H43A-01
Numerical and Experimental Investigations of Tsunami-Induced Sediment Transport
As demonstrated by the 2004 Indian Ocean Tsunami, high intensity wave runup and drawdown can mobilize substantial amount of sediment deposits. The resulting erosion and scour damage can undermine building foundations, roadways, sea walls, embankments, and may even lead to eventual collapse of the coastal structure. However, most previous models ignore the effect of tsunami-induced sediment transport and scour due to the complex physics and multiple scaling issues. In this work, a numerical model is presented for the simulation of solitary waves over a uniform sloped movable bed. The depth averaged nonlinear shallow water equations (SWEs) are used to model long waves, the wave profile gradient method is used to detect wave breaking, and the effect of energy-dissipation due to wave breaking is captured as a bore collapse. The effect of sediment transport is modeled via a passive scalar and is fully coupled with the SWEs to form a complete system. New forms of the erosion and deposition fluxes are introduced to account for the effect of the flow velocity, particle fall velocity, wave profile, and pore pressure gradient. The model is validated by comparing the numerical solutions to the measured bed changes obtained from the cross-shore sediment transport experiment under breaking solitary waves conducted by (Kobayashi & Lawrence 2004). To further validate the numerical model, two sets of experimental studies are planned for 2007. The first experimental study focuses on the effect of enhanced transport due to pore pressure gradients, and will be conducted at the 30-ft long flume in the hydraulics lab at the University of Hawaii using multiple grain sizes. The second experimental study focuses on tsunami-induced sediment transport, and will be conducted at the 160-ft long tsunami wave basin at Oregon State University using fine Oregon beach sand with D50=0.20mm. The experimental setups are shown, and scaling issues for the two movable bed physical models are discussed. Comparisons of the differences and similarities in transport mechanisms for tsunami cross-shore environment and river-type environment are presented.
H43A-02
Field Observation and Numerical Simulation on Sediment Yield Due to Freeze and Thaw Action
Because most parts of the Japanese islands are located within temperate to subarctic climatic zones, freeze and thaw action during winter can be observed frequently in most mountainous areas. The freeze-thaw action usually causes bedrock weathering and produces sediment. Erosion induced by freeze-thaw action appears to be a major source of sediment from mountainous areas to streams. However, these processes are little understood. In this research, a detailed field observation was conducted to elucidate mechanisms of freeze-thaw action and the sediment yield on a bare slope in southern Shiga Prefecture, Japan. The bedrock of the observed bare slope consists of granite and is highly weathered. Two small plots (plots 1 and 2) were established by partitioning these areas with plastic borders. For each plot, surface and subsurface temperatures at depths of 0, 10, 25 and 50 cm, as well as air temperature were continuously measured every 10 min during the observation period from December 2004 to April 2005. The sediment produced from weathered bedrock was collected and weighed once in a week from the plot 1. In plot 2, the sediment was not disturbed and remained on the bedrock until the end of observation period. From the observation results, it was shown that the freeze-thaw action occurred repeatedly, and the freezing front where the bedrock temperature is 0 degree C, reached a depth of 10 cm. Sediment was actively produced when freeze-thaw action was observed; when freeze-thaw action stopped in April 2005, sediment yield decreased markedly. Sediment yields from plots 1 and 2 were 108 and 44 kg/m2/year, respectively. This difference indicates that the sediment cover on the bedrock surface mitigated the effect of freeze-thaw action on sediment yield from the bedrock. These observed results were simulated by a simplified thermal conductivity analysis. Comparison between the observed and simulated results suggests that multiple episodes of freeze-thaw activity are necessary for the bedrock to be converted into sediment.
H43A-03
Suspended Sediment Load Prediction Using Artificial Neural Networks Approach
A multilayer perceptron (MLP) ANN with an error back propagation algorithm using historical daily and weekly hydroclimatological data (precipitation P(t), current discharge Q(t), antecedent discharge Q(t-1), and antecedent sediment load SL(t-1) ) is used to predict the suspended sediment load SL(t) at the selected monitoring stations. Performance of ANN was evaluated using different combinations of datasets (Input 1 = P(t), Q(t), Q(t-1), SL(t-1) , Input 2 = I-1 less P(t) and Input 3 = I-2 less Q(t-1), length of record for training (3 and 2 years) and temporal (daily and weekly) simulations. Comparison of the ANN model output with multiple linear regressions (MLR) was made. Daily simulations using Input 1 and three years of training and two years of testing (3*2) performed better (R2 and E of 0.85 and 0.72, respectively ) than the simulation with two years of training and three years of testing (2*3) (R2 and E of 0.64 and 0.46, respectively ). ANN predicted daily values using Input 1 and 3*2 architecture for Missouri (R2 = 0.97) and Mississippi (R2 = 0.96) were better than Rio Grande (R2 = 0.65). Daily predictions were better compared to weekly predictions for all three rivers. ANN predictions for Missouri and Mississippi were superior and Rio Grande was inferior compared to predictions with MLR. The modeling approach presented in this paper can be an efficient alternative to costly monitoring operations for sediment load monitoring programs where hydrological data is readily available. Key terms: ANN, sediment, sediment prediction, rivers, Mississippi, Missouri, Rio Grande
H43A-04
Hillslope-channel sediment linkages in steep forested catchments: sources to transport
Sources of eroded sediment and their transport were investigated in four headwater streams with different landslide histories in Nara, Japan, to establish relationships between sediment yield and sediment storage for different channel characteristics and to determine the connectivity of sediment sources on the hillslope with bedload transport. Two of the streams experienced debris flows in 2004; one of these retained a large amount of debris flow sediment (187.2 m3), while the other channel was flushed of most sediment (12.5 m3). Another stream impacted by a debris flow in 1989 retained 253.5 m3 of sediment. Both bedload transport and yield strongly depend on sediment storage in the catchments. The two streams in which large quantities of debris flow sediment were stored produced greater than one order of magnitude more bedload sediment than the two streams where less sediment was stored (12.5 and 79.1 m3, respectively). Bedload transport in the system affected by the older debris flow (1989) was characterized by partial mobility during most storms, while the system affected by the 2004 debris flow was characterized by equal mobility. In both catchments, grain-sizes of bedload sediment were characterized by similar or higher rates of finer materials compared with in- and near- channel deposits. Because eroded bank sediment is relatively fine, it is likely transported rapidly through the fluvial system, whereas coarser landslide sediment and sediment from uprooted trees are may move more slowly as bedload sediment once delivered to the channel. Episodic mass wasting appears to constitute very important inputs in these headwater streams and may greatly alter bedload transport. Sediment source-transport linkages must be understood to assess land use impacts in catchments.
H43A-05
Sediment from Wildfires: Production, Delivery, and Recovery at Different Spatial Scales
Sediment production from wildfires is of increasing concern in the western U.S. due to the severe effects on water quality and aquatic resources, and the increase in area burned due to increased forest densities and global warming. Over the past 6 years we have been measuring post-fire effects and sediment production from seven wildfires, three prescribed fires, and 40 unburned hillslopes in the Colorado Front Range. The resulting dataset includes a series of process-based studies, 600 plot-years of sediment production data at the hillslope scale, and changes in channel cross-sections at scales ranging from 0.01 ha to 6 km2. In unburned areas surface runoff and erosion is extremely rare. After a high-severity fire the threshold for overland flow drops to 7-10 mm hr- 1, and in the first two years after burning the median surface erosion rates are 5-10 Mg ha-1 yr-1. Hillslope runoff and sediment production rates drop to near-background levels within 3-5 years after burning, and recovery rates are primarily a function of vegetative regrowth. Approximately 80 percent of the post-fire sediment is derived from rill and channel incision in headwater areas, so nearly all of the eroded sediment is being delivered to the stream channel network. The downstream transport of this material adversely affects water quality and causes extensive aggradation. The relatively rapid recovery at the hillslope scale reduces surface runoff rates and sediment transport capacities in downstream channels. This means that the smaller streams cannot remove much of the accumulated sediment, and post-fire channel recovery may take hundreds of years as compared to the 3-5 years needed for hydrologic recovery at the hillslope scale.
H43A-06
Multiscale Analysis of Sediment Flux Data and Interpretation in Terms of River Planform Morphodynamics
Knowledge of the quantity and variability of sediment flux in a river is important for several hydrologic, hydraulic and ecological applications. While many studies have focused on the prediction of mean sediment flux, less work has been done on explaining the multi-scale sediment flux variability and its relation to the multi-scale variability of surface bed elevations. In the present study an attempt is made to quantify the variability of sediment fluxes at different temporal scales (accumulation intervals) and relate it to the variability of the evolving bed topography. Data were collected in a laboratory flume for different flows and for gravel and sand beds, at resolution of 1 sec (sediment flux), 10 secs (sonar data of bed elevation evolution) and 1 mm (bed topography laser transects). Our analysis shows that sediment flux series exhibit a rich multiscale structure (multiscaling) which is distinctly different in gravel versus sand bed rivers and directly reflects the multiscale variability of the river bed elevations. These results are explored towards: (a) predicting the variability (and extremes) of sediment flux rates at scales different than those at which measurements are available, and (b) predicting the statistics of sediment flux rates from the statistics of the more readily available bed topography data.
H43A-07
Epic Erosion Along Newly Constructed Roads in Yunnan, China
The recent expansion and construction of new mountain roads in northwestern Yunnan Province, China, poses problems related to landslides and surface erosion that are impacting the headwaters of three great river systems: the Salween, Mekong, and Yangtze. Many of these newer roads are simply blasted into unstable hillsides with virtually no attention paid to optimal road location, construction practices, and erosion control measures. During summer 2006, seven people traveling in a minivan along a newly constructed road to Weixi were killed by a landslide. A survey conducted along a this 23.5 km road section (4 yr old) in the headwaters of the Mekong River revealed epic levels of landslides and surface erosion. Based on a preliminary survey, the road erosion was categorized as moderately severe, severe, or very severe, and a representative 0.75 to 0.90 km stretch of road was then surveyed for both landslide (based on dimensional analysis) and surface erosion (based on soil pedestal height). Average mass wasting rates (9608 t ha-1yr-1) along the road were more than 13 times higher than surface erosion (720 t ha-1yr-1), even though surface erosion rates are among the highest reported for disturbed lands. Dry ravel constituted a minor proportion of the mass wasting: 4% in the severe erosion section of the road and 0.5-0.6% in the moderately severe and very severe sections. For the very severe erosion road section (6 km long), estimated landslide erosion alone was > 33,000 t ha- 1yr-1, 620 times the average landslide erosion from forest roads built in unstable terrain in western North America. These levels of landslide erosion along the Weixi road are the highest ever documented and are somewhat representative of erosion along new mountain roads in this region of Yunnan. Sediment produced from roads is highly connected to fluvial systems; we estimate that 80-95% of the direct sediment contributions into the headwaters of these rivers are attributable to road erosion and landslides. These epic sediment loads represent cumulative effects that may persist in these important transnational rivers for decades.
H43A-08
Rainfall Erosion of Intertidal Wetlands
A poorly quantified and mechanistically overlooked material cycling process in estuarine landscapes is rainfall- driven erosion of intertidal salt marsh and mudflat surfaces. During low tide rainsplash erosion and runoff may entrain carbon-rich sediment that, due to high cohesion, ordinarily is not mobilized by tidal currents and shallow water waves (e.g., a deachment limited landscape). Consequently, low tide rainfall may affect tidal creek network structure, creek network extension and nutrient cycling. Field manipulations and passive observations show that low tide rainfall events preferentially entrain highly nutritious intramarsh particulate matter, and in some cases with high concentrations of adsorbed metals. Once mobilized, the subtle topographic variations of the salt marsh landscape route runoff and the suspended load to intertidal creeks, and the subtidal water column. Hence low tide rainfall-runoff processes may enhance the cycling of, for example, benthic microalgae and their products, a primary carbon source for estuarine food webs. Once in the subtidal zone the material may be exported to the coastal ocean or it may be redeposited on the marsh surface with the next high tide, depending on tidal phase. Taken together these observations reveal one facet of salt marsh interactions between landscape structure- biological processes-physical processes.