Abstract Introduction >Methods Results Discussion Conclusions Acknowledgments Literature Cited Figures and Tables PDF Version

Study sites for monitoring of carbonate sediment production were established on a mud bank named Russell Bank and in basins near Manatee Keys and Captain Key in central Florida Bay. Study sites in western Florida Bay included mud banks named Nine Mile Bank and Barnes Key Bank and a basin located near Buchanon Keys (Fig. 1 and Table 1). Russell Bank and Nine Mile Bank monitoring sites were located on mud banks where sediment cores were previously taken (Brewster-Wingard and Ishman 1999; Halley and Roulier 1999; Robbins et al. 2000) or where permanent sediment elevation tables were installed in transects across the bank to measure long-term sediment accumulation rates and bank migration (Halley et al. 2001). Elevation profiles for the Nine Mile Bank and Russell Bank study sites are available from Halley et al. (2001). No elevation profiles are available for Barnes Key Bank. Monitoring sites located in basins near Buchanon Keys, Captain Key, and Manatee Keys were located in areas where previous sediment production studies have been conducted (Stockman et al. 1967; Nelson and Ginsburg 1986; Bosence 1989a; Walter and Burton 1990; Ku et al. 1999).

Four representative bottom types were chosen for monitoring in basins to reflect bottom-type categories previously mapped by Prager and Halley (1999). These representative bottom types included sparse and intermediate density Thalassia communities and mud bottom and hard bottom communities (Prager and Halley 1999). Dense seagrass beds were located on the tops of mud bank sites including Russell, Nine Mile, and Barnes Key Banks. Hard bottom communities located in the basin near Buchanon Keys were characterized by small, sparsely distributed corals (Porites sp. and Siderastrea sp.), sponges, octocorals, and calcareous and fleshy macroalgae.

Rates of net carbonate sediment production (calcification) were measured over 24-h time periods during winter (March) and summer (July- September) at each study location from March 1999 to March 2003. Rates of calcification were determined from precise measurements of total alkalinity (TA) using the alkalinity anomaly technique described by Smith and Key (1975) whereby calcification is equivalent to half of the change in TA measured over time or spatially along sampling transects. Prior to February 2002, TA was measured using the automated Gran titration method, equations, and automated titration system described in Millero et al. (1993). The automated titration system consisted of a plexiglas water-jacketed, fixed volume (200 ml) sample cell from the laboratory of Dr. Frank Millero (University of Miami, Rosenstiel School of Marine and Atmospheric Science), a Metrohm 665 Dosimat titrator, an Orion 720A pH meter, an Orion Ross glass pH electrode, and Orion double-junction Ag/AgCl reference electrode interfaced with a personal computer. The temperature of both the acid titrant and the sample cell were controlled to a constant value of 25 ± 0.05°C with a Lauda RE106 thermostated water bath. Standardized hydrogen chloride (HCl) used for titrations and standardized reference material (SRMs) used to determine the reliability of alkalinity measurements were provided by Dr. Millero. TA was calculated from the acid concentration, cell volume, salinity, temperature, measured emf, and volume of HCl using an automated titration program. SRMs and triplicates of samples were measured approximately once every ten samples to determine accuracy and precision of our titration system. Measurement of 34 SRMs yielded an accuracy of 0.003 mmol kg^-1. Triplicate sample analyses (n = 31) yielded an average precision of 0.001 mmol kg^-1. A correction factor was determined from the measured and reported SRM values and used to correct all TA measurements.

TA samples collected after February 2002 were measured using spectrophotometric techniques described by Yao and Byrne (1998). Absorbance measurements were made using an Ocean Optics USB2000 spectrometer and Ocean Optics software package OOIBase 32. Both water sample amount (approximately 130 g) and acid additions were determined gravimetrically using a Denver Instruments PI-214 analytical balance (± 0.1 mg). Water samples were placed in preweighed glass cells (Hellma Cells, Inc. Plainview, New York) and weighed again to determine exact sample weight by difference. Baseline absorbance was measured before adding bromocresol purple indicator (0.004 M) to each sample. Titrations were performed by addition of 0.100 N standardized HCl (± 0.0001 N) from a plastic syringe fitted with a Teflon syringe needle. The pH was measured continuously throughout the titration to an end point of approximately 4.3, and the weight of the acid was determined by difference in syringe weight before and after acid addition. At the end of each titration, the solution was purged with a stream of N₂ gas presaturated with H₂O. After purging, final absorbance measurements were made, and solution temperature was determined using a Hart Scientific 1521 (± 0.1°C) handheld thermometer. TA was calculated using the equations of Yao and Byrne (1998). Accuracy of our spectrophotometric alkalinity measurements was determined by comparison to certified reference material (CRM) from the laboratory of Dr. Andrew Dickson (Scripps Institute of Oceanography; see Dickson et al. 2003). CRMs and duplicates of samples were measured approximately once every ten samples to determine accuracy and precision of our titration system. Measurement of five CRMs yielded an accuracy of 0.004 mmol kg^-1. Duplicate sample analyses (n = 6) yielded an average precision of 0.001 mmol kg^-1. A correction factor was determined from the measured and reported CRM values and used to correct all TA measurements. Based on the alkalinity anomaly theory (Smith and Key 1975), precipitation of 1 mg kg^-1 seawater of CaCO₃ causes a change in TA of 0.020 mmol kg^-1 seawater. We report calcification rates calculated from our TA measurements to the nearest 1 mg, which is well above detection limits based on our precision and accuracy.

Temperature, conductivity, pH, and dissolved oxygen (DO) were also measured continuously during monitoring expeditions using an Orion Ross pH electrode (± 0.005 pH unit), salinity (± 0.1 psu) and temperature (± 0.1°C) probes, and a YSI DO meter and pressure-compensated field probe (± 0.01 mg L^-1). pH electrodes were calibrated using Tris seawater buffers prepared at an ionic strength of 0.7 and scaled to free hydrogen ion concentration scale (Millero 1996).

Calcification at mud bank study sites was determined using a flow-respirometry sampling strategy to measure spatial chemical changes across the banks (Smith 1973; Marsh and Smith 1978). Sampling sites were located upstream and downstream from 200 to 400 m transects across Russell Bank, Nine Mile Bank, and Barnes Key Bank. Mud banks usually have extremely level and uniform tops with sloping flanks (Halley et al. 2001). Upstream and downstream sampling sites were located at the edges of bank tops above the sloping flanks to minimize variation in bathymetry along transects. Measurements on Russell Bank were taken during March and September of 1999 and 2000. Measurements were performed during October 1998 and March 1999 on Nine Mile Bank and July 2000 on Barnes Key Bank. Water samples were collected and analyzed for TA at upstream and downstream sites approximately every 4 h throughout 24-h time intervals. In situ pH, temperature, and DO were measured at the time of water sample collection. Current directions and velocities were determined by deploying Sontek Acoustic Doppler Velocimeters on the bank tops near upstream or downstream sampling sites located near the flanks of banks. Calcification rates were determined from TA, current velocity, and sample-transect length, width, and depth. The change in TA (mol m^-3) was determined by difference in concentration between upstream and downstream sampling stations. The volume of water transported along a transect per unit time (m³ 4 h^-1) was calculated from the current velocity and length, width, and depth of each sample transect (assuming a transect width of 1 m). Transect lengths were determined from latitude and longitude of upstream and downstream sample locations. Transect area (m²) was determined from its length and an assumed width of 1 m. Rates of calcification, C (g CaCO₃ m^-2 4 h^-1), were calculated as half the change in TA (mol m^-3) times the volume of water transported along a transect per unit time, V (m³ 4 h^-1), divided by the area of a transect, A (m²), times the molecular weight of CaCO₃, MW (100.09 g mol^-1): C = 1/2TA X V/A X MW (Smith 1973; Smith and Key 1975; Barnes and Devereux 1984).

Calcification rates were determined in basins by measuring TA changes in a large incubation chamber (surface area of 11 m²), called the Submersible Habitat for Analyzing Reef Quality (SHARQ, U.S. Patent #6,467,424 B1), deployed on representative bottom type communities using methods previously described by Yates and Halley (2003). Bottom types included intermediate Thalassia beds adjacent to hard bottom communities located near Buchanon Keys, intermediate Thalassia beds adjacent to mud bottoms near Manatee Keys, and sparse seagrass beds near Captain Key and Manatee Keys. Intermediate seagrass beds were measured during March and September 1999 and 2000. Hard bottom sites were measured during March 2000. Mud bottom sites were measured during March and September 2000 and March 2003, and sparse seagrass beds were measured during March 2001 and March 2003. Salinity, temperature, pH, and DO were measured continuously through the incubation chamber's flowthrough analytical system throughout the duration of incubation periods (20-28 h). Fluorescein dye was injected into the incubation chamber during each deployment to determine incubation chamber volume, mixing rate, and leakage as described previously by Yates and Halley (2003). Water samples were removed from incubation chamber sample ports every 4 h for TA measurements. TA of ambient water (i.e., located outside of the incubation chamber) was measured every 4 h during 8 of 21 incubation chamber deployments. Rates of net calcification (C) were calculated for each 4-h interval between alkalinity measurements during chamber incubation periods using the equation C (g CaCO₃ m^-2 4 h^-1) = 1/2TA (mol m^-3 4 h^-1) X SHARQ volume (m³)/SHARQ surface area (m²) X MW (g mol^-1).

Net daytime carbonate sediment production rates (C_day, g CaCO₃ m^-2 d^-1) were calculated for both banks and basins by integrating calcification rate curves with respect to net calcification rates determined every 4 h throughout incubation periods from sunrise to sunset. Sediment production rates for 24 h (C_net, g CaCO₃ m^-2 24 h^-1) were calculated by integrating each data set over the entire diurnal cycle. A y-axis reference point of zero was used in all integrations. Nighttime rates (C_night, g CaCO₃ m^-2 night^-1) were determined by the difference.