4
from the control tank. The paired sampling canister/net arrangement allows for the residual from
the cod-end of one net from each pair to be processed for examination while filtration continues
via the other net, thereby avoiding clogging. In this way unimpaired filtration back and forth
between each pair of nets continues until a total of 3 to 7 m
3
has been processed from each of the
treated and control ballast tanks respectively. The sampling canisters are designed to allow
complete immersion of each net during the filtration process, thereby minimizing trauma to
filtered organisms. At the end of each trial (after five day hold times), the control and treated
ballast tanks were drained and processed as described above, with treated water first passing
through the BIO-SEA system for additional UV irradiation (but not filtration) prior to sampling.
The proportion and total concentration of live versus dead organisms > 50
µ
m were
determined using standard movement and response to stimuli techniques and this live/dead
analysis took place within two hours of collecting the individual samples. Three (3) m
3
was the
volume collected for control water upon filling and discharge from the control tanks and for
filling of treated tanks (high numbers of live organisms). Sample-sizes upon discharge after five-
day hold times averaged 4 m
3
(
trial 1 and 2) and 6 m
3
(
trial 3 and 4), or as large as allowed by
isokinetic sampling during de-ballasting at rates of approximately 100 m
3
/
h (trials 1 and 2) and
150
m
3
/
h (trials 3 and 4). Depending on concentrations, quantification of organisms > 50
µ
m in
initial samples (upon ballasting) and control samples may have required analysis of sub-samples
and extrapolation to the entire 3 m
3
.
The > 50
µ
m samples was then also fixed with buffered,
10%
formalin in 500ml Nalgene bottles and shipped to the Smithsonian Environmental Research
Center (SERC) for additional taxonomic evaluation. Total counts and general taxonomic
classification was conducted under a dissecting microscope at 25X, except for some taxa, which
were removed and identified using a compound microscope. Larval forms of invertebrates were
identified to higher taxonomic levels such as order (e.g., Decapoda) suborder (e.g.,
Balanomorpha) or class (e.g., Bivalvia). Adults were identified to species in most cases.
Live Organisms 10 - 50 µm in size:
Live organisms 10 - 50 µm in size were collected over the entire course of the
fill/discharge operation of each tank via an time-integrated sample system. Flow rate was
controlled by a values and flowmeters with the water emptying into a 100 l conical bottom,
fiberglass, algal-culture cylinder. Samples collected in the cylinder were aerated to keep entire
water column well mixed and in suspension. The sampling rate of the whole-water was 1.5
l/minute at 300 m
3
/
hour ballast-flow-rate (and less as flow-rates decrease). Samples drawn for
this size-range were subject to three distinct analyses and counts (described briefly below and in
detail in SOPs).
For live, unfiltered samples, 3 replicate samples were collected from each cylinder in 500
ml, opaque (brown) Nalgene bottles.
All of the live, unfiltered samples were processed or
examined within three hours of collection on the MERC Mobile Test Platform or at nearby
partner laboratories.
These samples were
stained using a combination of CMFDA (5-
chloromethylfluorescein diacetate) and FDA (fluorescein diacetate) as a selective live/viable
indicator. Samples stained with CMFDA+FDA, were incubated and observed on a Sedgewick
Rafter slide using a Olympus IX-51 inverted phase/fluorescent microscope. Cells were scored as
live when showing strong fluorescence signature under excitation (some cells also show
motility). This approach has been validated for use in the Chesapeake Bay (Steinberg et al.,
2011)
and provides the data for comparison to discharge standards.