H52B-01
Interactions Between Overland and Subsurface Runoff on Steep Forest Slopes: is Hortonian Overland Flow Important?
The prevailing notion that Hortonian overland flow is insignificant on temperate forest hillslopes has been challenged in poorly managed Japanese cypress (hinoki) plantations, due to overstocked stands and poor ground cover. Small (0.5 x 2 m) and hillslope-scale (8 x 25.5 m) runoff plots were established in forested sites of thinned and unthinned hinoki stands and mixed deciduous forests in Mie, central Japan. Additionally, small runoff plots, TDR probes, and recording tensiometers were installed in a forest site near Tokyo to assess the interactions of surface and subsurface flow paths. Runoff coefficients during larger storms (> 50 mm total rainfall) at Mie typically exceeded 0.10 for small plots, whereas hillslope scale plots rarely had runoff coefficients > 0.10. The contrast between plot sizes is more pronounced and consistent for the thinned hinoki forest with good ground cover and the deciduous forest site. In both of these sites, runoff coefficients from small plots were > 0.1 for total rainfall > 30 mm. In contrast, runoff coefficients from large plots for almost all storms were < 0.1, and larger storms had much lower runoff coefficients compared to smaller storms. These data suggest that much of the Hortonian overland flow re-infiltrates into soil organic horizon or matrix where it moves downslope as subsurface flow. Questions still remain whether overland flow measured in plots is really Hortonian. Field observations during major storms at similar sites indicate that runoff plots collect some shallow subsurface and preferential flow in the organic-rich biomat. Detailed soil water measurements in a deciduous forest near Tokyo revealed that 47% of rainfall (168 mm) in a large storm that infiltrated into the soil moved laterally as biomat flow; the remainder of the infiltrated rainwater percolated deeper into the mineral soil. Of the 9.3% of the rainfall that was measured as runoff, a portion appears to be biomat flow. Biomat flow represents a significant, albeit transient flow path that buffers against Hortonian overland flow even for thin litter layers. Water flowing through such near-surface biomats will be more rapid than subsurface runoff in the mineral soil, but much slower than true Hortonian overland flow and will have ample opportunity for vertical infiltration.
H52B-02
Low-dimensional modeling of hillslope subsurface flow: the relationship between rainfall, recharge, and unsaturated storage dynamics
We present a coupling between the one-dimensional Richards equation for vertical unsaturated flow and the one- dimensional hillslope-storage Boussinesq equation (hsB, see [Troch et al., 2003]) for lateral saturated flow along complex hillslopes. Here, the capillary fringe is included in the flow domain as an integral part of the Boussinesq aquifer. The coupling allows quantatitive investigation of the role of unsaturated storage in the relationship between rainfall and recharge. The coupled model (hsB-coupled) is compared to the original hsB model (hsB- original) and a three-dimensional Richards equation (RE) based model (taken to be the benchmark) on a set of seven synthetic hillslopes, ranging from convergent to divergent. Using hsB-original the watertables are overestimated and the outflow rates are generally underestimated, and there is no delay between rainfall and recharge. The coupled model, however, shows a remarkably good match with the RE model in terms of outflow rates, and the delay between rainfall and recharge is captured well. We also see a clear improvement in the match to the water tables, even though the values are still overestimated for some hillslope shapes, in particular the convergent slopes. We show that, for the hillslope configurations and scenarios examined in this paper, it is possible to reproduce hydrographs and water table dynamics with a good degree of accuracy using a low- dimensional hydrological model.
H52B-03
Effects of Gravel Bars on Nutrient Spiraling in Bedrock-Alluvium Streams
The importance of the connection between nutrient transport and local stream geomorphology is becoming increasingly important. Studies have shown that the interconnectivity of nutrient cycles in the downstream direction is in part controlled by the distribution and size of gravel bars in low order streams, as hyporheic flow occurs dominantly through alternate and mid-channel gravel bars. For this investigation multiple gravel bars in a 3rd order bedrock-alluvium stream were studied to determine general relationships between nutrient spiraling and hyporheic flow. The first goal was to understand (1) the extent to which water moves through hyporheic zones and (2) the basic chemistry of the hyporheic water. The second part of the study was to understand how nutrients, notably nitrogen, are affected in their cycling by the relatively long residence times encountered in gravel bars during hyporheic flow. Wells were installed along a 600 m reach of Panther Creek, KY in selected bars, as well as in a secondary location involving a grid installation pattern in one large bar. Results have shown that hyporheic flow through gravel bars is an important factor in influencing stream chemistry. Background water chemistry surveys have shown that certain parameters, specifically ammonium and nitrogen concentrations vary downstream, and that the dominant control over these changes is gravel bar location. Rhodamine WT was used in field tracer tests to track the travel times of water through bars as well as partitioning of water between the open channel and hyporheic flows. Further tests will be conducted utilizing a stable isotope study to determine how nitrogen is affected by hyporheic flow, and what implications this has for nutrient transport. We expect results to show that the spacing and size of gravel bars is a dominant control in key nutrient spiraling parameters, namely uptake lengths and overall nitrogen cycling rates. This has implications for how natural systems will respond to human impacts, both through the modification of the physical template of stream systems as well as increased anthropogenic loading of nitrogen.
H52B-04
Wind-Dispersion of Biomass in Patterned Vegetation Models: From Classical Diffusion to WALD
Vegetation spatial patterning observed in semi-arid and arid ecosystems worldwide provides an exciting system for the testing and validation of ecohydrological models that couple the evolution of the vegetation pattern to the hydrological regime. However, one fundamental limitation to these models is their represention of biomass dispersion, currently assumed to follow a classical diffusion processes with an empirical diffusion coefficient. We present a new approach for biomass movement based upon the WALD model, a mechanistic dispersal kernel for wind-dispersed seeds. The model can be parameterised based upon seed characteristics, release height, and the mean wind frequency distribution. The WALD model is shown to reproduce the velocities of biomass movement reconstructed from the pollen record to within an order of magnitude or better for a wide range of species considered in North America and Europe. Incorporating WALD in pattern-forming ecohydrological models is briefly discussed.
H52B-05
Hydrodynamic Dispersion in Turbulent Open-Channel Flow Over an Irregular Bed
Characterizing hydrodynamic dispersion in open-channel flow is a key element in environmental studies aimed at modeling the transport and cycling of nutrients and pollutants. We use a simple flow model together with a particle-tracking algorithm to explore first-order influences of bed topography on the hydrodynamic dispersion. The model is based on linearized versions of the shallow-water equations for flow over an irregular bed topography composed of alternate bars. Theoretical dispersion curves were generated by simultaneously releasing tracer particles across the channel at a fixed location and keeping track of their positions for various intervals of time and different channel geometries. Particles were subject to fluctuating motions mimicking effects of turbulence. The shape and length of the tail of the dispersion curve appears to depend primarily on the time elapsed since the particles were released. For short time intervals, the curve is characterized by a steep leading edge which later transforms into a peak with a less steeply sloping front. This transition occurs more rapidly with increasing bar amplitude, and also with increasing number of alternate bars in the section traveled - thus with shorter bar wavelengths. Rhodamine WT was used in a field dye test conducted on a 150 m straight reach of Panther Creek, KY. This section of the creek has an average channel width of 6.3m, and exhibits a loose alternate bar structure with wavelength of ~55 m and amplitude of ~0.1 m. The bed of the channel has an average slope of 0.01 and consists of coarse gravel with a D85 of 6 cm. Consistent with the modeling results, the tracer test revealed a relative steep leading front and slowing decaying tail. In both the simulated and field case, this tail is similar to the behavior predicted by "dead zone" models of dispersion, and is attributable mostly to spatial variations in the local flow (with superimposed fluctuating motions) associated with vertical velocity structure combined with shoaling and deepening over the bed topography.
H52B-06
Headwater Hydrology in Costa Rica
Tropical hydrology, especially headwater hydrology, has not received much attention in the scientific literature. Recently there have been reports on soil moisture and saturated conductivity but there are still few reports on streamflow. Scientists and managers often rely on information gathered in temperate climates that may or may not reflect tropical hydrology or biology. In temperate climates, especially in North America, knowledge about stream hydrology and biology of unmanged systems (reference condtions) is used to guide stream restoration efforts. Stream degradation is also a problem in the tropics, but there are few long-term studies of unmanaged streams that can be referred to when trying to understand how these streams historically functioned. We present a preliminary analysis of headwater streams in La Cangreja National Park, Costa Rica. Baseline data collection of streamflow, precipitation and air and water temperature was begun in January 2003. Permanent stream habitat plots as well as adjacent permanent riparian forest habitat plots were also installed. La Cangreja National Park lies in the tropical wet forest and premontane wet forest Holdridge Life Zones. There is a distinct rainy season from May to November. About 4000 mm of rain falls each year. Streamflow responds very quickly to rainfall during the rainy season. Stream levels have been observed to rise up to 0.3 m in 15 minutes during intense rainstorms and fall back to lower levels almost as quickly after the rainfall ceases. This flashy' response goes against common wisdom of forests acting as sponges and having slow responses to precipitation events.
H52B-07
Effect of vegetation change on hydrologic cycle in Changjiang and Yellow River catchments
In China, the climate and water budgets are very different between the northern and southern regions. In the semi-arid regions, such as Liaohe, Haihe, Ruanhe, and Yellow River catchments in north China, the runoffs are small or even zero during dry season (May - October) and are very sensitive to temperature increase and rainfall decrease. In the humid south regions like Changjiang River catchment, the runoffs are perennial and the base flow normally occupies a large portion of the total runoff volume, which often causes severe floods in the middle and lower regions. China is now to accelerate preparatory work on its ambitious project for driving water from the Changjiang River to the Yellow River catchments, so-called, the South-to-North Water Transfer Project. The chief objective of this project is to remedy the basic mismatch of Chinafs land and water resources. The new researches has begun about the optimum amount of water that can be transferred, the environmental consequences of the scheme for the Changjiang River, the overall economic and social consequences of the project and various individual technologies, including the development of large pumps. We applied the NIES Integrated Catchment-based Eco-hydrology (NICE) model (Nakayama and Watanabe, 2004, 2006; Nakayama et al., 2006, 2007) to the Changjiang and Yellow River catchments in order to simulate the relationship between the vegetation change and hydrologic cycle. We simulated the water/heat dynamics in the entire catchment with a resolution of 10 km mesh by using the NICE. The model reproduced excellently the river discharge, soil moisture, evapotranspiration, groundwater level, crop water use, crop productivity, et al. Furthermore, we evaluated the effect of irrigation in main crops such as rice, wheat, and maize on the water/heat budgets by comparing between both catchments. References Nakayama, T., and Watanabe, M., Water Resour. Res., 40(8), W08402, doi: 10.1029/2004WR003174, 2004. Nakayama, T., and Watanabe, M., Hydrol. Earth Syst. Sc. Discuss., 3, 2101-2144, 2006. Nakayama, T., et al., Hydrol. Process., 20(16), 3441-3466, doi: 10.1002/hyp.6142, 2006. Nakayama, T., et al., Sci. Total Environ., 373(1), 270-288, doi: 10.1016/j.scitotenv.2006.11.033, 2007.
H52B-08
Velocities of Water Pulses are the Key to Up-Scaling Subsurface Storm Flow
The velocity of laminar flow in soils is dynamically related to the thickness of water rivulets while the lateral extent of all the rivulets is proportional to the mobile component of volumetric soil moisture. The geometry of the pores thus constrains flow as it is expressed with the hydraulic conductivity at saturation. Flowing rivulets can be superimposed under the two conditions of (i) gravity being the only flow-driving force (i.e. hydraulic pressure and capillary potential become negligible at the hill-slope scale) and (ii) gravity being immediately balanced by viscosity. The rivulet approach is positioned between the realm of Darcy's law, where all the pores are filled with water, and the realm of the Richards' equation, where the diffusion of capillary potential increasingly dominates flow as soil moisture decreases. The rivulet approach to flow provides for a framework for dealing with fast flow that is usually associated with preferential flow. Measured vertical and lateral flow velocities will be presented. The resulting spectrum of time and length scales of subsurface storm flow will be discussed.