A new paper detailing research from the rainforests of Panama provides an unprecedented level of detail regarding the diversity and distribution of arthropod species from the soil to the forest canopy - the authors estimate that a 6,000 hectare forest hosts a total of around 25,000 arthropod species, a figure vastly outnumbering that of better-studied organisms.  

Most multicellular species on Earth are arthropods living in tropical forests. Yet, given the difficulties involved in merely tallying them, we know very little about their exact numbers– even at the scale of a single forest. A massive collaborative effort involving 102 researchers from 21 countries was necessary to collect and identify arthropods from all parts of the rainforest. During 2003-2004, the field team spent an effort of nearly 70 person- or trap-years in sampling the rainforest canopy from a construction crane, inflatable platforms, balloons, climbing ropes through forest layers, as well as crawling along the forest floor to sift soil, and trap and bait arthropods.

During the ensuing eight years, the team sorted and identified 130,000 arthropods, to a total of more than 6,000 species.

By scaling up the diversity values obtained from twelve intensively-sampled sites, the team was able to calculate that the rainforest reserve harbors in excess of 25,000 arthropod species. "This is a high number as it implies that for every species of vascular plant, bird or mammal in this forest, you will find 20, 83 and 312 species of arthropods, respectively," explains  Yves Basset, scientific coordinator of the CTFS Arthropod Initiative at the Smithsonian Tropical Research Institute. "What surprised us the most was that more than half of all species could be found in a single hectare of the forest."

"This is good news, as it means that to determine the species diversity of a tropical rainforest, we need not sample gigantic areas: a total of one hectare may suffice to get an idea of regional arthropod richness – provided that this total includes widely spaced plots representative of variation within the forest," said Tomas Roslin from the University of Helsinki, one of 35 co-authors.

"Another exciting finding was that the diversity of both herbivorous and non-herbivorous arthropods could be accurately predicted from the diversity of plants", says Basset. "While we have assigned immense resources to mapping our genes, resolving sub-atomic structures and searching for extra-terrestrial life, we have invested much less in exploring with whom we share the Earth. Why such research should be run on a shoe string budget just escapes me." 

 Published in Science.