HR: 09:30h
AN: H51A-07 [PDF]
TI: Using Distributed-Hydrology-Soil-Vegetation Model to Study Road Effects on Stream flow and Soil
Moisture
AU: * Cuo, L
EM: cuo@hawaii.edu
AF: Department of Geography, University of Hawaii at Manoa, 2424 Maile Way, Honolulu, HI 96822 United States
AU: Giambelluca, T W
EM: thomas@hawaii.edu
AF: Department of Geography, University of Hawaii at Manoa, 2424 Maile Way, Honolulu, HI 96822 United States
AU: Ziegler, A D
EM: thaihawk@hotmail.com
AF: Department of Geography, University of Hawaii at Manoa, 2424 Maile Way, Honolulu, HI 96822 United States
AU: Ziegler, A D
EM: thaihawk@hotmail.com
AF: Department of Geography, National University of Singapore, 1 Arts Link, Kent Ridge, Singapore,
117570SGP
Singapore
AU: Nullet, M A
EM: mnullet@hawaii.edu
AF: Department of Geography, University of Hawaii at Manoa, 2424 Maile Way, Honolulu, HI 96822 United States
AB:
The distributed-hydrology-soil-vegetation model (DHSVM) was applied in Pang Khum Experimental Watershed (PKEW), located near
19.05\deg N, 98.65\deg E in the mountainous region of northern Thailand, headwaters of the Chao Phraya River system. PKEW
has a highly seasonal rainfall regime, with 90% of the annual 1200-1400 mm rainfall occurring during the southwest summer
monsoon. The elevation of PKEW ranges from approximately 1100 to 1500 m. Total road area including road banks is about
1.2% of the basin area. About 57% of the road area occurs on slopes steeper than 10%. All roads are unpaved. Land cover
in PKEW is affected by swidden agriculture. Six land cover and nine soil classes are identified in the basin. We have been
working in the area since 1997 as part of the Thailand Roads Project (TRP). Within the basin, we are monitoring microclimate
at two sites, soil moisture at four sites, and rainfall at five sites. Streamflow is measured at the outlet.
Based on digital elevation data, DHSVM explicitly accounts for the spatial distribution of the stream and road networks, soil
depth, soil and vegetation types. The model run period, including warm up, calibration and validation, is from August 1997
to January 2001. Field measurements provide forcing data, calibration data, and guidance in parameter selection. Model
calibration and validation were done by aggregating simulated hourly soil moisture and stream flow into daily values and
comparing them with aggregated daily measurements.
For the calibration period, RMSEs of soil moisture and streamflow were lower than the observed variability as represented by
the standard deviation, median absolute deviation, and (for stream flow) interquartile range. Model performance drops in
validation period, but RMSEs remain near or lower than observed variability.
We ran DHSVM with and without roads to examine their effects. Significant effects of roads were found despite the very low
proportion of the watershed covered by roads and road banks. Streamflow for road and non-road cases was significantly
different (p $<$ 0.0001) based on the Wilcoxon signed rank test. In general, roads increase peak volume for short, intense
storms, but reduced whole-period discharge by about 5.5%. Soil moisture was affected in cells where roads occur. In cells
where water exited the road onto the hillslope, soil moisture was higher than it was without the road. In cells with roads,
but without water flowing onto the hillslope, soil moisture was higher in some cases and lower in others.
UR: http://webdata.soc.hawaii.edu/climate/Roads1/Roads.html
DE: 1803 Anthropogenic effects
DE: 1836 Hydrologic budget (1655)
DE: 1860 Runoff and streamflow
DE: 1866 Soil moisture
SC: Hydrology [H]
MN: 2003 Fall Meeting