Paul A. Conrads¹ and Edwin A. Roehl, Jr.^2
1USGS South Carolina Water Science Center, Columbia, SC, USA
²Advanced Data Mining, Greenville, SC, USA

The real-time Everglades Depth Estimation Network (EDEN) has been established to support a variety of scientific and water management purposes. The expansiveness of the Everglades, limited number of gaging stations, and extreme sensitivity of fauna to small changes in water depth has created a need for accurately predicting water depths at locations between the gages. This has been challenging because an ultra low gradient makes interactions between meteorology, vegetation, topology, and hydrology very complex. Linear techniques such as interpolation and ordinary least-squares regression and have under-performed because of the system’s non-linear dynamics. This paper presents an alternative approach that employs artificial neural network (ANN) models to perform multivariate, non-linear interpolation between gaging stations.

Using a combination of static and dynamic variables, predictions are generated in two modeling steps. The dynamic variables were 30-month time series of daily water depths at 16 stations and water levels relative to sea level at 3 other stations. Static variable values were obtained from a previously developed GIS application having a 400 square-meter grid. They included cell coordinates and percentages of vegetation types (slough, prairie, sawgrass, or upland) for approximately 2300 cells, covering 370 square kilometers. The first ANN model interpolates mean water depths (for the period of record) from input static variables and mean water depths and levels at the gaging stations. The second ANN model predicts day-to-day variability about the interpolated means using a combination of static and dynamic variable inputs. A complete interpolation at a given cell is computed by summing the outputs of both models. Six of the water-depth gages were withheld from model development to validate model accuracy. Prediction accuracy was greatly improved, resulting in an average root mean square prediction error at validation gages of only 3 centimeters (0.1 feet), or 4 percent of the dynamic range.

Contact Information: Paul A. Conrads, USGS South Carolina Water Science Center, Stephenson Center – Suite 129, 720 Gracern Road, Columbia SC, 29210 USA, Phone: 803.750.6140, Fax: 803.750.6181, Email: pconrads@usgs.gov

(This abstract is from the 2006 Greater Everglades Ecosystem Restoration Conference.)

Back to Project Homepage