Sustainable Pest Management for New York Urban Farmers

Our work in this project was demonstration focused, not pure research, as we ‘demo-trialed’ pest management techniques that already have proven potential for success. In this approach, although we do not replicate treatments to analytical levels, we employ rigorous data collection to support adoption. What was unique was the urban setting in which we adapted and deployed the management practices, as well as the site-specific pests and practices. However, at some sites we were able to collect quantifiable data that provided a very applied research approach with the ultimate goal of demonstrating impacts to the farmers.

 Releasing Ladybeetles to Manage Cabbage Whitefly

Scouting on 7/15/21 and 8/18/21 showed slightly less whitefly pressure near areas of Delphastus release. On one farm, there was a rating of 1.5 near the release location versus 2 to 2.5 at the other locations sampled. At the other farm, there was a rating of 2 near the release location versus 2.5 at most other locations scouted. Three Delphastus larvae were identified on collard leaves at one farm on 9/23/21, including clear evidence of cabbage whitefly predation. The presence of Delphastus larvae and evidence of cabbage whitefly predation, along with lessons learned about improving effectiveness of releasing Delphastus in outdoor settings and farmers’ stated interest in this biocontrol as an IPM strategy, suggested that we should expand on this trial in 2022.

In 2022, we trialed releasing Delphastus catalinae ladybeetles again on the same farms for control of cabbage whitefly (Aleyrodes prolotella). As in 2021, we found Delphastus larvae on brassicas (evidence of successful reproduction, since Delphastus are released as adults), with clear signs of predation of cabbage whitefly nymphs. However, as in 2021, we did not find conclusive evidence of a corresponding reduction in whitefly damage.

In 2023, at Red Hook, a count on 8/11/23 showed no significant difference in whitefly adults or nymphs between treatment and control plots. Treatment sections had significantly fewer egg clusters at 26 per leaf, compared to 41 per leaf in the control sections. This squares with the Delphastus supplier’s comment to us that the species (nymphs and especially adults) preferentially predates eggs versus nymphs or adults. On the 18 leaves scouted in treatment sections, we found only 3 Delphastus larvae and 1 adult, and none in the control sections.

Several Delphastus larvae were identified on leaves about 3 weeks after the biocontrol release at Pink Houses in 2023. Though this was not done as a controlled experiment and it was unclear whether the ladybeetles were effective as a control. Overall, while we continue to find ample evidence that Delphastus predates on cabbage whitefly and reproduces successfully while doing so, it remains unclear how to effectively use Delphastus as a control for cabbage whitefly.

Detecting Parasitoid Wasp of Cabbage Whitefly

An unanticipated finding in 2022: In the process of scouting for cabbage whitefly predation and Delphastus, we noticed the presence of a parasitoid wasp (Figure 6) which appeared to be parasitizing cabbage whitefly nymphs; the adult wasps were very common in the scouted plantings, with as many as 10 per leaf. The wasp resembled Encarsia formosa, a commonly purchased biocontrol of greenhouse whitefly, but we believed it to be Encarsia tricolor, which we subsequently learned has been observed as a naturally occurring cabbage whitefly parasitoid in Eurasia. A search of academic databases turned up no other documented observations of Encarsia tricolor parasitizing cabbage whitefly in the Americas.

In Fall 2023, we found and collected more unidentified Encarsia specimens at Pink Houses Community Farm. None were found at Red Hook Farm, as in past years, possibly due to their more regular regime of spraying insecticidal soap to suppress whiteflies (once weekly after first appearance, with an improved nozzle and new battery-operated backpack sprayer). Adult Encarsia were collected in 90% ethyl alcohol and sent to a USDA ARS research partner (Luke Kresslein, housed at the Smithsonian in Washington, D.C.), a taxonomist specializing in Encarsia, for identification. He reported that while the species is likely a close relative of Encarsia tricolor, it was determined to be a previously undescribed species. Project team members intend to submit a paper to describe and name it after the completion of this project. We brought some parasitized whitefly nymphs indoors and set up two insect rearing pens: one for Encarsia and cabbage whitefly together, and one for additional cabbage whitefly to provide a continual food source for the Encarsia. In late November, the whitefly population had increased enough to threaten the host plant’s health. On 12/1/23, the first confirmed Encarsia adult appeared. By 12/4/23, many more Encarsia adults had emerged, at least 4 per leaf. By 12/12/23, the whitefly population had crashed, with no egg clusters found in the rearing pen and very few adults remaining. Closer inspection showed that on some leaves, nearly every whitefly nymph had been successfully parasitized. While in a controlled environment, this demonstrated the potential of this Encarsia species to control cabbage whitefly.

Encarsia colonies were found at a new location in Brooklyn in 2024, on kale and collards, over 3 miles from the other sightings, suggesting the species may be more widely distributed and better established than previously known.

Adjusting Soil pH to Manage Pillbugs

After the second sulfur application in spring 2022, the treatment area’s pH was measured at 6.8, compared to 7.2 for the control area. While pillbug damage was reduced in general compared to 2021 (Figure 7) (possibly due to drier weather), Edgemere found nearly no pillbug damage in the treatment area in 2022 (Figure 8), compared to 15% of radishes damaged in the control area (Figure 9).

With an added yield equivalent to 375 bunches per acre, at $4/bunch, this amounts to $1,500/acre added revenue per year. The farm continued to add elemental sulfur ahead of radish plantings in 2023 and 2024, again finding nearly no pillbug feeding damage in treated areas each year. These results show promise in reducing pillbug feeding damage by lowering soil pH with elemental sulfur in soils with high levels of organic matter (>8% OM). With similar results achieved in 2022, 2023 and 2024, total added revenue at this farm was around $4,500/acre.

Taking a “Brassica Break” to Reduce Pest Pressure

In 2022, demonstration trials of a new approach to reduce brassica pest pressure was particularly successful. At all seven of farms implementing a brassica break, the first appearance of the previous years’ most significant spring brassica pests—cabbage whitefly, harlequin bug, or flea beetle—was delayed by two to seven weeks. This resulted in large increases in marketable product during brassica plantings’ head start. For example, at Kelly Street Garden, 163 lbs of collard greens were harvested before the first appearance of cabbage whitefly nymphs; the prior year, fewer than 35 lbs of collards were harvested before whitefly nymphs began affecting marketability. This 128 lb increase is over $350 more in revenue and amounts to over 23,000 lbs per acre and over $40,000 per acre in revenue.

In 2023, all 9 farms reported a later onset of cabbage whitefly, from 2 to 8 weeks. Two farms that used a modified Brassica break by leaving row cover on all brassicas until late June also reported a near absence of flea beetles, aphids, or harlequin bug until fall.

In 2024, 10 farms implemented a winter and/or spring Brassica Break and reported cabbage whitefly appearances delayed 3 to 7 weeks, with significant reductions in cabbage aphids and flea beetles as well.

Resistant Cucumber Varieties for Downy Mildew

Both farms trialing a downy mildew resistant cucumber variety (Brickyard) alongside non-resistant varieties (Longfellow and Marketmore 76) saw straightforward results. Cucurbit downy mildew caused rapid plant decline in the Longfellow and Marketmore beginning in late July 2023, with nearly no yield after 8/15/23, while Brickyard continued producing until mid-September. Both farmers intend to expand use of DM resistant cucumbers in the future.

Spotted Lanternfly Management

In 2023, while trialing low-impact, low-cost SLF traps, NYC experienced two heavy rain events between 6/16/23 and when we returned to check the traps on 6/29/23 at Snug Harbor. The tape had come unattached from most of the cardboard collars. The cucumber traps also were insufficient due to many points at which weeds reached above the collars, providing alternate routes for nymphs to crawl onto the cucumber plants and the trellising. That said, there may have been some reduction in SLF nymphs in one treatment section (bottom center, or “East-Middle,” in Figure 10). On okra, plants with sticky tape collars average 1.3 SLF nymphs per plant on 6/29/23, compared to 1.6 per plant on plants without traps, possibly indicating some effectiveness, but probably not justifying the labor of attaching the traps.

However, at Red Hook, we did see promising results on the fig tree (Figure 11). Although we were unable to get a thorough count of nymphs on the branches, we observed that of the 6 branches protected by a sticky tape collar, only 1 had SLF nymphs present on 6/29/23, while the majority of neighboring branches had at least 10 nymphs per branch. Although only one nymph was actually seen stuck to the tape, we watched several nymphs approach the sticky tape, then turn around. Even when attempting to force the nymphs to cross the tape by “herding” them down the branch with a hand, when they reached the tape, they preferred to walk around or even over our hand rather than venture onto the tape. This suggested that yellow sticky tape may be effective as a low-cost barrier for keeping SLF nymphs off of trees; at least for fig trees, which have smooth bark that allows the tape to stay attached for several weeks.

Releasing Biocontrols and Other Tactics to Manage Aphids

With using an integrated pest management approach of releasing biocontrols and using insect exclusion netting in 2022, Urban Fruits and Veggies (UFV) reported increases in crop yields with 25% increase in crop size, and an increase in crop quality and length of harvest window. UFV farm manager noted a reduction in labor time and costs. Prior to implementing these pest management practices, farm workers would spend 5 hours per week on pest management (observing, strategizing, handpicking) and after implementing these practices farm workers would spend 2 hours on pest management- net decrease of 3 hours per week. This equates to saving over $2200 on labor per year. In 2023, Aphidoletes aphidimyza larvae were observed feeding on aphids on cucumber leaves on 6/22/23 (Figure 12). We saw a decrease in aphid numbers after 4 biocontrol releases and the removal of the cucumber crop (Figure 13). The UFV farm manager noted again a reduction in labor time and costs. The farm manager reported the plants were cleaner with less pests and less money having to be spent on pest control products. He reported saving roughly 15 hours a month during the growing season on pest management in the greenhouse due to the biocontrol release. This equates to saving over $850 on labor during the growing season in 2023.

In 2022, the Brewster Street Farm manager noted the diatomaceous earth possibly had negative effect on lady beetles. The farm manager reported these practices resulted in an increase in crop quality and increased harvest window of high tunnel tomatoes by 4 weeks. The farm manager shared they “would have had to pull all tomato plants in July [2022] if not for the biocontrols.” In winter 2023, lady beetles were still alive and reduced the aphid population by 98.2% by 2/22/23 (Figure 14). On our final check on 3/10/23, we did not see any aphids in the high tunnel.

At the MAP Farm in 2022, no impact was observed from releasing biocontrols. This can partly be due to more communication needed from project members. In 2023, we worked closer with this site and included them in our urban scouting program. After discussions with the farmer manager, we focused on trialing methods for managing flea beetles in 2023.

Using insect exclusion netting to manage cucumber beetles

At Common Roots Urban Farm in 2021, insect netting appeared to provide sufficient protection of squash and cucumber plants from cucumber beetles. However, the cucumber plants did suffer from lack of pollination due to 1) not being a parthenocarpic variety and 2) the exclusion netting further inhibited pollinators from entering the tunnel. To remedy this, Common Roots moved frames of bees into the tunnel. Pollination thereafter seemed adequate but noticeably less. Twospotted Spider Mite (TSSM) damage was noted to be significant in the latter part of the season.

In 2022, Common Roots reported the insect netting appeared to provide sufficient protection of squash and cucumber plants from cucumber beetles. After releasing the biocontrol mite predator, farmers reported they noticed a decrease in TSSM damage and would have released the biocontrol a few more times but cost was prohibitive. Overall, Common Roots reported a 25% increase in cucumber yield in 2022. Prior to this project and implementing these sustainable pest management strategies, Common Roots reported $323 in cucumber revenue. In the second year of this project and after implementing these sustainable pest management strategies, Common Roots reported $1070 in cucumber revenue, a $747 increase in cucumber revenue over the course of this project. In 2023, the farmers at Common Roots took a break from farming to pursue other business opportunities. This increase in revenue was in the caterpillar tunnel which was 700 sq ft, this trial has demonstrated a potential $46485/acre increase in cucumber revenue by using insect exclusion netting on caterpillar tunnels.

Swede Midge Management

In 2021, Swede Midge (SM) was identified as a priority pest on Buffalo farms. Throughout the duration of the project, multiple urban farms implemented recommended SM management strategies. In 2022, UFV reported practicing crop rotation focusing brassica crops in raised beds that have not have brassicas in them for past 2-3 years and used Proteknet exclusion netting. They did not notice significant swede midge damage in 2022.

In 2022 at Brewster Street Farm, all raised beds affected by SM in the first year of this project had late season brassica crop residue removed and discarded. These beds were treated with Heterorhabditis bacteriophora nematodes in Fall 2021 and Spring 2022. In 2022, SM was still present on farm but less damage was observed compared to the previous year.

In 2023, we monitored the populations of SM starting mid-summer at two farms, WST and FLCF. At WST, we saw highest SM numbers in mid-July and the population decreases over the rest of the season with a small spike in late September and going dormant by the end of October (Figure 15). This aligned with typical SM population dynamics observed in our area. In New York, we expect to see peak emergence of adult swede midge from overwintered pupae in late spring. The high SM population in mid-July corresponded with the initial collards planting showing severe damage. The sharp decline in population we speculated is due to timely post-harvest crop destruction by the farmer. We did observe SM on sticky cards in the collards bed under insect netting which we speculated is due to SM already being present in the soil.

At FLCF, we saw moderate SM numbers at the end of August 2023 with a spike in early October which we speculate may be due to warm weather (Figure 16). The farmer saw pest management and educational value in monitoring the SM population. Based off our work in 2023, the farmer stopped growing Red Russian Kale during the summer season and explored new brassica varieties that may be less appealing to SM. In addition, the farmer planned to continue using insect exclusion netting with brassicas, especially in areas that have not have brassicas grown in them for at least 3 years. The farmer shared the trap was a “great teaching tool working with community gardeners and tour groups. It fits in with the growing practices of the farm. It shows the pest is there and is helpful to keep track of pest pressure. It helps to illustrate the importance of knowing a pest’s life cycle when choosing management tools.” The farmer reported on average 25 people visit the farm weekly from school and community groups, with over 1000 people visiting the farm throughout the growing season.  

In 2024 at FLCF, trap counts showed the swede midge population fluctuated depending on management strategy used (Figure 17). In our covered carrot bed (trap 3), we saw emergence first which we speculated is due to that the raised bed had a heavy infestation that overwintered. We see the SM population peaked and dropped mid-June remaining low until the carrots were harvested and the netting came off in August. This suggests the potential of using exclusion netting to crash the swede midge population in small scale growing spaces. Carrots showed no damage or reduced yield by being covered. The grower shared that covering the carrot bed with netting did make it harder to weed.

When monitoring background population pressure, trap 4 loosely followed trap 1 trends. Both traps were place in uncovered raised beds in different locations on the farm, yet trap 4 counts remained lower than trap 1 for entire season (Figure 17). We speculate that this may be due to crop type preference for SM- trap 4 was located in a bed planted with lacinato kale and trap 1 was located in a bed planted with collards. Additional speculation includes that SM might have overwintered in the trap 1 bed and not in the trap 4 bed, hence SM mostly stayed put as it was already happily feeding. Unexpectedly, averaging trap count data together, SM population peaked in early September which suggests SM might behave differently on urban farms than rural farms.

In our uncovered and covered collards beds (trap 1 and 2), we saw SM emergence first in the uncovered bed as we would expect. We saw low SM pressure slowly building in the covered bed in May and June. Towards the middle of the season, pressure rose and peaked on 8/15/24 and 9/8/24. Crop damage was consistently lower in the covered plot compared to the uncovered plot throughout the season (Figure 18). We observed crop damage tended to increase when trap counts increased.

We observed higher crop yields and quality on collards grown under cover. Our collards bed with netting produced higher yields than our collards bed without netting. Our collards bed with netting produced a total of 222.6 lbs and our collards bed without netting produced a total of 214.5 lbs. The bed with netting produced 8.1 lbs more than the bed without netting. This farm reported selling a bundle of collards for $2.50 and a bundle is roughly 2 pounds in 2024. This 8.1 lbs yield increase with netting translates to $10.13 increase of revenue. Our plot area was 28 ft². This trial has demonstrated a potential $15,713 / acre increase in collards revenue by using exclusion netting.

Beyond yields, leaf quality was observed to be better on covered collards compared to uncovered collards, especially earlier in the season. On 7/2/24, a farm volunteer helping to harvest collards in both beds said "I vote for netting, plants are much cleaner, taller and healthier looking.” At the end of the season, FLCF farm manager shared “Although the difference between the covered and uncovered beds was not significant in weight, the quality of the leaf in the covered beds was superior, particularly in the early half of the season. As the summer progressed and insect pressure under the cover increased, the quality of the leaf was nearly equal.”

Future work includes addressing when is best to take off the netting and investigating the best swede midge damage rating approach for various brassica crop types. Throughout the season, other brassica pests were present including imported cabbageworm, diamondback moth and aphids. The farm manager and project team members hypothesized taking the netting off sooner would have reduced later season damage and helped reduce pressure from other brassica pests trapped under the netting. Most research done on swede midge in our area was on heading brassicas. On urban farms, brassicas most grown and affected by swede midge are leafy greens (kale and collards). In 2024, FLCF farm manager developed an SM leaf damage scale that we plan to incorporate in future work (Figure 19).

Using Row Cover and Entomopathogenic Nematodes to Manage Flea Beetles

Eggplant plots without row cover initially showed higher damage levels than eggplant plots with row cover and eggplant damage levels appeared to be evening out among treatments as the season progressed (Figure 20). We hypothesize this could be due to plants growing more leaves. Eggplants showed higher damage levels than collards and flea beetle counts on sticky cards were consistently higher in eggplant plots (Figure 21). We hypothesize this could be due to flea beetle host preference. We believe we observed multiple species of flea beetles on plants and on sticky cards including the crucifer flea beetle, striped flea beetle and eggplant flea beetle. Flea beetles were present on sticky cards in each treatment.

Eggplant plots with row cover produced higher yields than eggplant plots without row cover (Figure 22). Our two plots with row cover produced a total of 164.75 lbs and our two plots without row cover produced a total of 127.1 lbs. This farm reported selling a quart of African Eggplant (1 lb) for $5 in 2023. This 37.65 lbs yield increase with row cover translates to an $188.25 increase of revenue. Our single plot area was 37.5 ft² and thus our row cover plot area was 75 ft². This trial has demonstrated a potential $109335.60 / acre increase in African eggplant revenue by using row cover. Farmers qualitatively reported horticultural benefits to row cover as well. On 6/22/23, MAP farmer shared “there are so many more flowers on eggplants under row cover.” Project team members and MAP farmers noted darker green foliage on plants under row cover.

Lesson learned from this trial on best practices for using row cover include: regular weeding and drip irrigation are helpful for row cover success and using mulches can help; row cover may be most successful in areas where crops have not been previously grown for a few years or in areas that are known to be free of targeted pest.

MAP Farmer Katie reported the trial had numerous benefits for the farm’s social mission. She saw great value in the trial demonstration information sign (Figure 23) posted in front of the trial as a way to passively educators farm visitors exploring the farm. She shared hosting the trial was “fun and a great example for our community to see a field trial in action and be exposed to the process of agricultural research.” Katie estimates over 500 people interacted with the 2023 demonstration trial.