scouting procedures are available in the annual summaries of cooperating-grower data.

Crops included tomatoes, cucumbers, melons, eggplant, greens, peppers and strawberries (all in protective culture). Pests included Two Spotted Spider Mites, Aphids, Thrips and White flies. Diseases included Powdery Mildew and Leaf Mold. In year one 2 farms hosted trials, in year two 11 farms, in year three 5 farms, and year four 5 hosted trials.

Growers were exposed to the principles behind biological pest/disease control in various educational formats such as winter meetings and on-farm demonstration trials. At these meetings control data from our trials was presented in handouts and Powerpoint presentations.
An evaluation of grower adoption was taken at the conclusion of year three via a survey mailed to 65 growers engaged through the project. A program assistant entered the data into a spreadsheet to calculate percent responses for a number of categories (see evaluation section for details).

Research results and discussion:

Our first Milestone “30 greenhouse/high tunnel vegetable growers will complete pest/disease surveys, ranking issues of greatest relevance” was accomplished in the fall of 2007 with 28 respondents ranking the pests and diseases they faced in greenhouse or high tunnel agriculture. See attached summary table of responses. In hindsight this survey did not play the intended roll of focusing which pests or diseases we were to work with, as this became highly site specific based on who was willing to cooperate with our project. However the survey did help us in other ways. It served to announce the start our project to many growers, and was also used to develop a list of potential cooperators. Looking back at the respondents we find that 6 of them went on to become demonstration-trial hosts.

Our next Milestone “80 growers will attend winter meetings and/or one of 4 on-farm demonstration farms for biological control” was met in the following methods via 23 on-farm trials and 4 on-farm workshops, 7 formal winter meetings, 20 articles and 1 webinar. Our rough estimate of attendance at all meetings is 445 people, which includes industry representative, university faculty, extension educators and farmers. This Milestone was important to make sure our message reached as many as people as possible, however adoption we found was best supported by on-farm face-time with growers. The Cornell Vegetable Program technician made more than 300 farm visits to support this project.

Our Performance Target “Of the 80 farmers who attend demonstration farms and educational meetings as part of this project, 15 will adopt biological control of pests and diseases in greenhouse/high tunnel vegetable production within 3 years” was met based on the results of our survey from the fall of 2010. These results are available in the attached file “LNE07-262 Performance Target Verification Results”. Of the 23 respondents to the survey 20 indicated they had increased their knowledge of pest management (Table 10), with 14 planning to use natural pest control in the future (Table 12) and 10 implementing to date (Table 11). However, the wording of the survey may have been somewhat confusing to respondents. For example only 10 indicated they implemented natural controls, yet 20 indicated they were using cultural controls, 16 indicated they were using low risk fungicides (Table 7), 13 were growing recommended disease resistant varieties (Tables 1 and 2) and 10 had implemented organic insecticides (Table 8). According to the stated project definition “these controls could include predatory insects, resistant varieties and microbial fungicides”. Thus if we examine the individual practices growers implemented we find that 20 of the 23 respondents increased their knowledge and those 20 had implemented natural pest controls (Table 9). This is 87% of the respondents.

Among natural enemies, Phytoseulis persimilis was the most commonly released by respondents to control Two Spotted Spider Mite, which was the most common arthropod targeted (Table 4). Although our project worked with P. persmilis successful control was not consistent. For example in Case Study #9 (see attached LNE07-262 Case Studies) although some control of Spider Mites on Greenhouse Cucumbers was achieved, eventually the crop succumbed. In Case Study 17 tomatoes were defended by P. persimilis, but cucumbers again succumbed. Our most convincing success with this predator was Case Study 22, where Spider Mite numbers were up to almost 45 per strawberry leaf before P. persimilis dropped the population to just over one pest mite per leaf. The number of persimilis was also tracked at this location, showing a direct relationship between the number of beneficial mites with the decrease in TSSM.

Our opinion is that P. persimils is best used in greenhouses with thermostatically controlled temperature and more stable relative humidity, versus a high tunnel where temperatures and humidity fluctuate widely. It should also be noted that we observed great differences in the vitality of the shipped material from different distributors. By the end of the project we greatly favored one distributor for P. persimilis. As a highly specialized predator it is also more difficult to employ in preventative programs. P. persimilis preys solely upon Two Spotted Spider Mites and in the absence of prey will perish. Thus we often promoted and demonstrated success with generalist predators.

One of these generalists was Amblyseius cucumeris. This beneficial mite was one of our repeated success stories. For example in Case Study #1 we were able to reduce thrips numbers from over 33 per leaf to 0. Other successful trials can be found in Case Studies 14, 16 and 19. As a generalist predator A. cucumeris lends itself to preventative approaches as it can survive on pollen or other arthropods in absence of pests such as Thrips. In our survey 33% of respondents implementing biological control used A. cucumeris, second only to P. persimilis (44%); although we feel it is in several aspects it is a superior option. These advantages include lower cost, higher survivability, broader prey range and options for preventative releases.

Aside from predatory beneficials, the project also worked with parasitic beneficials. Specifically wasps that parasitize aphids. The relationship between host and parasite is highly specific and more complex than a simple prey-predator system. In our first attempt to use parasitic wasps to control aphids (Case Study #4). Due to the placement of the row covers at this site, aphid populations were allowed to increase without being detected. Another problem that we faced was the temperature. As we got later into October nighttime temperatures were falling below freezing. This leads us to a question worth addressing; what types of biological control agents would work best in cooler temperatures, or even hotter temperatures? The challenge of aphids on high tunnel winter greens is discussed in Section 10.

We did demonstrate success with parasitic wasps to control aphids in warm season crops such as in Case Studies #8, 17 and 21. We feel that these parasites can be effective tools when applied early to warm season aphid infestations. Releasing more than one parasite is also advised to cover a range of aphid species. Waiting for lab confirmation of aphid species will not work for a biological control program that relies on releases early in the population curve.

For disease control in protected settings our project promoted ‘low-risk’ fungicides that are either organic or have a low Field Use Environmental Impact Quotient (per NYS IPM values). Our survey results were positive for this with 11 positive responses for microbial fungicides (or soil-premixes) and 16 positive responses for implementation of low risk fungicides (Table 7). Yet, our project’s experience is that resistant varieties are among the best tools for natural disease control. According to our survey 56% of responding growers exposed to our program are now growing resistant varieties. In tomatoes Geronimo is the hybrid of choice to manage Brown Leaf Mold and Diva cucumbers are the choice for Powdery Mildew control. These are both hybrids, and although derived from traditional plant breeding techniques, their adoption may be hindered as some growers favor open pollinated varieties. Horticultural issues such as fruit quality and plant vigor may also be a deciding factor for organic and direct marketers. Despite these issues, our program will continue to promote varietal resistance as an important management tool for all growers.

Participation Summary


Educational approach:

In year 1, nearly 30 surveys were completed by farmers throughout New York indicating pests and diseases of major concern in high tunnel and greenhouse vegetables. Our first field season found us focused on spider mites, aphids, thrips, leaf mold and powdery mildew on crops such as cucumbers, peppers, and tomatoes.

We accomplished several milestones towards the goal of increased adoption of biological control. These include the recruitment of four farms to host trials in 2008, two on-farm demonstration trials in year one and several winter meetings focused on biological control.
At one of our year one demonstration farms we implemented a successful biological control program for thrips on greenhouse cucumbers. We were able to bring the thrips population down to zero without pesticides. Fortunately this coincided with the date of our on-farm field day, with over 40 farmers in attendance.

In 2008 we worked with nine different growers, at a total of twelve different on-farm demonstration sites. The program technician made 109 on-farm scouting visits during the 2008 season. Out of the twelve sites that we scouted, nine locations had pest populations significant enough to release a biological control agent. The other three locations were scouted on a weekly basis until it was evident that damaging levels of pests would not interfere with the crop.

We also had an on-site grower meeting, where 40 growers were in attendance. There was also an opportunity to share research results at a University sponsored meeting that was solely focused on the use of biological control practices. Here, nearly 100 individuals (growers, researchers, industry representatives, etc) were able to interact and share successes, concerns, and opinions with each other. The results from the year one on-farm demonstration trial were also presented.

Scouting, performed by the project leader and technician, continued at 9 different cooperating sites in 2009, including over 85 on-farm scouting visits by the technician alone. Interest from five new growers to the program was established during the year. Five of the nine regularly scouted farms presented an opportunity to demonstrate/trial biological controls. Of these sites one farmer hosted 45 growers for an on-farm demonstration meeting in August 2009.
Over 100 growers attended a winter meeting (January 2009) that included industry, grower and university speakers. At this meeting the basics of pest and biological control were covered. Also, the results from the 2008 on-farm trials were presented. The PI presented project results to the statewide Organic Program Work Team, with teleconference sites in Long Island, Geneva, Albany and Ithaca, NY, reaching 60 faculty, extension educators and farmers.
Progress toward our Performance Target was notable in 2009 as some growers began to order their own biological controls or otherwise use natural controls such as resistant varieties.

The final year of this project was focused on grower adoption, including more on-farm visits, grower meetings, and a final adoption survey. Five farms cooperated with regular data collection. The technician made 118 on-farm scouting visits, in 7 different counties, from February to December 2010. Forty growers attended an on-farm twilight meeting in July 2010 where natural pest management was promoted with supporting project data. Again, in 2010, over 100 growers attended a winter meeting (January 2010) that included industry, grower and university speakers. At this meeting biological control biology and on-farm success stories were covered. A webinar hosted by the Great Lakes Vegetable Working Group reached over 100 sites across the country with natural pest management information. A final winter meeting in December saw an attendance of 20 vegetable growers.

No milestones

Additional Project Outcomes

Project outcomes:

Impacts of Results/Outcomes

The verification process was successful in that we were able to certify the accomplishment of our Performance Target “15 farmers will adopt biological control of pests and diseases in greenhouse/high tunnel vegetable production within 3 years”. Indeed our verification indicated that 20 farmers implemented some form of natural pest management by the end of the project.

However there are opportunities for improvement. Our program likely reached hundreds of growers via publications, webpages and meetings; yet we only had a database of 65 growers with mailing addresses we could reach with written surveys. Of those 65, 23 responded (35%). Fortunately these responses were sufficient to certify or Performance Target, but yet we are left without the full level of adoption across the region. We need to improve upon how we record our program contacts. This may become simpler as growers move towards the use of more personal computers and mobile devices. Yet, hard copies will continue to be necessary to reach those who are not ‘online’, including the many Old Order Amish that use our program. In the future assigning this task to someone, with periodic review of the contact list could help develop better contact lists which will then lead to fuller verification.

As agriculture educators and researchers, verifying social measures is not our background. We do realize though that we need help in crafting our verification tools to capture this data. Social scientists with evaluation experience have been enlisted for future projects.

The adoption of natural pest control in protected settings permits growers in certain situation to market a crop that would otherwise be unsalelable as in Case Study # 22 (strawberry hanging baskets).

The information generated in this project can be of use to extension educators without a background in biological control. The beneficial species we used with greatest success, for example A. cucumeris, can be implemented with a degree of confidence. Timings of both pest outbreaks and effective release dates found on our charts can be used by both educators and growers to predict when to implement biocontrol.

Economic Analysis

Although not described in the original proposal we did ask growers to provide economic impact data in our verification process. We asked them “Can you put a dollar value on your adoption of natural pest control? If so, please explain.”

There were five responses:
a. Natural pest control can save an entire crop. It is highly effective and valuable.
b. Zerotol has saved on Botrytis.
c. Natural pest control takes less times, works 24-7, seems to do a great job. Great public relation tool.
d. I can't quit put a dollar value on my crop, but I can say we were able to control the mites with beneficials, to the point where we could continue harvesting.
e. We never used anything for pest control.

Do biological controls pay? According to our estimates this question cannot be answered with a yes or no. For example at one research site our total bill for biological control in 2007 (Case Study #1) was $479 to control thrips on cucumbers.
• Our extrapolated yield for 2400 square feet was 260 bushels.
• At $20 per bushel, gross per tunnel is $5200.
• At 3 cukes/$1, gross per tunnel is 11,200 (retail pricing).

With this gross return our investment was about 4% of revenue at the retail price. An insecticide could have been applied, at a lower input cost, particularly if the amount of active ingredient required to treat the tunnel is charged only for a one-time use over such a small acreage. However for many organic growers this is not an option. Furthermore if we calculate in the labor costs of mixing, applying, storing, posting and reporting the pesticide use the true application cost will rise. 2007 was also our first year using beneficials and we were inefficient in the amount of material and timing. Comparing Case Study #1 to Case Study #14 (2009) we have achieved similar results in comparable situations with half the expense on biologicals. Thus, experience in biological control can save money.

Farmer Adoption

Per Section 5, to verify farmer adoption 65 surveys were mailed to those farmers contacted in the program, with 23 responses; 20 of those indicating adoption.

Assessment of Project Approach and Areas of Further Study:

Areas needing additional study

Through this project a clearly separate area of study was elucidated-winter pest management in high tunnels and greenhouses. Over the past three years of the project several changes have taken place. Indeed we were successful in promoting natural pest management which was successfully adopted by farmers on tomatoes, cucumbers and other warm-season crops. At the same time, these same farmers have responded to market demand for year-round local produce by growing storage vegetables as well as cold hardy greens in high tunnels. However, our success in managing pests biologically in the warm season was not immediately transferable to winter production of these greens. Releases of biological controls in the cold season were ineffective. Aphid infestations made some salad mixes unmarketable. Farmers have noted that some varieties of greens were ‘magnets’ for aphids, while other remained clean. Is this attractiveness specific to the crop, or are there varieties within the crops that are more likely to ‘pull’ aphids? Should one of these be included as a potential trap crop to concentrate management efforts, and keep the remainder of the tunnel clean? As most high tunnels for winter greens also serve as summer production centers for warm-season crops, what role do these play in the presence of winter aphid populations? Could beneficial insects work if introduced early enough? Are there OMRI spray materials that would work in conjunction with biological controls?

We are pleased to recently begun a new NESARE project “Sustainable Pest Management in High Tunnel Winter Greens Production” (LNE10-302) to answer these questions.

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.