Assessing biodiversity 'sampling packages': how similar are arthropod assemblages in different tropical rainforests?

Arthropod sampling protocols have been applied to four areas of undisturbed rainforest in Australia and Borneo at sites located between 28 degrees S and 4 degrees N. Seven different trapping methods have been used in each case following a strictly comparable protocol applied during single wet-season surveys at each location. A number of large scale patterns across sites correlated with latitude, altitude and/or biogeographical history are evident at the ordinal level. The results are used to evaluate two key ideas in biodiversity survey. The first tests questions of the constancy of particular trap types in terms of the target animals. The second attempts to design an efficient multiple method sampling 'package' for rapid biodiversity assessment. Across all sites similar traps behaved in a similar manner in terms of the profile of arthropods that they caught although Malaise traps and pitfalls are particularly 'constant' in this regard. Subsets of both trapping methods and target taxa are identified which will maximise the opportunities for detecting between site differences.