SHORT COMMUNICATION EVALUATING BIODIVERSITY THROUGH FUNCTIONAL GROUPS: A COMPARISON OF FUNCTIONAL GROUPS AND BIODIVERSITY MEASURES Richard Moles and Kevin Hayes Richard Moles (corresponding author, e-mail rnchard moles@ulie), Department of Chemical and Environmental Sciences, University of Limerick, Limerick, Republic of Ireland, Kevin Hayes, Department of Mathematics and Statistics, University of Limerick, Limerick, Republic of Ireland Received 12 February 2001 Read 16 March 2002 Published 30 November 2002 INTRODUCTION Quantification of blodiversity is crucial to development of policy aimed at the conservation of flora and fauna, and ecologists are therefore under pressure to arrive at an agreed mode of evaluating biodiversity (Hunter 1996). However, difficulties arise in defining blodiversity because it is studied at different levels, usually distinguished as the genetic, species, community and landscape levels, each of which constitutes a different facet of the subject (Noss 1992). Where a decision ismade to focus on the community level, further problems arise in quantifying biodiversity. Communities are to a greater or lesser extent multispecific, and taxonomic and time constraints prevent the identification of all species present. This has resulted in the adoption of indicator taxa and functional groups as a pragmatic means of approximating community biodiversity (McLaren et al. 1998, Williams 1998). Some studies have found relationships between functional groups, for example butterfly and vegetation communities, and have suggested that a single functional group may be used as an indicator of landscape-scale ecological processes (Oostermeljer and van Swaay 1998; Negi and Gadgil 2002). However, it has been argued that such functional groups are often selected on the basis of popular attractiveness and ease of survey, rather than on ecological criteria such as importance to ecosystem functioning (Hunter 1996), and that the outcomes of biodiversity quantification based on such indicators may relate more to functional group choice than to attributes of the community (Walker 1989). If it is decided to attempt quantification of the biodiversity of a community through a survey of indicator functional groups, then a method of quantification must be selected. Many such methods have been suggested, but perhaps themost frequently adopted methods are (1) to count the number of species and (2) to combine the number of species and the distnbution of individuals among species within a diversity index (Magurran 1988). This short commuunication aims to provide information relevant to answering the following two questions Does the choice of indicator functional group affect the estimation of blodiversity? Does the choice of quantification method affect the estimation of blodiversity? METHODS To assess the effect on blodiversity of a range of farming practices in County Down, Northern Ireland, surveys of flora and fauna were carned out in 1971-2 in five lowland farmland blocks, each approximately 18ha in extent (Moles 1974) and representative of a > 4km2 landscape unit Dunng the surveys, fields in two blocks (Gl and G2) were under permanent grass, another two blocks (Ti and T2) were planted with mixed crops, including cereal, root and green crops, and the fifth block (Ml) contained amixture of grass (38% by area) and tilled (62% by area) fields. Surveys were undertaken of six functional groups: Breeding birds,Wintenng birds, Butterflies, Moths, Woody plants and Forbs (Moles 1974) Selection of functional groups was based on (a) the assumption that to facilitate interblock comparisons, indicators should represent awide spectrum of genetic diversity, and (b) the pragmatic requirement that they be quickly and reliably identifiable to species For Breeding birds, five counts per block were carned out between mud-May and the end ofJune, each count lasting 3h and starting lh after dawn on days without mist, rain or strong winds. Over-flying individuals were not counted The location of each individual was recorded on maps The presence of breeding pairs was establhshed through comparnson of the five count maps, and terntory boundanes were estimated on the basis of observation of occupied nests, juveniles, singing BIOLOGY AND ENVIRONMENT PROCEEDINGS OF THEROYAL IRISHACADEMY, VOL 102B, No 2, 113-117 (2002) C ROYAL IRISHACADEMY 113 BIOLOGY AND ENVIRONMENT males, fighting, and carrying of nest matenal and food. Where individuals were recorded at simular locations on four or five counts, or where nesting was confirmed, this was accepted as evidence that apairwas resident Where individuals were recorded...