Richard F. Keim

Time scale and intensity dependency in multiplicative cascades for temporal rainfall disaggregation

[1] Multiplicative random cascades (MRCs) can parsimoniously generate highly intermittent patterns similar to those in rainfall. The elemental MRC model parameter is the cascade weight, which determines how rainfall at one scale is partitioned at the next smallest scale in the cascade. While it is known that the probability density of these weights may vary with both time scale and rainfall intensity, nearly all previous studies have considered either time scale or intensity separately. We examined the simultaneous dependency of the weights on both factors and assessed the impacts of explicitly including these dependencies in the MRC model. On the basis of the observed relationships between cascade weights and time scale and intensity, four progressively more “dependent” models were constructed to disaggregate a long time series of daily rainfall to hourly intervals. We found that inclusion of the intensity dependency on the model parameters that generate dry intervals greatly improved performance. For the relatively small range of time scales over which the rainfall was disaggregated, varying model parameters with time scale resulted in minor improvement.

Throughfall and Stemflow in Wooded Ecosystems

Incident precipitation is routed to the subcanopy by throughfall and stemflow. Throughfall is defined as the precipitation that passes directly through a canopy or is initially intercepted by aboveground vegetative surfaces and subsequently drips from the canopy, whereas stemflow is the precipitation that drains from outlying leaves and branches and is channeled to the bole (or stem) of plants. Throughfall and stemflow inputs constitute the majority of incident precipitation (Table 21.1) and are of critical importance to wooded ecosystems, ranging from 70 to 90% of incoming precipitation in most cases with the remainder lost to interception (Levia and Frost 2003). The inputs of throughfall and stemflow are highly variable over space and through time with consequent “hot spots” and “hot moments” of water and solute inputs from the canopy to the subcanopy (Stout and McMahon 1961; Levia 2003; Zimmermann et al. 2007). There are marked differences in the routing of intercepted water to the forest floor via throughfall (Keim and Skaugset 2004) and stemflow (Herwitz 1987) in terms of flowpaths and residence times along vegetative surfaces, which result in notable differences in solute concentrations and mass fluxes of canopy leachates (Levia and Frost 2003; Zimmermann et al. 2007). Recent work has documented the demonstrable effects of throughfall and stemflow to the hydrology and biogeochemistry of hillslopes (Keim et al. 2006a; Liang et al. 2009). Liang et al. (2009), for example, have reported that stemflow has led to a root-induced bypass flow infiltration process on hillslopes, a coupled mechanism termed “double-funneling” (Johnson and Lehmann 2006) discussed further in Chap. 24. Stemflow also has been documented to contribute to the enrichment of soils beneath shrubs in semiarid climates, leading to a “fertile island” effect (Whitford et al. 1997), whereas the spatial distribution of fine roots was observed to mirror throughfall inputs (Ford and Deans 1978).

Digital terrain modeling of small stream channels with a total-station theodolite

Abstract A Digital Terrain Model (DTM) is an alternative to traditional measures of stream channel morphology that allows for extraction of many different types of data. This paper describes a method of creating high-resolution Digital Terrain Models of stream channels using an electronic, digital, total-station theodolite and standard methods of land surveying, and also includes considerations unique to hydrological application. Included is a detailed description of one application in Oregon, and also suggestions of how to apply the method in other research of morphology.

Physical Aquatic Habitat II. Pools and Cover Affected by Large Woody Debris in Three Western Oregon Streams

Abstract Large woody debris (LWD) is important in affecting stream channel morphology and aquatic habitat. Although the greatest effects on streams of the Pacific Northwest have been by LWD from large conifers, many riparian forests in the region are dominated by red alder Alnus rubra. The effects of the small size and short life of LWD from red alders on channel morphology may be different from that of conifers and are poorly understood. We added LWD (primarily red alder) to three third-order streams in the Oregon Coast Range and used digital terrain models to evaluate physical habitat for salmonids over 3 years. Total residual pool volume increased in two streams, but in the one with the lowest gradient it did not change in the treated portion and even decreased in the untreated portion. In all streams, both the relative proportion and absolute amount of residual pool volume from deep pools increased from their pretreatment values. Cover from LWD in pools increased after treatment and remained high, but...

Floodplain biogeochemical processing of floodwaters in the Atchafalaya River Basin during the Mississippi River flood of 2011

The 2011 flood in the Lower Mississippi resulted in the second highest recorded river flow diverted into the Atchafalaya River Basin (ARB). The higher water levels during the flood peak resulted in high hydrologic connectivity between the Atchafalaya River and floodplain, with up to 50% of the Atchafalaya River water moving off channel. Water quality samples were collected throughout the ARB over the course of the flood event. Significant nitrate (NO3−) reduction (75%) occurred within the floodplain, resulting in a total NO3− reduction of 16.6% over the flood. The floodplain was a small but measurable source of dissolved reactive phosphorus and ammonium (NH4+). Collectively, these results from this large flood event suggest that enhancing river-floodplain connectivity through freshwater diversions will reduce NO3− loads to the Gulf of Mexico during large annual floods.

Applying geostatistics to quantify distributions of large woody debris in streams

Abstract Large woody debris (LWD) strongly influences morphology and aquatic habitat in streams within forested watersheds. Previous studies have used data to describe spatial distributions of LWD in qualitative terms. We used high-resolution spatial data collected from a forested stream in western Oregon to conduct three geostatistical analyses. The data were collected at five time periods prior to and following the addition of LWD into the stream. Each analysis is examined for its potential and shortcomings in quantifying distributions. Our findings indicate that semi-variograms, when used with a raster spatial data structure, can be useful quantitative descriptions of LWD distributions.

The role of stable isotopes in understanding rainfall interception processes: a review

The isotopic composition of water transmitted by the canopy as throughfall or stemflow reflects a suite of processes modifying rainfall. Factors that affect isotopic composition of canopy water include fractionation, exchange between liquid and vapor, and selective transmittance of temporally varying rainfall along varying canopy flowpaths. Despite frequent attribution of canopy effects on isotopic composition of throughfall to evaporative fractionation, data suggest exchange and selection are more likely the dominant factors. Temporal variability in canopy effects is generally consistent with either exchange or selection, but spatial variability is generally more consistent with selection. However, most investigations to date have not collected data sufficient to unambiguously identify controlling processes. Using isotopic data for improved understanding of physical processes and water routing in the canopy requires recognizing how these factors and processes lead to patterns of isotopic variability, and then applying this understanding towards focused data collection and analysis.

Physical Aquatic Habitat I. Errors Associated with Measurement and Estimation of Residual Pool Volumes

Abstract Residual pools have been used as a measure of physical aquatic habitat, but measurement techniques usually entail estimates or unquantified errors. In order for residual pools to be a useful metric of stream channels, it is important for the errors associated with their measurement to be quantified. We used precise, digital terrain models of a third-order mountain stream in Oregon to identify the sensitivity of measurements to error and the consequences of those errors in calculating the volumes of residual pools. Precise quantification of the elevations of the crests of riffles is critical for precise measurement of the volume of individual residual pools because they serve as control points for the water surface elevations of residual pools. Reach-level estimates of the total volume of residual pools, however, are relatively insensitive to errors incurred in measuring individual pools. This insensitivity indicates that reach-level measurements of residual pool volume may be more appropriate tha...

Perirheic mixing and biogeochemical processing in flow‐through and backwater floodplain wetlands

Inundation hydrology and associated processes control biogeochemical processing in floodplains. To better understand how hydrologic connectivity, residence time, and intrafloodplain mixing vary in floodplain wetlands, we examined how water quality of two contrasting areas in the floodplain of the Atchafalaya River—a flow-through and a backwater wetland—responded to an annual flood pulse. Large, synoptic sampling campaigns occurred in both wetlands during the rising limb, peak, and falling limb of the hydrograph. Using a combination of conservative and reactive tracers, we inferred three dominant processes that occurred over the course of the flood pulse: flushing (rising limb), advective transport (peak), and organic matter accumulation (falling limb). Biogeochemistry of the two wetlands was similar during the peak while the river overflowed into both. However, during the rising and falling limbs, flow in the backwater wetland experienced much greater residence time. This led to the accumulation of dissolved organic matter and dissolved phosphorus. There were also elevated ratios of dissolved organic carbon to nitrate in the backwater wetland, suggesting nitrogen removal was limited by nitrate transported into the floodplain there. Collectively, our results suggest inclusion of a temporal component into the perirheic concept more fully describes inundation hydrology and biogeochemistry in large river floodplain.

Conservation and Use of Coastal Wetland Forests in Louisiana

U.S. Geological Survey, National Wetlands Research Center, 700 Cajundome Blvd., Lafayette, LA 70506 School of Renewable Natural Resources, Louisiana State University Ag Center, Renewable Natural Resources Building, Baton Rouge, LA 70803 Baruch Institute of Coastal Ecology and Forest Science, Clemson University, Department of Forestry and Natural Resources, PO Box 596, Georgetown, SC 29442 Department of Oceanography and Coastal Sciences and Coastal Ecology Institute, School of the Coast & Environment, Louisiana State University, Baton Rouge, LA 70803 Center for Bottomland Hardwoods Research, USDA-Forest Service Southern Hardwoods Laboratory, PO Box 227, Stoneville, MS 38776 USGS Louisiana Cooperative Fish and Wildlife Research Unit, Louisiana State University Ag Center, School of Renewable Natural Resources, Baton Rouge, LA 70803 Savannah River Ecology Laboratory, PO Drawer E, Aiken, SC 29802 Department of Biological Sciences, Southeastern Louisiana University, Box 10736, Hammond, LA 70402