Location: Pest Management and Biocontrol Research
Project Number: 2020-30400-001-007-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: Aug 1, 2026
End Date: Jan 30, 2028
Objective:
The cotton seed bug (CSB), Oxycarenus hyalinipennis, is a notorious pest of cotton. It feeds on cotton seeds, and when crushed during harvest and ginning, they stain cotton red, reducing the quality of highly preferred white cotton fiber. This pest, native to Africa, became established in urban areas of California in 2019. Facilitating their successful invasion was the capacity of CSB to feed on the seeds of a very large number of plant species that are related to cotton. These non-cotton host plant relatives include common ornamental species like hibiscus, weeds, like cheese weed, and native plants, like mallows. CSB is using species of non-cotton host plants to spread throughout urban and wilderness areas. Because of this spread, CSB threatens to invade cotton producing regions, threatening the profitability of US cotton producers. Unfortunately, mitigation efforts are hampered by a dearth of information on CSB biology, particularly regarding how populations might develop in US cotton once they invade major cotton producing areas.
To address this informational deficit, we will undertake the development of a proactive approach to managing CSB. This project has three objectives: (1) to study the invasion processes that result in CSB moving from non-cotton host plants, like cheese weed, a common weed that grows abundantly in agricultural areas, into cotton fields; (2) to measure the rate at which CSB populations increase in cotton over the course of a growing season, and quantify the damage this pest causes to cotton seeds and lint; and (3) to identity natural enemy species in CSB infested cotton and to the assess the impacts these natural enemies have on CSB populations infesting cotton. The experiments will collect data from isolated cotton fields planted in already infested areas. Together, data from these objectives will provide key information on the propensity and time of year CSB is likely to invade cotton from non-cotton host plants, the speed at which populations develop in cotton, the likely economic cost of unchecked invasion, and the levels of control natural enemies may provide in cotton. The information will be used to develop a management strategy for growers that can be shared before CSB becomes established in cotton production regions.
Approach:
Field experiments proposed here will be run in experimental cotton plots in southern California that are hundreds of miles away from commercial cotton production areas in California. Two 1-acre plots will be set up at Agricultural Operations, University of California Riverside. Each plot will consist of half an acre of Upland cotton and half an acre of ThryvOn cotton. ThryvOn is marketed for protection against piercing-sucking insect pests, like CSB. Sampling of CSB and natural enemies will commence once cotton plants have five true leaves and before the first square (i.e., flower bud) is present. Once this phenological growth stage is reached, plant vacuum sampling of six short row stations, consisting of five plants, will be conducted every two weeks in Upland cotton and ThryvOn cotton sections of experimental plots. This will result in 30 sampled Upland cotton plants and 30 ThryvOn cotton plants per plot for a total of 60 cotton plants sampled every two weeks. Vacuum samples will be labeled by date, sampled row, and cotton type then frozen. Dead insects will be sorted into functional groups: (1) CSB nymphs and adults, (2) natural enemies, with a particular focus on predators (e.g., Geocoris, Orius, Nabis, spiders, lacewing larvae, and ladybug larvae and adults), (3) other sucking and chewing insects that can damage cotton bolls and provide an entry point for CSB.
CSB and natural enemy data will be recorded by sample date, row, and cotton variety. Cotton plant phenology will be recorded each time vacuum sampling is done. Collected data will include, plant height (cm), number of nodes, mainstem node of first square, number of squares, flowers, green bolls, and open bolls. These data will be used to calculate mean number of and percent plants with squares, flowers, and bolls over time. Sampling will commence before CSB is present in the crop as this will establish the “zero” baseline and subsequent sampling will capture early colonization and natural enemy presence before reproductive structures appear in the crop. CSB biology dictates that once cotton plants enter reproductive development, especially from formation of the first square through to boll opening, the crop will be most vulnerable to this pest and this is when population build up is most likely to begin.
Generalized linear mixed models with a negative binomial distribution will be used to analyze CSB population growth. The response variable will be CSB, fixed effects will include cotton type, sample date, and cotton plant growth stages. Random model effects will be plot, and sample row station nested within plot and cotton type. Similarly for predator count data, generalized linear mixed models with either a Poisson or negative binomial distribution. The latter accounts for over-dispersion due to zero-inflated counts. The response variable will be predator density, and model fixed effects will include cotton variety, sampling data, and CSB densities. Random model effects will be plot, and sample row station nested within plot and cotton type. Predator-prey ratios will be used to assess the magnitude of pressure applied by natural enemies on CSB populations.