97th ESA Annual Meeting (August 5 -- 10, 2012)

COS 161-4 - Modeling the effects of climate change and invasive species cover on stream flow in tropical island watersheds using the distributed hydrology-soil-vegetation model

Thursday, August 9, 2012: 2:30 PM
E145, Oregon Convention Center
Ayron M. Strauch1, Christian P. Giardina2, Richard MacKenzie2 and Greg Bruland3, (1)Natural Resources and Environmental Management, University of Hawaii at Manoa, Honolulu, HI, (2)Institute of Pacific Islands Forestry, USDA Forest Service, Hilo, HI, (3)Biology and Environmental Studies, Principia College, Elsah, IL
Background/Question/Methods

In tropical island systems, freshwater is a vital resource for agriculture, development, consumption, and native biota. While freshwater resources are threatened by climate change and invasive species, few studies have examined how these stressors or their potential interactions affect stream flow in tropical watersheds. We used the Distributed Hydrology-Soil-Vegetation Model (DHSVM) to assess the effect of climate warming, altered precipitation and expanding invasive species cover on stream flow in 87 watersheds located along the Hilo-Hamakua Coast on Hawaii Island. Watersheds were located along a highly constrained precipitation gradient spanning 3,500 mm in mean annual precipitation (MAP), while all other variables (landcover, geomorphology, topography) were held constant, thus providing a space-for-time substitution examining the impacts of changing MAP. Using the DHSVM calibrated for a six-year period (water years 2006 to 2011), we assessed how 1) mean annual temperature change; 2) MAP change; and 3) change in Psidium cattleianum (strawberry guava) cover will influence daily, monthly, and annual water flow as well as flow variability.

Results/Conclusions

Changes in MAP were the principle driver of stream flow while increased spread of P. cattleianum cover also reduced downstream flow. We also found a negative relationship between modeled flow variability (Q10:Q90) and MAP, and a positive relationship between modeled flow variability and watershed size across the precipitation gradient. Furthermore, watersheds with higher MAP experienced a greater loss of mean monthly stream flow compared to similarly-sized watersheds with lower MAP, whereas the percent of total flow lost was greater in lower MAP streams. Watersheds with lower MAP had much more variable stream flow compared to similarly-sized watersheds with higher MAP. We conclude that predicted climate and vegetation changes in tropical environments will decrease stream flow and increase flow variability, with anticipated impacts on human and biological systems, especially downstream water users and in-stream freshwater biota. Restoration of degraded and/or invaded forest will reduce the impact of climate change variables on stream flow.