The Tropical Rain Forest Canopy: A Method Providing Total Access

Abstract

An aerial rope network was constructed, using three emergent trees as supports, which provides access to a large volume of tropical rain forest from ground level to above the canopy's upper surface. The virtually unexplored canopy community thus becomes accessible for a broad range of scientific research. THE CANOPY OF THE TROPICAL FOREST possesses one of the most complex and diverse communities on the earth, yet there have been few effective methods for studying this aerial zone, and none of these provides comprehensive access to a large volume of forest. To observe the canopy, which varies in height from about 15 meters to above 60 meters, early investigators built towers and platforms in tall trees (Hingston 1932, Bates 1944, McClure 1966). Immobile structures such as these proved to be biased observation posts due to the high spatial heterogeneity in plant species and associated animal activities of tropical forests (Elton 1973). The usefulness of treetop platforms was expanded by Muul and Liat (1970), who built transects of catwalks extending hundreds of meters at various heights in the canopy. This method faciliated observations, but access to surrounding vegetation was severely limited. Further, catwalks were not ecologically benign since they offered new routes for canopy animals that in turn could influence colonization patterns on nearby limbs by epiphytes (Perry 1978b). Using advances made by Denison et al. (1972), Perry (1978a) developed highly mobile tree-climbing methods that made the peripheral regions of every strong tree and the volume of forest under its crown accessible to investigation. Nevertheless, important regions of the forest remained inaccessible: i.e., the upper surface of the canopy with its high activity of insect and bird species, and a large fraction of the forest which is composed of weak trees unsafe for climbing, whose heights commonly reach 30 meters. To gain complete access to the latter regions for studies of pollination biology, we developed an aerial network of ropes, which provided access to about an acre of forest from ground level to above the historically inaccessible upper surface. The study site was Finca La Selva, a field station owned by the Organization for Tropical Studies, located near Puerto Viejo, Heredia Province, Costa Rica. The work required only two people and was begun on 8 March, and ended on 1 April 1979. Following the lead of Muul and Liat (1970), who used trees as structural supports for their catwalks in a Malaysian forest, our minimum canopy research facility was constructed using three emergent trees in undisturbed lowland rain forest of the southeast corner of the University of Washington Research Plot I. The trees formed a triangle of approximately 100 meters on a side. Each was taller than the surrounding canopy by about 15 meters, a height that was essential to the function of the web. A small platform was constructed in one tree at a height of 32 meters which serves for sleeping, equipment storage, and access to the web. From the platform we shot 20 lb. test monofilament lines, using a crossbow and arrows, to and between the two opposite trees, making a continuous loop (fig. 1). The position of these initial lines was adjusted to avoid abrasion against limbs, which would weaken the ropes when the web was in operation. The lines were replaced by 200 lb. test braided Nylon cords that were strong enough for use in pulling the 8600 lb. test, 1/2 inch in diameter, perimeter rope into place. The rope brand is Continental Dacron over Dacron, and is available through Continental Western Corporation, 2931 South Avil Avenue, Commerce, California 90040. Dacron was selected because it is nearly stretch-free and resistant to sunlight, though this rope does abrade easily. Because of this hazard, all ropes, especially where they extend into a tree crown, should be clearly visible so they can be inspected regularly. Rodents should not be overlooked as a source for abrasion since they may BIOTROPICA 13(4): 283-285 1981 283 This content downloaded from 157.55.39.72 on Wed, 14 Sep 2016 06:13:03 UTC All use subject to http://about.jstor.org/terms