During periods of flow, water samples were collected for the determination of nitrogen and phosphorus species that truly represent the stream cross-section. This was accomplished by collecting depth- integrated samples and point (grab) samples from mid-stream at 0.5- and 1.0-m depths for comparison. These samples were collected mostly upstream of the gated control structures at locations where most of the historical data were collected. Additionally, hydrologic data were collected concurrent with each sampling event for use in the determination of instantaneous discharge. Based on these data, statistical comparisons were made between point (grab) and depth-integrated samples, and models were developed to estimate total nitrogen and total phosphorus loads.

Depth-integrated and point (grab) samples for nutrients were collected from the gated control structures at the east coast canal sites in Miami-Dade County during periods of flow. Most of the samples were collected near low tide when the gates at the sites usually are open. Depth-integrated samples were collected by means of the equal-width-increment method, which commonly is used when a discharge measurement is not made before sampling. In this method, the width of the stream is subdivided into equal-width intervals with a sampling vertical associated with each interval.

In each cross section for the study, 8 to 10 verticals were used. The first vertical selected was half the distance of the first interval from the edge of the stream bank, and the other verticals were equally spaced apart across the stream from the first vertical. Because flow velocities of the east coast canals in Miami-Dade County are nearly always less than 2 ft/s (feet per second), the weighted-bottle method was used to collect each sample. The weighted bottle does not sample isokinetically (nozzle velocity equal to stream velocity); however, it can be used during low flows and when differences in water quality across the stream are believed to be insignificant. In the initial step, the vertical with the greatest velocity is selected; the weighted bottle is then lowered and raised at a constant rate so that it is not overfilled when returned to the surface. For each specific site, this transit rate was maintained throughout for all the verticals in the cross section, and a sample from each vertical was composited in a churn splitter for processing. The transit rate as well as the nozzle size was varied for each specific site in order to prevent overfilling of the bottle.

During sample processing, the water in the churn was stirred by the churn disc at a rate of about 9 in/s (inches per second) to minimize error, with care being exercised to prevent the churn disc from breaking the surface. A total of 125 mL (milliliters) of this composite sample was siphoned off into an amber polyethylene bottle, chilled immediately, and within 48 hours was shipped to the USGS Water Quality Service Unit in Ocala, Fla., for analysis. The constituents determined were total organic nitrogen, ammonia, nitrite, nitrate, nitrite plus nitrate nitrogen, total phosphorus, and orthophosphate. A point (grab) sample was collected concurrently with each depth-integrated sample. Point (grab) samples, depending on the site, were collected at 0.5 or 1.0 m below the surface near the centroid of flow by using a Niskin bottle. This bottle is spring loaded so that a messenger can trip it shut at the appropriate depth. After collection of the point (grab) sample, 125 mL of the sample was transferred from the Niskin bottle into an amber polyethylene bottle, chilled, and shipped for analysis according to protocol.

Before each sampling trip, the Niskin bottle, weighted bottle, and churn splitter were cleaned with a dilute nonphosphate detergent and then rinsed with tap water followed by deionized water. Two rinses with native water were required for the samplers before sample collection at each east coast canal site. Between sampling sites, the samplers were rinsed with deionized water before being rinsed with native water.

To ensure the integrity of the field data collected for the study, quality assurance samples (equipment blanks, field blanks, and split samples) were used extensively in the data-collection phase. Blank solutions essentially are samples free of the analytes being determined in the environmental samples. An equipment blank is a blank solution that is processed through all of the equipment used in the collection and processing of the environmental samples. Field blanks are actually equipment blanks done in the field and are subject to all aspects of the data-collection efforts as the environmental samples, including processing, preservation, transport, and laboratory handling. Two types of field and equipment blanks were used: one for the Niskin bottle and the other for the weighted bottle and churn splitter. These two field blank solutions consisted of inorganic blank water prepared at the USGS Water Quality Service Unit and were processed and analyzed according to protocol. Analytical results from the equipment blanks and field blanks indicated that most of the concentrations for the individual nitrogen and phosphorus species were below the detection limits for the analytical methods.

Split samples were collected concurrently by the USGS and DERM to verify interlaboratory accuracy. The sampling data consisted of point (grab) samples that were collected from the same Niskin bottle by both agencies. The samples were sent to the USGS Water Quality Service Unit and to the DERM laboratory for analysis.

To top of page