SOUTH DAKOTA COUNTY RISK ASSESSMENT MODEL FOR CORN INSECT PESTS
North Central Agricultural Research Laboratory
2013 Annual Report
1a.Objectives (from AD-416):
The objective of this seed grant proposal is to develop a two stage risk assessment tool. The outcome anticipated is a prototype risk assessment model that provides reliable forecasting capabilities for estimating the risk from a specific arthropod becoming established in corn production areas in South Dakota at the county level. The risk assessment will be based on county characteristics favorable to establishment of high consequence arthropods in a specific South Dakota county.
1b.Approach (from AD-416):
A two stage model is proposed. The first stage of the model will generate ordinal quantitative risk assessment estimates for arthropod establishment-risk based on county characteristics using a “scorecard qualitative assessment approach.” The second stage generates a probability estimate (a cardinal measure) for an arthropod of high economic consequence becoming established in a South Dakota county, based on the first stage risk assessment rankings. We intend to collect data on: a) county attributes associated with risk establishment, and b) county level arthropod infestation data based on standardized sampling procedures targeting seven herbivorous pests that represent different taxonomic and feeding guilds consistently problematic for conventional corn fields in South Dakota. The data will be used to calibrate the two-stage model.
The first stage of the two-stage model will be designed in accordance with the methodology developed in the country risk analysis literature. The country risk analysis methodology will be modified to provide qualitative risk rankings for county attributes correlated with successful pest establishment. The second stage will develop a logistic regression model that will relate the incidence of successful establishment in a county (sample data) to variables expected to influence the likelihood of successful establishment (county risk attributes).
Our proposed high consequence corn infesting arthropod risk assessment project overcomes two flaws identified in the non-indigenous species (NIS) risk assessment literature:.
1)our proposed county level risk assessment model provides cardinal rankings rather than ordinal rankings and thus provides a probability ranking for pest establishment risk; and.
2)our proposed model focuses on only the establishment stage of invasion at the county level.
The primary objective of the project was to develop a better understanding of the insects currently inhabiting South Dakota cornfields. A secondary objective was to develop a risk prediction model for insect corn pests. The corn pest survey project collected insect data during the tassel stage of corn production in eastern South Dakota. In 2010, the survey team collected insect samples from 28 counties in eastern South Dakota and in 2011 the survey team visited 25 counties. Only refuge corn acres were surveyed. All insect functional groups were documented during the collection process. Survey reports were written for each survey year. Individual farm specific reports on insect populations were provided to each producer. Each producer also received a detailed report on aggregate insect populations found in eastern South Dakota refuge corn fields. Thirty-eight corn producers participated in this study over the 2-yr period. Insects collected were categorized as caterpillars, aphids, or rootworms. The caterpillar pests were made up of western bean cutworm, European corn borer, and corn earworm. The corn rootworms included western, northern, and southern rootworms and aphids (which were primarily corn aphids). These pests never exceeded economic thresholds at any of the sites. European corn borer was the most consistent pest found, but the maximum infestation level was 20% of the plants infested at a single site. A likely factor for this low level of pests may be the area-wide suppression of pests due to high adoption rates of genetically modified (GM) crops in South Dakota. We found a diverse community of beneficial insects (insect predators) at all of the farms we sampled, including spiders, lacewings, pirate bugs, lady beetles, and syrphid flies. Densities ranged from 1-8 predators per plant. The average planting density in SD is 31,174 plants per acre, giving a back of the envelope estimate of about 249,000 beneficial insects per acre living just in the plant foliage. Scientific manuscripts are in preparation on these surveys.