The tentative locations of all Pilot year and field years 2 and 3 current meter and sediment trap arrays were presented (figure 1 and figure 2) and discussed. There was some concern about the lack of current meters on the western side of the lake. However, the main objective of the deployment strategy is to position our moorings at the best possible location for observing the offshore migration of the plume and to have a sufficient mooring density so as to be able to conserve mass over as broad a range of time scales as possible from the data alone. Current information in other regions of the study area (and whole lake) will depend on the hydrodynamic model, lagrangian drifter data and intermittant measurements from the benthic layer tripod(s).
There is an ambitious pilot program planned with 10 arrays of current meters and 6 arrays of 9 (23 sample) sequencing traps scheduled to be deployed October/November, 1997. Retrieval - servicing - redeployment of the current meters will be in 2 stages (June and July, 1998 and 1999) of approximately half of the arrays each time. This will provide continuous coverage for all 3 field years. Trap retrieval will follow the July current meter cruise with redeployment in August. Sampling during the month long interim will be with non-sequencing traps of the same design, allowing complete temporal coverage for the 3 field years.
Current meter and trap data and trap sub-samples will be made available to program participants as soon as possible (ca. 2 months after retrieval).
Current Meter Details
Two types of current meters will be used and are indicated in figure 1. The first type is a Vector Averaging Current Meter (VACM) and the second type is an Acoustic Doppler Current Profiler (ADCP). The VACM moorings will have 2 instruments each with one near the bottom and the other approximately 12m below the surface. Water velocities and temperature will be sampled at 15 min intervals and approximately 2 additional thermistor/data loggers will be positioned on the same mooring between the 2 VACMs. The ADCPs will continuously sample and produce an averaged current profile every 30 minutes. In the years following the Pilot Year an additional seven current meter moorings will be deployed as well as up to 10 thermistor string moorings positioned along two transects in the primary study area.
Sediment Trap Details
GLERL has been using cylindrical sediment traps in its lake studies programs since 1977. The traps are usually deployed as anchored arrays using subsurface buoyed 3/8" steel cable. Cylindrical traps have a high collection efficiency in lake environments and have proved satisfactory in many lake studies. The accuracy of calculated fluxes is poorly understood, but depends on the trap design, the types of particles in the fluid and the currents at the site.
In 1994 - 96 GLERL deployed sediment traps in the "A" region. Results from these deployments have provided critical background input to our COP experimental design. Mass fluxes in the nearshore region (water depth 30-35m) exceeded 500g/m2/d during brief periods in the spring and fall and these massive samples caused the traps to fail. In order to successfully sample during these events, we had to reduce the trap opening from 8" ID to 2" ID.
In the COP project we will deploy 3 types of traps. Our primary collections, in the nearshore and bottom boundary region, will be with a cylindrical sequencing trap with an inner diameter of 2" (5 cm) and an aspect ratio of 8:1 above the funnel. A larger (to collect sufficient mass for constituent analysis) offshore version has an ID of 8" (20 cm) with the same aspect ratio. A computer-controlled carrousel contains 23 sixty ml polyethylene bottles, which rotate under the funnel at preprogrammed intervals. The sequencing traps are deployed with the collection funnel feeding to an empty opening (no collection bottle). After a preprogrammed period of time the carousel will move the first collection bottle under the funnel. The remaining 22 bottles will follow in a preprogrammed sequence. After retrieval, the sample bottles are removed from the traps and to the laboratory in cold storage (4°C). The traps have on-board intelligence that records the time of each sequence and various system checks.
For approximately one month per year, while the sequencing traps are back at GLERL for servicing, we will deploy an equal number of simple traps 10 cm in diameter with an aspect ratio of 5:1 above the funnel opening to a 500 ml polyethylene bottle. The 500 ml bottles in the simple traps were poisoned with 25 ml of chloroform and filled with distilled water prior to deployment. The 60 ml polyethylene collection bottles in the sequencing trap were poisoned with 6 ml of chloroform and filled with distilled water immediately prior to deployment. This concentration of chloroform is an effective preservative and results in a supersaturated solution, with beads of chloroform remaining after retrieval.
In order to estimate trap collection precision and intercalibrate between the 10 cm diameter traps and the 20 cm diameter sequencing traps, a series of deployments were made in regions with a wide range of fluxes on specially constructed brackets to assure identical depth and exposure. The 20 cm traps used in these tests did not have sequencing capability, but were identical in other aspects. When deployed in replicate, both the 10 cm and 20 cm traps showed good repeatability with paired t-test showing equal means (P<0.05) in all 4 comparisons. The 10 cm traps replicate with an average difference between pairs of ± 11% and the 20 cm traps replicate with an average difference between pairs of ± 14%. There was little bias (slope = 1.05) between the two types of traps.
A similar intercalibration will be conducted with the 2" ID traps. During the pilot year, pairs of 2" and 8" ID sequencing traps will be co-located at a low and medium flux site (46 data pairs). In addition, 4 pairs of non-sequencing traps (identical in all aspects with the 2" and 8" sequencing traps) will be deployed near bottom at depths ranging from 20-40m near GLERLs home port (Muskegon, MI) to collect a large number of data in regions of high flux.
After arrival at the lab, the trap samples are allowed to settle in a refrigerator for a day, then overlying water is carefully siphoned off and the residual is freeze dried. After drying, samples are weighed and transferred into precleaned scintillation vials for storage in a freezer. All trap samples will be weighed on an analytical balance calibrated to within ± 1mg with known standard weights during each weighing session. Virtually all samples are greater that 100 mg, thus all mass weights have an accuracy and precision of less than 1 % (coefficient of variation).
Workshop Reports
EEGLE Homepage Greg Lang, EEGLE Web Manager,
gregory.lang@noaa.gov