HYDROLOGIC DESCRIPTION OF THIRTY-ONE ISOLATED WETLANDS EMBEDDED IN A LONGLEAF PINE / WIRE GRASS FOREST: INFLUENCE OF CONNECTIVITY TO REFERENCE CONDITION AND AGRICULTURAL STORM RUNOFF

Small isolated wetlands occurring in dense clusters on the agriculturally developed Dougherty Plain represent substantial surface water storage in this physiographic province. The Dougherty Plain is a karstic region in southwest Georgia with numerous episodic and intermittent streams as well as isolated wetlands embedded within uplands. Hydrologic characterization of small isolated wetlands can yield insight to their biogeochemical and habitat provision roles within the landscape. The goal of this study was to characterize the stage frequency of 31 isolated wetlands using probabilistic methods. The wetlands analyzed by this study are embedded within a long leaf pine / wire grass ecosystem with several episodic stormwater flow paths. Our analysis specifically focused on comparing wetlands based on connectivity to stormwater runoff channels. We compared wetlands isolated from episodic flow (NC, n=19) to wetlands connected to either agriculturally generated runoff (AC, n=6) or forest runoff (RC, n=6). Over 14,000 bi-weekly wetland stage observations were used in this study (8,473 for NC wetlands, 2612 for AC wetlands, and 3,075 for RC wetlands). Empirical cumulative distribution functions (ECDF) were calculated for each wetland’s stage data and the mean ECDF for each wetland type was compared with a Kolmogorov-Smirnov tests which indicated that all wetland classes differed. Interestingly, mean dry bed probability (probability that a wetland is dry) and mean maximum observed stage probability (probability of wetlands being at maximum observed stage) did not differ among the three types of wetlands. Our results indicate that while wetlands connected to episodic flows differ in overall stage frequency distribution but not in frequency at the extreme ends of the distribution (dry or maximum observed stage). Therefore even if isolated wetlands do not hold water more frequently or stay at maximum observed stage longer than episodically connected wetlands, they do more frequently exhibit stages from the middle of the distribution. Which means that these isolated wetlands have a higher probability of water being available for wildlife use and are much more likely to retain nutrient and energy inputs than connected wetlands. Connected (RC or AC) likely have their lower probability of inundation due to export of water.