1
Deciphering the Mysterious Waters of Wakulla Springs Poornachanda Deenadayalan 1 , Casey Luzius 2 , Sarah Ellen Johnston 2 , Phoebe Zito 2 , Robert G. M. Spencer 2 , David C. Podgorski 2 1 James S. Rickards High School, Tallahassee, Florida 32301 2 Depertment of Earth, Ocean and Atmospheric Science, National High Magnetic Field Laboratory, Florida State University, Tallahassee, Florida 32306 Introduction Wakulla Springs is one of the largest and deepest freshwater springs on Earth. It is located within a karst landscape, characterized by numerous swallets (sinkholes), conduits (underground rivers) and springs (Figure 1). Karst aquifers are highly susceptible to contamination due to their hydrological connectivity between surface and subsurface waters. Surface water and associated dissolved organic matter (DOM) from both natural sources (Apalachicola National Forest) and anthropogenic sources (wastewater facility, spray field, agriculture, and diffuse inputs from the city of Tallahassee) are transported through conduit networks and discharged at Wakulla Springs Vent. In recent decades Wakulla Springs has experienced more days of darker water compared to clear water, which impacts local tourism. Investigations on inorganic nutrients have been undertaken, however, they cannot account for the increase in water color and concurrent decrease in visibility at the spring vent. Research Motivation The objective of this project is to determine if DOM composition can be utilized to estimate the fraction of clear and dark water sources that contribute to form the Wakulla Springs Vent. Simple mixing experiments were completed with water sampled directly from a conduit known to transport dark water and another that transports clear groundwater to the vent. The dissolved organic carbon (DOC) concentration, optical properties (fluorescence and absorbance) and composition by Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) were determined for the DOM in each mixture and compared with that from the spring vent. Figure 1. Map of the conduit network that supplies water to Wakulla Springs (left). Runoff from the city and dark tannin-rich water from the forest may infiltrate the conduit network and discharge at the vent, negatively impacting water clarity in the spring basin. Figure 2. Box plots showing the range of DOC concentrations at each sampling site from August 2015-16 (left). Water discharge (Q) hydrograph for Wakulla River (spring headwater) versus DOC concentration at each sampling site over the year sampling campaign. Green symbols represent clear and black represent afforested sites. The light blue symbols indicate the spring vent and dark blue are down river sites. Previous Research Samples were collected bi-weekly from August 2015-16 to delineate the role of seasonality, particularly with respect to hydrology, on DOM composition throughout the conduit network. DOC concentrations varied in the conduits that flow into Wakulla Springs depending on source and discharge (Q) throughout the year (Figure 2). DOC concentrations from the afforested sites ranged from 0.5 mg/L to 7.0 mg/L, varying with precipitation. DOC concentrations for the clear water sample sites ranged from 0.11 mg/L to 2.89 mg/L and showed little variation throughout the year (Figure 2). Figure 3. Six component PARAFAC model. Components 1, 2, and 3 are indicators of terrestrially derived DOM. Components 4, 5, and 6 are indicators of DOM derived from microbial (natural groundwater) and anthropogenic sources. Each sample collected during the 2015-2016 sampling campaign was analyzed by fluorescence spectroscopy. Parallel factor (PARAFAC) analysis was utilized to identify the statistically significant components. Figure 3 shows the six components that were identified. Components 1-3 are associated with relatively large, aromatic compounds that are common to dark water systems. Components 4-6 correspond to relatively small, aliphatic compounds associated with clear water systems. The 2015 -16 campaign showed that different conduits transported either clear or dark water to the vent at Wakulla Springs. Table 1. Percentage of clear and dark water that were mixed to generate each sample for comparison versus the vent. Figure 4. Principal component analysis of the DOM obtained from each of the 2 endmembers (1 & 10), 8 mixtures and that sampled from the spring vent. DOM Mixing Model The principal component analysis (PCA) plot shown in Figure 4 includes loadings values obtained from several analytical methods (DOC, optical spectroscopy, mass spectrometry). The plot shows that almost all of the variance (91.3%) may be explained along the x-axis. This variance can be explained simply as changes in aromaticity of the DOM. The DOM clear water endmember (1) is comprised of blue-shifted optical signatures and molecular formulas with high H/C whereas the dark water endmember (10) is comprised of red-shifted optical signatures and formulas with low H/ C. The laboratory mixing experiments (Figure 4) show a linear response between DOM composition and percent contribution of clear/dark water sources. As such, the model indicates that the water exiting the vent is comprised of 30-40% clear and 60-70% dark water. Acknowledgements Work supported by the Fish & Wildlife Foundation of Florida, Inc. and the National Science Foundation through DMR 1157490. The authors thank Edward Ball Wakulla Springs State Park, FGS, and the NWFWMD for their assistance. PCA 1 (91.3%) PCA 2 (4.77%) Sample ID Clear (%) Dark (%) 1 100 0 2 75 25 3 50 50 4 40 60 5 30 70 6 20 80 7 15 85 8 10 90 9 5 95 10 0 100 Date DOC (mg/L) 7 6 5 4 3 2 1 0 Clear Clear Clear Clear Afforested Afforested Afforested Vent Bridge 8 6 4 2 0 09/01/15 10/01/15 11/01/15 12/01/15 01/01/16 02/01/16 03/01/16 04/01/16 05/01/16 06/01/16 07/01/16 08/01/16 1800 1600 600 400 DOC (mg/L) 1400 1200 1000 800 Q (m 3 /s) Sample Site Date
