Link to USGS home page



Venture Capital Project Final Report, July 2004


Quantifying the importance of fog drip to ecosystem hydrology and water resources in windward and leeward tropical montane cloud forests on East Maui, Hawaii


Research Team

Martha Scholl, USGS, WRD, National Research Program, Reston, VA

Stephen Gingerich, USGS, WRD, Hawaii District, Honolulu, HI

Lloyd Loope, USGS, BRD, Haleakala Field Station, Makawao, HI

Thomas Giambelluca and Michael Nullet, University of Hawaii, Geography Department, Honolulu, HI



Understanding the hydrology of tropical montane cloud forests has become essential as deforestation of mountain areas proceeds at an increased rate worldwide.  Accurate water balances that include cloud water or fog (precipitation not captured by a standard rain gage) are needed.  It is also important to quantify fog contribution to recharge and stream flow, and to understand the role of fog water in the forest ecosystem in terms of nutrient and pollutant inputs, and as a moisture source during rainless periods.  Previous studies have shown that the isotopic signature of fog tends to be more enriched in the heavier isotopes 2H and 18O than that of rain, due to differing condensation history and temperature.  Monitoring the isotopic signature of surface water, soil water and tree sap therefore may yield estimates of the contribution of fog water to streamflow, recharge and transpiration.  Instrumentation to measure fog input includes passive and active cloud water collectors, throughfall collectors, and visibility or droplet size measurements in combination with meteorological sensors (such as temperature, humidity, wind speed and direction measurements).  Distinguishing fog events, fog / rain events, and rain events aids in determining the rates of cloud water deposition to a forest, and how much water input would be missed by using conventional rain gage-based water balance techniques.

Quantifying fog deposition is most difficult in cloud forests that receive mixed precipitation, as there are no established methods to separate fog deposition from rain deposition on passive collectors.  Another measurement problem occurs in areas with low water content clouds that may not deposit measurable amounts on instruments, but still have important effects on the forest via interception of droplets by the canopy and reduction of transpiration from plants.  Throughfall measurements provide another way to assess fog deposition, but quantifying the surface area of the vegetation that scavenges the water from the clouds remains a challenge.  This study involved two years of monitoring and isotopic analyses at two sites, one each on the windward and leeward sides of Haleakala volcano (East Maui).  By gaining an understanding of the role of fog in forest hydrology on both sides of the island, we hope to learn what factors are most important in restoration of the forests on the leeward side.  On the windward side, quantifying the amount of fog drip in streamflow and transpiration will help in predicting possible effects of forest degradation on the water resource.  The techniques developed during this study will have application to montane cloud forests worldwide. 



How much does fog contribute to cloud forest water budgets?  Do plants utilize fog water, rain water, or both?  How much does fog contribute to streamflow?  What factors contribute to fog-interception efficiency of the forest?  To help answer these questions, we proposed to use stable isotopes in combination with intensive monitoring of precipitation input (fog and rain) to quantify and compare the role of fog drip in the ecohydrology of montane forests on the windward and leeward sides of East Maui. 



Our approach consisted of two years of data collection at two mountain forest sites, one each on the windward and leeward sides of East Maui (finished December 2003), followed by analysis of the data and development of new isotopic mixing models to identify water sources in forested catchments, new methods for comparing fog and rain deposition on instruments with deposition on the forest canopy, and the inclusion of fog as a component of the regional water balance (ongoing). 

Cumulative rainfall, cumulative fog deposition, and grab samples of streamflow, transpiration water (tree sap), and soil water were taken approximately monthly and analyzed for the stable isotopes deuterium and oxygen-18, to trace fog water through the hydrologic cycle.  Wind speed and direction, incident radiation, air, surface and soil temperature, soil heat flux and soil moisture, relative humidity, and rainfall were measured with weather stations, in order to characterize site climatology, to help separate rain events from fog events, and to obtain data for calculation of potential evapotranspiration.  Fog deposition (amount and frequency) was measured on a vertical 1-m2 screen connected to a tipping bucket rain gage.  Four throughfall collectors with three 6-m long collection troughs (0.7 m2 collection area) were placed under the canopy at each site.  An optical visibility sensor provided information on the occurrence of fog events that did not deposit measurable amounts of precipitation on the instruments.

The leeward fog site (Auwahi) was adjacent to the site of the BRD/UH projects “Developing a Listening Post in the Tropical Pacific: Sensitivity of Hawaiian High Elevation and Aquatic Ecosystems to Global Change” and the Auwahi Dryland Forest Restoration.  This site contains numerous endangered endemic species.  Dry forests were the most diverse ecosystems in the Hawaiian Islands, and Auwahi is one of the few remaining patches of dry forest left.  The area is sparsely forested, on a steep slope within pastureland at 1220 m altitude, near the lower altitude boundary of the fog zone.  The windward site was in the upper Honomanu Stream drainage basin within the Nature Conservancy’s Waikamoi Preserve.  The forest here also contains endangered native species, and represents the upper boundary of the cloud/rain forest, where it transitions to open shrub land at 1950 m altitude. 


Results and Preliminary Conclusions


Data analysis is ongoing at this time, and several proceedings papers and journal papers are in preparation.  Results thus far are summarized below:  

Stable Isotopes:

At the windward site (Waikamoi), the mountainside is frequently enveloped in raining clouds.  The isotopic composition of stream water at each sampling date was best explained by invoking a mixture of shallow groundwater and the collected fog water for the previous sampling period, indicating that cloud water contributes substantially to stream flow at this site.  The soil water did not have isotopic composition similar to the stream water, it had a larger proportion of precipitation from large rainstorms.  Metrosideros polymorpha (ohia) tree sap isotopic composition was sometimes identical to fog water, almost never similar to soil water, and generally appeared to be a mixture of fog and rainfall. 

At the leeward site (Auwahi), most water input is from infrequent large storm systems, while the clouds that envelop the site on a more frequent basis have relatively low water content.  Soil water isotopic composition reflected the large rainstorms, with little cloud water input.  Isotopic composition of M. polymorpha sap at this site was similar to the soil water, suggesting that the trees tap a deeper water source than at the windward site.  Future analyses will include detailed isotopic mixing models to quantify the different sources of water in the components of the catchment water balance, including vapor sources in the orographic cloud on the windward slope.  

Cloud Water Input: 

Results from analysis of the cloud water collector data indicate that cloud water interception adds significantly to rainfall at both dry (leeward) and wet (windward) cloud forest sites in Hawai'i.  Cloud water interception was equivalent to 268 mm yr-1 at Auwahi and 1073 mm yr-1 at Waikamoi.  Measured throughfall was about 65% of incident rainfall at Auwahi, and 119% of incident rainfall at Waikamoi.  Throughfall was dominated by rainfall at both the Waikamoi and Auwahi sites, and was significantly influenced by fog only at the windward site.  Very high rates of cloud water flux and cloud water interception were estimated for the windward site.  At the leeward site, a high percentage of cloud water flux was intercepted, but most of the intercepted cloud water was evaporated before reaching the ground.  We are currently analyzing additional data from the two sites and refining methods of analysis. The canopy water balance estimates will be integrated with estimates of cloud water input based on the stable isotope analysis.



Scholl, M.A., Eugster, W., and Burkard, R., in press, Understanding the Role of Fog in Forest Hydrology:  Stable Isotopes as Tools for Determining Input and Partitioning of Cloud Water in Montane Forests, in:  Juvik, J., Bruijnzeel, L.A., Scatena, F.N., and Bubb, P., Proceedings of the Second International Symposium: Science for Conserving and Managing Tropical Montane Cloud Forests, Waimea, Hawaii, July 27 – August 1, 2004.

Giambelluca, T.W., DeLay, J.K., Nullet, M.A., Scholl, M.A., and Gingerich, S.B., in press, Cloud water interception at wet and dry forest sites on Haleakala, Maui, Hawai'i, in:  Juvik, J., Bruijnzeel, L.A., Scatena, F.N., and Bubb, P., Proceedings of the Second International Symposium: Science for Conserving and Managing Tropical Montane Cloud Forests, Waimea, Hawaii, July 27 – August 1, 2004. 

Scholl, M.A., Gingerich S.B., Giambelluca, T.W., Nullet, M., Loope, L.L., 2003, Quantifying the Role of Cloud Water in the Hydrology of Two Montane Forest Sites on East Maui, Hawaii, in: EOS, Transactions of the American Geophysical Union, v. 84, no. 46, p. F687, Fall Meeting, December 8-12, 2003

Gingerich, S. B., and M. A. Scholl, 2003, The contribution of cloud water to recharge on the volcanic island of Maui, Hawaii.  Presented at International Union of Geodesy and Geophysics Meeting, July 2003, Sapporo, Japan.

Yuen, E., 2003, Will water return to Kaho'olawe?  Maui forests hold tantalizing clues, Environment Hawaii, v. 13, no. 9, p. 1-7.

Scholl, M.A., Gingerich S.B., Giambelluca, T.W., Nullet, M., Loope, L.L., 2002, The role of fog in ecosystem hydrology: initial results from investigations using stable isotopes of water in Hawaiian cloud forests in: EOS, Transactions of the American Geophysical Union, v. 83, no. 19, p. S155, Spring Meeting, May 28-31, 2002




U.S. Department of the Interior, U.S. Geological Survey, Reston, VA, USA
Maintainer: mascholl @
Last modification: 08/20/04 11:56 (mas)

Privacy Statement || Disclaimer || Accessibility