Sequential sampling for panicle caterpillars (Lepidoptera: Noctuidae) in sorghum

Area of Science:

• Agricultural Entomology
• Pest Management
• Statistical Ecology

Background:

• Panicle caterpillars (corn earworm and fall armyworm) are significant pests of sorghum in the Great Plains.
• Current fixed-size sampling methods (30 beat bucket samples) are labor-intensive and may not be optimal.
• Understanding pest distribution is crucial for developing efficient sampling strategies.

Purpose of the Study:

• To develop and validate a sequential sampling plan for panicle caterpillars in sorghum.
• To improve the efficiency of pest monitoring compared to traditional fixed-sample size methods.
• To establish economic and safe thresholds for pest management decisions.

Main Methods:

• Utilized Wald's sequential probability ratio test for a negative binomial distribution.
• Sampled 115 sorghum fields across Kansas, Oklahoma, and Texas (June-August 2010).
• Developed a plan with high (0.5 caterpillars/panicle) and low (0.20 caterpillars/panicle) thresholds and fixed error rates (α=0.10, β=0.05).

Main Results:

• Panicle caterpillar counts exhibited an aggregated distribution.
• The sequential plan required fewer samples than the fixed 30-sample method when densities were high or low.
• Validation in 2013 showed strong agreement (r²=0.93) between expected and actual samples, with correct field classifications.

Conclusions:

• The developed sequential sampling plan is more efficient than current fixed-size methods for sorghum panicle caterpillars.
• This plan optimizes sampling effort, reducing labor when pest densities are far from thresholds.
• Accurate pest monitoring through sequential sampling supports timely and effective integrated pest management strategies.