Visualize Complexity, Discover Solutions, Shatter Limits
|
|
Visualize Complexity, Discover Solutions, Shatter Limits |
|
Case Study: Cooper River-Charleston, SC |
|
|
Client: |
Cooper River User's Group |
| Project: |
Non-Point Source Pollution Impact
|
|
Issues:
|
The Cooper River estuary flows through Charleston, SC, and is bounded by the harbor in the south and a hydroelectric dam on Lake Moultrie, some 75 kilometers north. Discharges into the river by several industrial and municipal wastewater treatment plants are controlled and monitored by the SC Department of Health and Environmental Control (SCDHEC). These discharges constitute “point sources" of loading (available oxygen is consumed, or “loaded” by bacteria and other pollutants). Dissolved oxygen concentration (DO), a measure of the available oxygen, is also impacted by the "non-point sources" of rainfall run-off and tidal flushing of interior wetlands. |
| ADMi's Role: |
A study conducted by researchers from the U.S. Geological
Survey (USGS, Columbia, SC) and Advanced Data Mining, LLC (ADMi) compared
two approaches to modeling the Cooper River’s water quality. They were
state-of-the-art finite-difference modeling and artificial neural
 networks
(a machine learning method). The study concluded that the data mining models
were much more accurate, required 80% less time to develop, and offered a
number of user-friendly and real-time deployment options that are
unmatchable by finite-difference models. Subsequent research identified the
ability of the methods developed to obtain optimized, real-time control of
the river’s salt-front via dam releases to protect an inland freshwater
reservoir. |
|
Results:
|
In the latest study, the Charleston Commissioners of Public Works of Charleston, SC, contributed to the development of special signal processing technology, called signal decomposition, which isolated the causes of variability in the river’s DO signals. This allows the relative impacts of tides, ambient conditions, and point and non-point sources to be calculated with precision. A key finding was that typical non-point source impacts could be many times greater than that of the point sources. Work continues on the Cooper River, and is now focused on developing a real-time, internet-based alarming and visualization system for salt-front control to be used by the USGS. |
|
Home | Products | Services | Training | Technologies | News | Links | Contact |
Send mail to
info@advdataming.com with questions or comments about this web site.
Copyright © 2003 Advanced Data Mining, LLC
