Vegetation can play a major role in the hydrologic ecosystem service tradeoffs resulting from land use change: by affecting the volume of rain water that reaches the ground surface, vegetation affects water supply. But because canopy interception of rainfall is affected in complex and competing ways by forest structure and ambient weather conditions, both of which vary at small and large scales, predicting the impacts of vegetation change is challenging. We explore rainfall and cloud interception in two native forests sites on leeward Hawai'i Island and find that although our study forests are superficially similar, with identical dominant species and no history of logging, throughfall in one forest is nearly double that in the other. Using micrometeorological and vegetation data collected over 20 months, we examine the dominance of different hydrologic processes at each site. Direct fog water input accounted for at least 12% of total water input in the North forest site, an average of about 0.1 mm/day, and 27% of total water input in the South forest, an average of 0.3 mm/day. In the North forest, canopy interception of rainfall dominates, and annual throughfall amounts to only 64% of rainfall. A large canopy surface area and low rainfall rates cause the high rate of interception. Direct interception of clouds by the canopy dominates in the South forest, where throughfall is 113% of rainfall. Increased throughfall at this site is not attributable to increased fogginess. Instead, taller trees and a denser mid-canopy increase canopy surface area, causing increased cloud interception. The denser forest structure is likely a result of cattle exclusion and limited grazing in this forest. This study illustrates the effects of subtle differences in vegetation structure on hydrologic fluxes and, by extension, the hydrologic effects of land use change. It also underscores the importance of replicate sites in ecohydrologic investigations.
Bibliographical noteFunding Information:
This project would not have been possible without the support and friendship of J, D, K, and J Fergerstrom; J, E, and M Greenwell; and the staff of the Kamehameha Schools Land Assets Office in Kona. Kamehameha Schools, Kealakekua Ranch, and Palani Ranch graciously allowed us to erect weather stations on their property. Jody Fergerstrom provided invaluable field assistance. The authors are most grateful for financial support from The Center for Conservation Biology, IPER Rudolf Research Fellowship, The Koret Foundation, The Moore Family Foundation, The Lucille and David Packard Stanford Graduate Fellowship, McGee Stanford School of Earth Sciences Research Funds, National Science Foundation Graduate Research Fellowship, Peter and Helen Bing, The William C. and Jeanne M. Landreth IPER Student Fellowship, and The Winslow Foundation. C. Allred, G. Asner, J. Berry, T.K. Duarte, D. Fergerstrom, C. Field, M. Garces and the UH Infrasound Lab, J. Goldstein, L. Goulder, N. Hannahs, G. Hendrickson, F. Hughes, C. Hunnings, A. Kagawa, R. Martin, P. Matson, G. Mendoza, L. Pejchar, H. Tallis, B. Thompson, P. Vitousek, A. Wolf, S. Wolny, and two anonymous reviewers provided insight and advice.
- Canopy interception
- Fog drip
- Rainfall partitioning
- Tropical montane cloud forest