Final report for GS18-189
Cover crops offer a range of benefits that include suppressing disease in subsequent cash crops, providing habitat for beneficial insects, and promoting pollinators. Overwintering cover crops can provide early floral resources for pollinators and refuge to other beneficial insects like natural predators before cash crops are planted. These attributes may help bridge beneficial insects into the crop following cover crop termination.
Cover crops have also been found to reduce severity of some watermelon diseases. Based on these benefits, it is worthwhile to investigate the potential of different cover crops for use as an integrated pest management strategy in watermelon production. Such a strategy would benefit watermelon yield while cutting costs and pesticide use, making production more sustainable. The grant provided funding for 1-year of this research project designed to determine the cover crop monocultures or mixtures best suited for the watermelon disease suppression, pollinator production, and bridging beneficial insects into watermelon production systems.
To achieve this goal, insects were sampled in various cover crop treatments before termination and after termination before planting watermelon. Disease and insects were evaluated in the subsequently planted watermelon. Pollinators were sampled in the cover crops and watermelon during bloom as well. Findings were presented to watermelon growers during talks and at field day watermelon workshops.
An important finding from this research was that when compared to the control with no cover crops, none of the winter cover crops investigated were found to increase major watermelon pests or disease. In general, cover crops that flowered including the mustard, crimson clover, and pea attracted more beneficial insects. These cover crops also hosted aphids that served as a food source for beneficial insects like lady beetles and parasitoid wasps. Although these aphids were present in the cover crops, they were not bridged into the watermelon because they were not melon aphids. The grass cover crops tended to support fewer insects and often fewer pests compared to the control, however, outbreaks of rice stink bugs occurred and were found to be more prevalent in the grasses. Because there were insects present before and after cover crop termination, a bridging effect likely occurred. The bridging effect was not very clear because pest pressure was low in the watermelon. Most of the beneficial insects in the cover crops were natural enemies of aphids like lady beetles and parasitoid wasps. Because melon aphids were not abundant in the watermelon, there was no food source for the natural enemies, which may explain why they were not very common in the watermelon either.
It was difficult to determine whether the cover crops impacted watermelon disease because fungicides were applied in order to ensure that their yield was preserved as part of a larger project investigating the effects of cover crops on watermelon. Disease incidence was consequentially low and was the same across all treatments in the field. In order to determine the true effects of cover crops on watermelon disease, it might be necessary to perform a more controlled study in which the plants are inoculated with the pathogens. It is important to note that none of the cover crops bridged diseases into the watermelon.
Pollinator sampling revealed that native bees played a very important role in watermelon pollination. Cucurbit specialists were very common throughout the field, much more so than honey bees. Sampling in the cover crops revealed that only those that flower before termination serve as resources for pollinators. Mustard was one of the cover crops to flower before termination. However, when canola and Austrian pea were permitted to grow until bloom, they also supported bees that had emerged in the spring. Depending on fall planting date and winter weather conditions, other flowering cover crops may have potential to flower before termination and provide early season resources for bees as well.
Based on the findings of this project, growers that seek to promote pollinators should plant a flowering winter cover crop like mustard or canola that blooms before termination in the spring. If growers would like to keep general agricultural pests low, then they should use a grass cover crop like black oats, cereal rye, or wheat. If melon aphids are an issue, growers should plant mustard or a legume to increase natural enemies of aphids. It is also important for growers to know that if they are growing a legume nearby they should avoid using legume cover crops because they could increase problematic pests like pea aphids. If growers are going to be producing cereals, then they should avoid grass cover crops that were shown to support rice stink bug. Ultimately, growers should choose cover crops not only based on their potential impact on beneficial insects but also based on other benefits like nutrient contribution and their impact on soil health and yield.
Objective 1: Evaluate winter cover crops and cover crop mixtures to determine which are best suited for increasing natural enemies.
Objective 2: Evaluate winter cover crops and cover crop mixtures to determine which are best suite for disease suppression in watermelon.
Objective 3: Determine the effect of different winter cover crops and cover crop mixtures on pollinator abundance and diversity, especially for pollinators essential to watermelon production.
Objective 4: Participate field days to disseminate information on cover crops in watermelon production systems.
Research Sites and Plot Setup
Research was conducted at two field sites, the Southwest Research and Extension Center located in Hope, AR (33.7107°N, 93.5573°W) and the Vegetable Research Center located in Kibler, AR (35.3791°N, 94.2333°W). The sites will be referred to as Hope and Kibler respectively. Plot setup varied by location, however, both locations had 5 replications of 12 treatments in a randomized complete block design. Hope plots were 3.7 m by 9.1 m with 3.0 m allies between blocks and 1.5 m. allies between treatments. Kibler plots were 3.7 m by 9.1m with 3.0 m allies between blocks and no allies between treatments. The 12 cover crop treatments consisted of black oats (Avena strigosa Schreb), cereal rye (Secale cereale L.), winter wheat (Triticum aestivum L.), Austrian pea (Pisum sativum subsp. arvense L.), crimson clover (Trifolium incarnatum L.), Austrian pea with black oats, Austrian pea with cereal rye, Austrian pea with winter wheat, crimson clover with black oats, mustard (Sinapis alba L.), and a mix of mustard, Austrian pea and black oats, as well as a control with no cover crops (Table 1). Cover crop treatments were chosen based on cover crop species popular amongst Arkansas growers and well-adapted for the southeastern U.S. (Roberts et al., 2018; Clark, 2007). Cover crop seed was sourced from Southern Soil Solutions Inc. (Clarendon, AR).
Plots were established when cover crop seeds were hand broadcasted uniformly throughout the plots between mid-September and early October (Table 1). Seeding rates for cover crops in mixtures were adjusted based on standard recommendations for mixtures (Clark, 2007). Prior to planting, Austrian pea and crimson clover seed was inoculated with nitrogen-fixing rhizobia bacteria (Graph-Ex SA™, ABM®, Van Wert, OH), as is standard for legume cover crops in order to ensure they attain their full potential for nitrogen fixation (Clark, 2007). Austrian pea seed was broadcasted first and raked in due to a deeper planting depth requirement. In Kibler, cover crop seed was watered for establishment. Cover crops were not irrigated during the growing season.
Cover Crop Termination
Cover crops were terminated in the spring approximately two weeks before transplanting the watermelon. Hope cover crops were terminated by strip-tilling. The cover crops were mowed down at the center of the plot, leaving strips of cover crop standing on both sides. The mowed cover crops were then tilled into the soil and a raised plasticulture bed was formed. Kibler cover crops were no-till terminated by roller crimper followed by an application of glyphosate (Cornerstone® Plus by WinField® United, St. Paul, MN). Cover crop residue was then left on the soil surface.
After cover crop termination, drip irrigation line was placed down the center of the plots at Kibler. Drip tape irrigation was utilized at Hope installed underneath the plastic mulch. Watermelon transplants were grown from seed in the greenhouse. Transplants of the cultivar ‘Jubilee’ were transplanted on 4/30/18 in Hope 2018, 4/24/19 in Hope 2019, 4/20/18 in Kibler 2018, and 4/23/19 in Kibler 2019. ‘Jubilee’ watermelon seed was sourced from Sustainable Seed Company (Chico, CA) for the 2018 season and NeSeed™ (Hartford, CT) for 2019. Hope transplants were planted into raised beds while Kibler transplants were planted directly into the cover crop residue or on bare ground in control plots along the drip tape. Transplants were spaced 0.3 m apart with 9 plants per plot. In Kibler 2018, ‘Jubilee’ transplants died and the field was replanted with transplants of the cultivar ‘790’.
Krista™ K soluble potassium nitrate fertilizer (Yara, Tampa, FL) was drip applied per recommendations listed in the Southeastern Vegetable Crop Handbook (Kemble et al., 2019). Pesticide application was minimal (Table 2). Ridomil Gold® SL (Syngenta®, Wilmington, DE) was applied through the drip tape at transplant and again 30 days later. A fungicide spray program was utilized in which either Bravo® C/M (Fermenta Plant Protection Company, Painesville, OH) or Bravo Weather Stik® (ADAMA, Raleigh, NC) with Kocide®3000 (Certis USA, Columbia, MD) per labeled rates were applied every 14 days. About two applications were made each year. Insecticides were to be applied if a major pest outbreak threatened yield. The only insecticide application to occur during this study was at Kibler in 2019 when Sevin® XLR Plus (Tessenderlo Kerley Inc., Phoenix, AZ) at the labeled rate was applied to manage an outbreak of striped cucumber beetles that exceeded the economic threshold. Rally® and Inspire Super® for powdery and downy mildew were on hand if outbreaks of these diseases had occurred but were not needed.
Insects were sampled in the cover crops just prior to cover crop termination and just before transplanting the watermelon. Cover crop insects were sampled by sweep netting with a 15 inch-diameter aerial net (BioQuip® Professional Series Insect Net, Rancho Dominguez, CA). Sweep netting consisted of ten consecutive 180° back and forth sweeps moving down the center of the plot. Sweep net insect sampling methods were adapted from Tillman et al. (2004) and modified to ensure a comprehensive sample that would accurately represent insect populations in the plots. Cover crop samples were later identified to family using Borror and DeLong’s Introduction to the Study of Insects, 7th edition. (Triplehorn & Johnson, 2005). All arthropods in the samples were identified to family when possible.
Once the watermelon was transplanted, insects in the field were assessed weekly up until harvest in July. Watermelon insect data consisted of field counts of pest and beneficial insects found on or in the watermelon foliage. During the seedling stage, 5 plants per plot were chosen at random for the insect field counts. As plants matured, 5 vines were selected randomly per plot for evaluation. In 2018 when watermelon vines began to fruit, insects were sampled with 5 back and forth sweeps of watermelon foliage. All watermelon insects were identified to family by sight in the field. Major watermelon pests were identified to species.
Watermelon disease ratings were taken at Hope and Kibler during the 2018 and 2019 field seasons. Disease severity ratings based on a visual estimate of symptomatic foliage on all plants within each plot was taken weekly from the time of planting until harvest. Individual ratings were taken for angular leaf spot, anthracnose, gummy stem blight, powdery mildew, and downy mildew. An overall disease rating was taken for each plot as well. The ratings were on a one to ten scale in which 0 = 0% disease, 1 = 1 to 10% disease, 2 = 11 to 20% disease, 3 = 21 to 30% disease, 4 = 31 to 40% disease, 5 = 41 to 50% disease, 6 = 51 to 60% disease, 7 = 61 to 70% disease, 8 = 71 to 80% disease, 9 = 81 to 90% disease, and 10 = 91 to 100% disease. Disease ratings were observational and rating methodology was modified from methods used by Zhou and Everts (2012) to rate gummy stem blight and anthracnose severity on watermelon foliage. When a potential disease was found, plant samples were sent to the University of Arkansas Plant Health Clinic for disease identification.
Pollinator Sampling Technique
Pollinator sampling took place in both the 2018 and 2019 field seasons. Locations and cover crop treatments sampled varied between the years; however, the sampling technique remained consistent. Pollinators were sampled passively using blue vane traps (OakStump Farms® blue vane trap, SpringStar® Inc., Woodinville, WA), which has been shown by Stephen and Rao (2007) to be an efficient sampling technique for studies on bee diversity in the presence of floral resources. One vane trap wired at approximately bloom height to a u-post was placed at the center of each of the sampled plots. Vane traps were filled approximately one-third of the way full with soapy water. Soapy water was mixed by the gallon using about 3 drops of unscented dish soap (Dawn® Ultra Free & Gentle, Procter & Gamble Company, Cincinnati, OH) as a surfactant. Traps were only set during periods without precipitation or harsh weather conditions that may have impacted pollinator visitation or disturbed the traps. Samples were dumped into a strainer and insects were transferred to 70% ethanol for storage until they could be processed. 70% ethanol is the standard fluid used to preserve insect samples (Triplehorn & Johnson, 2005).
2018 Kibler Pollinator Sampling
Pollinators were sampled at Kibler in 2018 before cover crop termination and during watermelon bloom. The only cover crop to flower before termination was mustard treatments, thus only three treatments were sampled: the control with no cover crops, the mustard treatment, and the mustard/Austrian pea/black oats treatment. Cover crops were sampled twice before termination with sampling dates approximately two weeks apart. During sampling, all mustard was flowering. Pollinators were also sampled twice approximately two weeks apart when the watermelon was in bloom. Traps were set for 48 hours during each sampling date.
2019 Hope Pollinator Sampling
Pollinator sampling took place once, during which the vane traps were set in all treatments for approximately 24 hours. Traps were only set for 24 hours due to weather and time constraints. This sample occurred one day after cover crops had been terminated via strip-tilling. Vane traps were set up in the center of each plot about 0.6 m away from cover crops left standing on either side. If blooms were present, vane traps were wired at about bloom height. If no blooms were present, vane traps were wired at approximately average bloom height about 0.6 m off the ground.
2019 Fayetteville Pollinator Sampling
The study set in Fayetteville at the Arkansas Agricultural Research and Extension Center was solely for the purpose of looking at the effect of different flowering winter cover crops on pollinators and watermelon was not planted. Fayetteville had 4 replications of 7 treatments set up in a randomized complete block design. The treatments included a control with no cover crops, mustard (Sinapis alba L.), canola (Brassica napus L.), Austrian pea (Pisum sativum subsp. arvense L.), mustard with canola, mustard with Austrian pea, and canola with Austrian pea (Table 2). The plots each measured 9.1 m by 4.3 m separated by 1.5 m allies. As with Hope and Kibler, cover crop seeds were hand broadcast and uniformly distributed throughout the plots in the fall. The Austrian pea was inoculated and broadcast first. Following the Austrian pea, the field was rolled to fulfill its deeper planting depth requirement. The canola and mustard seeds were left on the soil surface and watered in with sprinkler irrigation after plant. Fayetteville cover crops were not terminated because watermelon was not produced at this location.
Pollinators were sampled in the spring when the cover crops were in full bloom. Approximate percentages of flowering cover crops within each plot were noted. The vane traps were set for approximately 48 hours during each sample. Traps were sampled twice with the two sampling dates about two weeks apart.
Once identified to family, arthropods were generally categorized as being agricultural pests, beneficials, or neither. The beneficial categorization included natural enemies like predators and parasitoids, as well as pollinators. Mean insects per treatment were calculated and analyzed for differences using SAS® 9.4 (SAS Institute Inc., Cary, NC) one-way ANOVA by PROC GLIMMIX and means separation by least squares.
Disease severity rating data was analyzed using SAS® 9.4 (SAS Institute Inc., Cary, NC) ANOVA by PROC GLIMMIX and means separation by least squares.
Bees were sorted out from the vane trap pollinator samples and identified to genus using the Discover Life Bee Genera guide (https://www.discoverlife.org). Mean bee abundance and mean bee richness for each treatment was calculated and compared using SAS® 9.4 (SAS Institute Inc., Cary, NC) ANOVA by PROC GLIMMIX and means separation by least squares. Richness is a measure of diversity equivalent to the number of different genera.
Clark, A. (Ed.). 2007. Managing Cover Crops Profitably, 3rd edition. Beltsville, MD: Sustainable Agriculture Network.
Kemble, J.M., Meadows, I.M., Jennings, K.M., & Walgenbach, J.F. (Eds). 2019. 2019 Southeastern Vegetable Crop Handbook.
Roberts, T., Ortel, C., Hoegenauer, K., Wright, H., & Durre, T. 2018. Understanding cover crops. University of Arkansas Research and Extension. 9 Aug. 2019. <https://www.uaex.edu/publications/pdf/FSA-2156.pdf>
Stephen, W. P., & Rao, S. 2007. Sampling native bees in proximity to a highly competitive food resource (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society, 80, 369-377.
Tillman, G., Schomberg, H., Phatak, S., Mullinix, B., Lachnicht, S., Timper, P., & Olson, D. 2004. Influence of cover crops on insect pests and predators in conservation tillage cotton. Journal of Economic Entomology, 97, 1217-1232.
Triplehorn, C.A. and N.F. Johnson. 2005. Borror and DeLong’s Introduction to the Study of Insects, 7th edition. Blemont, CA: Thomson Brooks/Cole.
Zhou, X. G., & Everts, K. L. 2012. Anthracnose and gummy stem blight are reduced on watermelon grown on a no-till hairy vetch cover crop. Plant Disease, 96, 431-436.
Impact of Cover Crops on Beneficial and Pest Insects in Watermelon
Ultimately, the results varied greatly from year to year and among cover crop treatments. Other factors outside the cover crop treatments likely contributed to our inconsistent results from year to year. The major finding of this study is that the cover crops investigated did not serve as alternate hosts to the major pests of watermelon including cucumber beetles, squash bugs, and melon aphids. The results of the Hope 2018 trial support that mustard was an ideal cover crop to increase beneficial insects because it attracted aphids that served as a food source for lady beetles and parasitoid wasps. In 2019, most of the mustard was winter-killed and the results were inconsistent with the previous year. Based on generalizations, it seems that flowering cover crops are better for providing resources to beneficial insects. It was more common that treatments containing mustard or one of the legumes had greater beneficials. The lady beetles especially seemed to be attracted to the crimson clover, which was exemplified in Kibler. The grasses tended to support fewer insects in general, but this also meant there were often fewer pests compared to the control. If general pests are an issue, it may be better to plant a grass cover crop that supports fewer insects.
This study also sought to investigate the effects of strip-tilled cover crops, no-till, and whether insects were bridged into the watermelon. The strip-tilled cover crops at Hope should have increased beneficial insects in the watermelon in theory, but unlike Tillman et al. (2004), these data did not show definitive evidence of an effect where the cover crops left standing became refuge for the beneficial insects after cover crop termination. However, there was never much pest pressure in the watermelon. The watermelon in the control treatments with no cover crops had no major pest outbreaks that threatened yield, which indicates pest pressure was low throughout the field. The lack of pests may have meant that the predators did not have a sufficient food source after the cover crops were terminated. In a different system where pest pressure is greater, perhaps there would be different results. Because the strip-tilled system and the no-till system were at different study sites, comparisons cannot be made between the two. Although, there were similar trends in the results from the two sites. It would be interesting to compare strip-till and no-till at the same location in order to determine the impact of the cover crop termination method on cover crop insects.
As far as a green bridge, both pest and beneficial insects were present before and after cover crop termination. It seems that a green bridge effect was possible. However, without greater pest pressure in the watermelon to support natural enemy populations, there is no way to tell how this green bridge may have benefited IPM. Further study is necessary to determine if cover crops bridge beneficials into watermelon that help reduce pests and benefit yield. Cover crops have the potential to impact insect populations in a production system, but this impact must be further explored. However, a flowering cover crop that is not an alternate host to the cash crop is likely to be the best bet for attracting beneficial insects.
Impact of Cover Crops on Watermelon Disease
Watermelon disease was found to be the same across all cover crop treatments and the control with no cover crops. In general, disease incidence was low. In order to more accurately test the impact of cover crops on watermelon diseases, watermelon plants or field sites would likely need to be inoculated with the pathogen to ensure the diseases are present. Because the disease data were collected as part of a larger study that investigated the effect of cover crops on other aspects of watermelon production, inoculating watermelons with the pathogens to test for disease was not feasible. It was necessary to apply preventative pesticides to protect yields, although a much-reduced spray program was utilized. As such, the common diseases of watermelon aside from angular leaf spot did not occur at Hope or Kibler in either year. Because disease incidence and severity were low across the field, the cover crop treatments evaluated were not found to serve as hosts for common watermelon disease. In future studies, growing cover crops and watermelon in a more controlled environment and inoculating with the pathogens to be tested would likely provide more conclusive results on whether certain cover crops have an impact on disease.
Impact of Cover Crops on Pollinators
In summary, the results of this part of the study demonstrated that flowering winter cover crops can provide early season resources for bees if they survive the winter and flower before cover crop termination. Whether cover crops flower before termination can be impacted by planting date and weather. Mustard in the year when it was not winter-killed and canola in the following year seemed to be attractive for a variety of bee genera. In the second year of the study, Austrian pea may have been less attractive; however, future replications of the study are necessary to confirm.
Furthermore, the importance of native bees was highlighted when watermelon was sampled during bloom. The data show that honey bees were not likely a major contributor to watermelon pollination in this trial. Bees are the most important pollinators of watermelon and there is evidence that native bees can provide pollination services sufficient for the demanding pollination requirements of watermelon (Andersen, 2011; Kremen, Williams, and Thorp, 2002). The abundance of native bees collected during watermelon bloom supports this claim. The cucurbit specialists found in high numbers were likely to have been important in providing pollination services as they are well adapted to access watermelon floral rewards (Wilson & Carril, 2015; Hurd, Linsley, & Whitaker, 1971). Growers should be aware that these specialists and many other native bees are ground nesters, often found nesting in or nearby the field. In addition to providing flowering cover crops as floral resources, steps can be taken to manage agricultural systems to better preserve bee habitat (Nicholls & Aliteri, 2013). In conclusion, a variety of factors like the survival of the cover crops, sampling techniques, weather, and other environmental conditions likely influenced these preliminary findings. As discussed, steps could be taken to improve upon this study and more accurately assess cover crops that might be used to promote pollinator populations.
Andersen, C. 2011, April. Home gardening series: watermelons. University of Arkansas, Division of Agriculture Cooperative Extension Service. 15 Oct. 2017. <https://www.uaex.edu/publications/PDF/FSA-6012.pdf>
Hurd, P. D., Linsley, E. G., & Whitaker, T. W. 1971. Squash and gourd bees (Peponapis, Xenoglossa) and the origin of the cultivated Cucurbita. Evolution, 25, 218-234.
Kremen, C., Williams, N. M., & Thorp, R. W. 2002. Crop pollination from native bees at risk from agricultural intensification. Proceedings of the National Academy of Sciences, 99, 16812-16816.
Nicholls, C. I., & Altieri, M. A. 2013. Plant biodiversity enhances bees and other insect pollinators in agroecosystems. A review. Agronomy for Sustainable Development, 33, 257-274.
Tillman, G., Schomberg, H., Phatak, S., Mullinix, B., Lachnicht, S., Timper, P., & Olson, D. 2004. Influence of cover crops on insect pests and predators in conservation tillage cotton. Journal of Economic Entomology, 97, 1217-1232.
Wilson, J. S., & Carril, O. J. M. 2015. The Bees in Your Backyard: A Guide to North America's Bees. Princeton, NJ: Princeton University Press.
Educational & Outreach Activities
Preliminary research was presented in a talk entitled "Can cover crops serve as a bridge for beneficial insects in watermelon production systems?” at the 2018 Joint Annual Entomological Society of America meeting in Vancouver. Another presentation was given at the 2019 Horticulture Industries Show in Fayetteville, AR. A talk entitled "Can cover crops serve as a bridge for beneficial insects in watermelon production systems?" was given at the Arkansas Entomological Society meeting in Barling, AR in February 2019. The research was also presented as a poster at the 2019 National Sustainability Summit in Tampa, FL. I talked about the findings of the research and recommendations to growers during the watermelon workshops at the Kibler, AR Vegetable Research Station and Hope, AR Southwest Research and Extension Center field days in the in the spring of 2019.
We will likely be publishing two journal articles from the research in the near future. One article will be on using cover crops to increase beneficial insects and decrease pests in watermelon and the second article will be on using cover crops to promote pollinators.
This project investigated the benefits of winter cover crops for integrated pest management. The findings showed that cover crops can help support natural enemies that may help decrease pests. It also found that cover crops can also help support pollinators by providing early season resources. Pollinator conservation is very important because they provide pollination services that produce many crops and have been in decline around the world because of factors like habitat destruction and global warming. Pollinators are an essential part of sustainable agriculture.
Using cover crops as an integrated management practice can help make agriculture more sustainable by contributing to pest control and ideally reducing the need for pesticide applications. Reducing pesticide applications not only limits harmful environmental effects like run off and non-target effects, but can also save the growers money. Cover crops can benefit overall plant health and improve yields as well. Ultimately, cover cropping is a sustainable and economically beneficial practice for growers.
Before this project, I did not know much about sustainable agriculture or much about agricultural practices in general. During the course of this project, I learned that cover crops can be a major part of sustainable agriculture. Not only can they benefit soil and plant health, but they can also be used in an integrated pest management strategy. Cover crops can increase beneficial insects that in turn may contribute to pest control. They can also help suppress diseases in the cash crop. After this project, I've become much more aware of the aspects of sustainable agriculture and how cover crops can play an important role.
In future studies, it would be interesting to look at the impact of cover crops on diseases in watermelon more directly. Field inoculations with the pathogen could help to more accurately determine if cover crops affect watermelon disease. It may also be beneficial to look at additional flowering winter cover crops and various fall planting dates to determine which cover crops could flower before termination and provide resources to pollinators. Investigating the impact of cover crops on insects in other production systems could be interesting as well.