A quantitative method for assigning abundance classifications to leaf beetles (Coleoptera: Chrysomelidae) found in Kentucky

Background/Question/Methods

Little is known regarding the distribution and relative abundance of leaf beetles (Coleoptera: Chrysomelidae) within the United States.  This extremely diverse family of phytophagous beetles includes many species reliant on a single species or family of host plants, making them likely indicators of biodiversity.  In addition, while relative abundance classifications can be quite helpful in comparing species composition data, the assignment of those categories is often subjective.  Our objectives were 1) to use all known specimen data for leaf beetles from Kentucky to develop an index for abundance classification of those species, 2) to compare index-based classifications to subjective classifications assigned to a subset of species, and 3) to use the index-based classifications to compare species composition of leaf beetles from sites of varying habitat quality: a remnant native grassland, a semi-natural prairie adjacent to a railroad, and a research farm.  We first assigned each of 149 species collected during recent intensive collection efforts one of the following classifications: abundant, frequent, infrequent, local, or rare.  Using museum specimen data (10,969 beetles), we then calculated a simple abundance index as A = C + N + T, where A is a species’ abundance index, C is the number of Kentucky counties in which a species has been found, N is the number of specimens collected outside state nature preserves, and T is the percentage of total specimens the species represents.   

Results/Conclusions

We identified 288 taxa, including 252 known species whose calculated indices ranged from 1.01-740.55.  We classified 19 species as abundant, 20 frequent, 43 infrequent, 40 local, and 130 rare.  Of the 149 species included in the initial subjectively ranked subset, 27 (18%) differed in subjective and index-based classification.  When the index-based rankings were applied to species from the three habitats, we found the greatest percentage of rare species in the native grassland (42%), and conversely, the greatest percentage of abundant species from the farm (37%).  The railroad prairie was intermediate between the two, with comparable proportions of rare and abundant species (18% vs. 13%).  Our results indicate that quantitative data can be used to establish abundance classifications in an approach that is likely more consistent than subjective assignments.  Once established, these classifications can be used to compare species composition of different habitats beyond the basic measures of species richness and diversity, or in cases where the dataset does not allow for calculation of these most commonly used biodiversity measures.