Sampling plan for the coffee leaf miner Leucoptera coffeella with sex pheromone traps T. Bacca 1,2 , E. R. Lima 2 , M. C. Picanc ¸o 2 , R. N. C. Guedes 2,3 & J. H. M. Viana 4 1 Faculdad de Ciencias Agrı ´colas, Universidad de Narin ˜ o, Torobajo, Pasto, Narin ˜ o, Colombia 2 Departamento de Biologia Animal, Universidade Federal de Vic ¸ osa, Vic ¸ osa, Brazil 3 Department of Biology, Carleton University, Ottawa, Canada 4 Nu ´ cleo de Agroecossistemas e Sustentabilidade Agrı ´cola, Embrapa Milho e Sorgo, Sete Lagoas, Brazil Introduction Integrated pest management programmes require precise and reliable methods to estimate density of pest insects and their natural enemies. The use of sampling plans allows the estimation of population density for pest control decision-making (Norris et al. 2003). Conventional and sequential sampling plans are frequently used to sample populations of the coffee leaf miner Leucoptera coffeella (Gue ´ rin- Me ´ neville & Perrottet) (Lep., Lyonetiidae), the main coffee pest in Brazil and a key coffee pest through- out neotropical America (Pereira et al. 2007). These sampling plans require large amounts of leaves to determine the percentage of damaged leaves, demanding a great sample effort and leading to a delay in the decision-making process (Gravena 1983; Bearzoti and Aquino 1994; Souza et al. 1998; Vieira et al. 1999). Traps with attractants, such as pheromones, have been successfully used for insect-pest sampling because they are a quick method and easy to use (Jones 1998). The use of this technique for the cof- fee leaf miner is possible because its sex pheromone is known and male capture with pheromone traps is correlated with pest damage (Francke et al. 1988; Keywords Coffea arabica, coffee pest, Lyonetiidae, sample size, spatial analysis, trap density Correspondence Tito Bacca (corresponding author), Faculdad de Ciencias Agrı ´colas, Universidad de Narin ˜ o, Torobajo, Pasto, Narin ˜ o, Colombia. E-mail: titobacca@gmail.com Received: July 6, 2007; accepted: November 23, 2007. doi: 10.1111/j.1439-0418.2007.01264.x Abstract The population density of the coffee leaf miner Leucoptera coffeella (Gue ´- rin-Me ´ neville & Perrottet) (Lep., Lyonetiidae) can be estimated using pheromone traps in coffee fields as male capture reflects this pest dam- age based on previous correlational study. However, the spatial distribu- tion of pheromone traps and their density are necessary to optimize the sampling procedure with pheromone traps. Therefore, the objectives of the present study were to determine the pheromone trap density required per hectare to sample coffee leaf miner populations and to determine the spatial distribution of the males of this pest species. The males were sampled every 8 days in 12 consecutive evaluations. Taylor’s power law and frequency distributions were used to recognize the distri- bution of the male capture data, which followed a negative binomial distribution. A common K was obtained, allowing the establishment of a single conventional sampling plan for the 12 fields investigated. The adjusted sampling plan requires eight traps in an area of 30 ha for a 25% precision error. Kriging-generated maps allowed the simulation of male captures for 8, 12 and 20 traps per 30 ha and the results were compared with those obtained with absolute sampling resulting in R 2 -values of 0.30, 0.57 and 0.60 respectively. The traps were able to identify the more highly infested areas within the field and are a precise and efficient tool for sampling populations of L. coffeella. J. Appl. Entomol. 430 J. Appl. Entomol. 132 (2008) 430–438 ª 2008 The Authors Journal compilation ª 2008 Blackwell Verlag, Berlin