Binomial sampling to estimate rust mite (Acari: Eriophyidae) densities on orange fruit

Area of Science:

• Agricultural Entomology
• Pest Management
• Quantitative Ecology

Background:

• Accurate estimation of citrus pest populations is crucial for effective management.
• Traditional mite counting methods can be labor-intensive and time-consuming.
• Developing efficient sampling strategies is key for integrated pest management programs.

Purpose of the Study:

• To investigate the efficacy of binomial sampling for estimating citrus rust mite (Phyllocoptruta oleivora and Aculops pelekassi) densities on oranges (Citrus sinensis).
• To determine the relationship between the proportion of infested samples and mean mite density.
• To assess the viability of binomial sampling as an alternative to absolute mite counts.

Main Methods:

• Collected data from 600 sample units (1-cm2 per fruit) across a 4-ha orange grove.
• Analyzed the relationship between the proportion of samples with mites (P0) and mean mite density using a linear model.
• Validated the fitted binomial parameters with independent data sets and projected applicability to smaller sample sizes.

Main Results:

• A significant linear relationship (r2 = 0.89) was established between ln(-ln(1 -P0)) and ln(mean).
• Fitted binomial parameters accurately described validation data and were projected to be applicable to sampling plans with as few as 48 samples.
• Confidence limits for mean estimates increased with higher proportions of infested samples, indicating reduced precision at high densities.

Conclusions:

• Binomial sampling is a viable alternative to absolute mite counts for estimating citrus mite densities, particularly for growers with low management thresholds.
• While effective, the value of binomial sampling with a zero tally threshold decreases as infestation levels rise.
• Adjusting the tally threshold to two mites per sample marginally improved estimates at higher mite densities.