In lieu of an abstract, here is a brief excerpt of the content:

Southeastern Geographer Vol. 27, No. 2, November 1987, pp. 71-89 HYDROLOGICAL SIMULATION OF OKEFENOKEE SWAMP UPLAND WATERSHEDS USING A DISTRIBUTED MODEL: THE EXAMPLE OF THE BLACK RIVER CATCHMENT* George A. Brook and Chin-Hong Sun INTRODUCTION. The Okefenokee Swamp is one of the largest freshwater wetland complexes in the United States. It is a National Wildlife Refuge, a designated wilderness area, and formerly a National Science Foundation Long-Term Ecological Research (LTER) site. About 56% of the 3,925 km2 watershed is occupied by uplands that drain into the swamp. The largest area of uplands, called the "terrace uplands" by Brook and Hyatt (J), lies to the west and northwest of the swamp (Fig. 1). The hydrology of the swamp is obviously a critical aspect of its ecology. Long-term hydrological budgets based on water balance estimates , assuming no change in water storage, suggest that the swamp receives 72—78% of its water directly as precipitation, the remainder coming largely as streamflow from the terrace upland region. (2) This streamflow component of recharge has been estimated in the past by residuals using water balance models for the uplands. Although the general conceptual water balance models that have been developed have improved our knowledge of the swamp watershed hydrology, they have not provided a means of modeling the temporal and spatial variations of streamflow into the swamp—variations that have a significant impact on the swamp fauna and flora. (3) In an attempt to better understand the flux of water—and ultimately the flux of nutrients—into the swamp from the uplands, a distributed hydrologie simulation model COASTAL was developed for Okefenokee upland watersheds to simulate daily streamflow into the swamp. In the absence of daily streamflow data for any of the terrace upland catch- * University of Georgia Okefenokee Ecosystem Investigations, Publication No. 75. This research was supported by National Science Foundation grants DEB 8110639 and DEB 8114823, and by U.S. Department of the Interior grant G-900(3). Dr. Brook is Professor of Geography at the University of Georgia, Athens, GA 30602 and Dr. Sun is Associate Professor of Geography at National Taiwan University, Taipei, Taiwan. 72 Southeastern Geographer Waycross Waycross Ridge Cowhouse Sßv Island\3 Homerville:: ^e Cypress / Mary's Moniac Camp Cornelia Terrace Uplands Ridge Uplands Islands and Island-like Uplands Black River Watershed Stream Gage Site Drainage Divides Within the Okefenokee Watershed Fig. 1. Characteristics ofthe Okefenokee Swamp watershed and the location of the Black River catchment. ments in the swamp watershed, the model was tested on the Hurricane Creek watershed 40 km to the northwest. This watershed, which was gauged by the U.S. Geological Survey between 1951 and 1971, was also used by Lumb to test the capabilities of the Stanford, Kentucky, Georgia Tech, Kansas, and TVA models to predict hydrological conditions in Georgia Lower Coastal Plain watersheds . (4) COASTAL outperformed the five models tested by Lumb in Vol. XXVII, No. 2 73 predicting streamflow from the Hurricane Creek watershed suggesting that the assumptions on which it is based are essentially correct. (5) The means of eight yearly correlation coefficients between daily measured and simulated streamflow for the period 1957—64 were 0.66, 0.71, 0.70, 0.40, and 0.54 for the Stanford, Kentucky, Georgia Tech, Kansas, and TVA models, respectively, and 0.85 for COASTAL. (6) On November 9, 1984, while COASTAL was being developed and tested using Hurricane Creek watershed data, a digital stage recorder was installed in the Black River watershed in the northern part of the Okefenokee Swamp terrace uplands region (Fig. 1). Daily streamflow data were collected with two objectives in mind: 1) to calibrate COASTAL for an Okefenokee upland watershed and assess its potential to predict streamflow into the swamp; and 2) to determine to what extent parameters developed in calibration with Hurricane Creek data could be transferred to Okefenokee upland watersheds —many with very little or no streamflow data. THE WATERSHED HYDROLOGY MODEL COASTAL. In developing COASTAL it was assumed that in Georgia Lower Coastal Plain watersheds with permeable sandy soils ground water discharge accounts for the majority of streamflow—overland flow contributes small amounts of streamflow only after intense or long...

pdf