Series
Georgia Water Resources Conference

Permanent Link
https://hdl.handle.net/1853/70886
Series Type
Event Series
Description
Associated Organization(s)
Associated Organization(s)
Organizational Unit

Filters

2011 - 2013 4
4
4
4
Reset filters

Settings
Author: Dowd, John F. × End date: 2013 × Start date: 2010 × Item Type: Publication ×

Publication Search Results

Now showing 1 - 4 of 4
Thumbnail Image
Item

Comparison of Digital Filter Hydrograph Separation with Geochemical Separation

2013-04 , Mason-Deese, William , Dowd, John F. , Cary, Richard H.

Aquatic Stormflow is defined as flow resulting directly from a storm event, while baseflow is thought to be groundwater flow that continuously occurs, most predominantly during non-storm periods. While conceptually these concepts are convenient, it is difficult to ascertain the actual flow paths of each component. In this paper we will compare common digital filters used to estimate baseflow with a geochemically derived baseflow separation. A Dynamic End Member Mixing Analysis (DEMMA) on Panola Mountain, Georgia was used by Cary (2011) to separate four stream flow components using naturally occurring chemical tracers for 22 storm events. DEMMA relies on intensive runoff and chemical sampling, and uses the flow and chemistry hysteresis to separate the hydrograph. Several digital filters were compared to the DEMMA hydrographs. While parameterized differently, each was a recursive procedure that acts as a low pass filter. In general the digital filters over estimate true baseflow for Panola (that is, true groundwater flow), and more closely resemble contributions from subsurface flow (that is soil) pathways. The one parameter filters are insensitive to calibration, although simple to use because the parameter is usually not modified. The two parameter filter (Eckhardt, 2005) was more robust in its range, but sensitive to calibration. This research provides some insight into the flow paths the digital filters may be approximating.

View
Thumbnail Image
Item

Beyond Correlation: the Search for Causal Relationships Between Flow Percentiles and Watershed Variables

2011-04 , Ssegane, Herbert , Tollner, E. W. , Mohamoud, Yusuf , Rasmussen, Todd C. , Dowd, John F.

The study explored use of causal feature selection algorithms to select dominant watershed variables that drive high, medium, and low flows. A two step approach was implemented. The first step minimized variable redundancy by examining variable relevance, variable redundancy, and conditional relevance of variable pairs whose correlation was greater than 0.9. The second step used six algorithms that seek to reconstruct a Bayesian network structure around a target variable for each flow percentile. Nineteen (19) flow percentiles were used to characterize high, medium, and low flow conditions of 26 Piedmont watersheds in the Mid-Atlantic. The algorithms included: (1) Grow-Shrink (GS); (2) interleaved-Incremental Association Markov Boundary (interIAMB) (3) Incremental Association Markov Boundary with Peter-Clark (IAMBnPC); (4) Local Causal Discovery (LCD2); (5) HITON-PC; and (6) HITON-MB. A new method was developed to quantify the reliability of each algorithm and its performance was compared to existing reliability methods. The effect of the initial number of variables on the final variable set selected by each algorithm was tested. Fusion of the algorithms was used to determine the overall dominant features for each flow percentile.

View
Thumbnail Image
Item

Site Investigation of Southern Historic Cattle Dip Vats

2011-04 , McKinnon, Robert J. , Fischer, Teresa A. , Lodato, Michael N. , Dowd, John F.

From the early 1900s to 1961, cattle ranchers in Southern states, particularly Arkansas, Georgia, Florida, and Texas, were required to control ticks on cattle to prevent Texas Cattle Fever, a disease that impacted cattle in the South. A United States Department of Agriculture program provided for the construction and operation of over 3,200 cattle dip vats. Solutions containing arsenic were initially used, until the ticks developed a resistance to the arsenic solution. Around the mid-1940s, arsenical pesticides were replaced by organochlorinated pesticides. The historic operations of cattle dip vats have lingering impacts on groundwater and soil that pose a threat to human health and the environment. Geosyntec Consultants has contracted with the Florida Department of Environmental Protection to evaluate groundwater and soil impacts associated with historic cattle dip vats that are located in the Florida State Park System. A multi-faceted approach was used to characterize contaminant migration in groundwater and soils with respect to historical site operations, archaeological preservation, and environmental health. Several case studies at state parks located across Florida are presented, including methods of delineation and remediation. The approaches employed in these studies will assist others in future investigations of historic cattle dip vats that exist across the South, including Georgia.

View
Thumbnail Image
Item

Determining Watershed Flow Pathways Using Geochemistry and Timing

2011-04 , Cary, R. H. , Dowd, John F. , Peters, N. E.

Investigating storm runoff generation in watersheds is an area of ongoing hydrologic research. Geochemical tracer studies, such as static end-member mixing analysis (EMMA) and hysteresis loop analysis, have been used to evaluate these processes. While EMMA can assess the relative input of flow pathways for individual stream water quality samples collected during a storm, it cannot quantify their contributions continuously. Hysteresis loops of stream discharge versus geochemical tracer concentration can be used to estimate relative inputs of basic end-member pathways, but this approach only suggests the timing and dominance of flow pathways and these patterns alone cannot quantify their contributions. We propose a new method that incorporates both hysteresis loops and geochemical tracer studies to quantify runoff contributions from watershed flow pathways during a storm. The approach involves estimating relative tracer concentrations of four end-members, along with estimating the percentage of total stream discharge from each end-member. The method has been applied to a 22 year dataset from Panola Mountain Research Watershed, Georgia and has identified two distinct watershed responses to rain events. The responses appear be related to a threshold of 50-60mm of total rain.

View