Sometimes a camera hundreds of miles up in the sky offers the best
perspective on microscopic sea plants. |
|
If you could see absolutely everything through that
mirror, youd observe a space teeming with all kinds
of life, from the biggest whale to the most microscopic
bits of plant-like aquatic organisms called algae. And if
you could gaze infinitely skyward past the bright sun,
youd catch a glimpse of a satellite that biologists
are using to study this vast web of life, starting with the
algae.
Its funny to think that scientists are using
high-tech space satellites, and not microscopes, to study
such tiny lifeforms. But the approach makes perfect sense
to marine biologists such as Raphael Kudela of the
University of California at Santa Cruz, who is trying to
understand the balance of life in the ocean. In the marine
realm, as on land, small creatures get eaten by bigger
animals, which get devoured by still larger predators, and
so on. Thus energy in the form of food travels up the
so-called food chain. But exactly what controls the growth
of sea life? "Where does the energy go? These are
biologists big, driving questions," says
Kudela.
Scientists know that one important kind of algae,
called phytoplankton, forms the base of the food chain. The
microscopic phytoplankton hold a key to many riddles about
everything from global warming to overfishing. If
scientists could measure exactly how much phytoplankton
floats around in the ocean, for instance, they could make
reasonable predictions about how many fish, whales, sea
turtles, or sharks the ocean can feed. And knowing that
answer would help answer the urgent question of how much
humans can fish without upsetting the oceans
ecosystem or pushing different species towards extinction.
Alternatively, monitoring how phytoplankton respond to
changes in climate from natural or manmade causes might
offer clues to the planets state of health.
Now, studies by Kudela and a colleague are taking a
closer look at these tiny fish-snacks. Kudela has been
using satellite pictures of the Pacific Ocean off central
California in a new formula for predicting the amount of
phytoplankton, in pounds, that will become food for fish in
the area on any given day, month, or year. This new
phytoplankton growth model, which Kudela developed together
with marine biologist Francisco Chavez of the Monterey Bay
Aquarium Research Institute (MBARI), combines data
collected from boat cruises, aquarium moorings, and most
importantly, the satellite.
The best thing about using a satellite to do algae
research, say the biologists, is that it gives them a view
they cant get anywhere else. "Satellites are the
only way to get that really big picture," says Kudela.
"On a boat you can only go out so far and get so much
data. Youre always wondering if youre getting
the big picture or not."
WHEN SCIENTISTS USE SATELLITES to study the land or the
sea, they call it remote sensing. First used decades ago to
observe dry land, the technology taught scientists about
the complex seasonal and yearly changes of plants, animals,
and geography. The tool was so helpful that marine
researchers soon wanted to see whether it would unlock the
secrets of the oceans, too.
Kudela studies pictures of Monterey Bay and beyond
taken by a satellite called the Sea-Viewing Wide Field
Sensor, or SeaWiFS, which is owned by NASA and Oribmage, a
private company. The three-foot-long, torpedo-shaped
satellite was lofted into space five years ago by a cruise
missile launched from the back of a 747 jet. Every day,
SeaWiFS beams images from all over the globe to more than
80 subscribers worldwide, including Kudela.
SeaWiFS takes pictures at eight different wavelengths
of light. It takes one photograph each of violet, blue,
yellow, green, and red light and infrared heat that
radiates back to space from the oceans surface.
Images recorded at the other two wavelengths allow
researchers to correct for the scattering of light that
occurs in the earths atmosphere. Each color reveals
something different about the ocean. Green light, for
example, tells scientists how much green chlorophylla
pigment found in algaeis floating at the surface.
SeaWiFS data has proven a goldmine for a range of
applications. Scientists use the satellite to track the
movement of surface ocean currents that are otherwise
invisible to the eye. Because water currents each have
their own unique temperature, the satellite can detect
them; it measures temperature by reading the infrared heat
signal at the oceans surface. The satellite is
sensitive enough to distinguish between currents only
one-twentieth of a degree apart. SeaWiFS can also follow
currents of dissolved sediment and pollutants as they run
from rivers and dissipate into the ocean.
Still other investigators use satellite photos to track
phytoplankton to figure out what ocean conditions give rise
to blooms of toxic algae. Called red tide, these dangerous
blooms poison fish, seals, and shellfish. They can also
make humans terribly sick.
Phytoplankton are generally underrated, Kudela says,
but theyre important to the planet. Fully half of the
plants at the bottom of the global food chain live and grow
in the ocean. Anchoring that chain, phytoplankton feed on
inorganic nutrients such as carbon, nitrogen, and ammonia
that enrich cold ocean water. Chlorophyll pigments in the
algae absorb sunlight and allow them to turn those
nutrients into more phytoplankton, via a process called
photosynthesis. Because of this talent for making something
out of nothing, ocean biologists call them "primary
producers."
Most biologists working on the problem of primary
production first calculate the weight of phytoplankton
produced in a given time frame. Then they roughly estimate
what percentage of the algae will be eaten by animals.
Kudela and Chavez say theyve created a better formula
that skips the first step and instead directly calculates
how many pounds of phytoplankton travel up the food chain
into the bellies of crustaceans, fish, whales, and the
like. The researchers call this measure "new primary
production."
"Most biologists are still looking at primary
production," Kudela says. "But the new primary
production is really what you want to be measuring if
youre interested in where most of those nutrients are
going."
To arrive at their calculation, they plug five pieces
of data into a computer equation. The first three come from
SeaWiFS: the surface temperature of the ocean water, the
amount of chlorophyll in the water, and information about
which direction the chlorophyll is moving. The researchers
also include water temperature at about six hundred feet
below the surface, and wind speed and direction in their
calculations.
"Its an integrated observing system,"
says Chavez. "We use data from ship point
measurements, moorings, and satellites." Each data
source provides essential information, but the satellite
provides a valuable comprehensive view of Monterey Bay.
Case in point: An interesting pattern emerged when the
researchers compared images of the open ocean with those
taken near the coastline. Far offshore, the ocean reflects
only a deep blue color back to SeaWiFS. Thats because
theres very little phytoplankton floating at the
surface so far from land. Most of it grows in shallow
waters off continental coasts, where cold, mineral-rich
water from the ocean floor slides up the continental shelf
to the surface. This water has likely been flowing along
the ocean floor for thousands of years, absorbing nutrients
released by decomposing plants and fish before reaching the
surface. One plume of this water wells up right off the
coast of Davenport. From there, one current travels north
towards Alaska, while another moves south into Monterey
Bay.
So far, Chavez and Kudela are pleased to report that
their formula is accurate within a factor of two at
predicting new primary production in Monterey Bay. That
means that if the mathematical model predicts two pounds of
phytoplankton per cubic yard of ocean water, the actual
value would lie somewhere between one and four pounds.
With a reliable formula in hand, the scientists can
estimate how much food is in the ocean to support fish.
Laws of nature say that an animal population will grow if
theres ample food, but shrink if substenance becomes
scarce. Instead of trying to count all the fish hiding deep
in the ocean-an impossible task-scientists can calculate
population numbers by simply looking at how much food the
fish have. Then, in turn, they can estimate how many fish
could be harvested without harming the survival of
different species.
With the oceans in increasing danger from overfishing,
Kudela and Chavezs formula will add some certainty to
a field dominated by guesswork. "Many people are
worried that were taking more out of the oceans than
we can support," says Kudela. He hopes environmental
agencies will someday use the new primary production
estimates to set limits on fishing. The formula could even
help provide independent scientific verification of whether
fishermen are really restricting their catches to the
numbers they report, he says.
RECENTLY, KUDELA AND CHAVEZ were able to test their model
by retrospectively crunching data collected under the
extreme conditions of the El Nino weather pattern during
the winter of 1997 to 1998. Along the central California
coast during that period, many seals and sea lions washed
up on shore, dead or dying of starvation. Seventy-five
percent of sea lion pups died that breeding season, and the
average weight of a weaned elephant seal was the lowest
ever recorded. To a lesser extent, the same thing had
happened during the 1992 - 1993 El Nino.
Why? El Nino had tipped the oceans balance and
brought abnormally warm water to California. Scientists
know that phytoplankton cannot thrive in warm water, which
lacks the nutrient richness of the old, cold water from the
ocean floor. The devastation at the base of the food chain
rippled up to the top. Without phytoplankton, the fish had
little to eat. And without fish, animals starved.
Oddly, however, readings from the moorings operated by
the Monterey Bay Aquarium six and twelve miles offshore
didnt record much of a difference in phytoplankton
levels, and this puzzled researchers. Some theorized that
the inner bay had sheltered a large amount of phytoplankton
from the warm water, and that the real damage had occurred
just beyond their instruments. But they couldnt prove
it.
Fortunately, SeaWiFS, which was launched in the summer
of 1997, had recorded the whole thing. When Kudela and
Chavez finished looking at its pictures, they found proof
that the theory was right. Warm water had reduced the
phytoplankton population two hundred miles offshore to just
one fifth of its usual size. Most fish had migrated out of
the bay and beyond to colder waters, where they could find
food. But many of the sea lions and seals that stayed
behind starved, because so little was left for them to
eat.
Whats more, when Kudela and Chavez ran the
SeaWiFs data through their formula, results showed that 98
percent of the phytoplankton that normally becomes food was
killed or failed to grow in the warm El Nino waters. The
extent of the mass starvation made even more sense.
Beyond the California coast, the new model can be
customized to help other researchers studying other parts
of the ocean. A complex environment like Monterey Bay, with
wind and currents coming from all directions, requires
different numbers than a more stable system, like the open
ocean. Andrew Thomas, a marine biologist at the University
of Maine at Orono, uses SeaWiFS data to study red tide in
the Gulf of Maine, and to track how ocean color changes as
a result of upwelling currentsone way of monitoring
the health of the oceans. Thomas says a formula to predict
new primary production would be highly useful to his
studies. "Its one of the missing parts we have
to have," he says.
Keeping track of the yearly cycle of new primary
production is one more tool biologists can use to keep tabs
on ocean conditions, says Kudela. "You can watch to
see the impacts that humans have had over time," he
says. With so much uncertainty about how global warming
will affect the oceans, monitoring them closely has become
especially important.
Scientists forecast another El Nino for the coming
winter of 2002-2003. Californians everywhere are groaning
with dread, but Kudela and Chavez are getting excited.
Its another chance for them to further test and
refine their formula under another set of extreme weather
conditions. Next winters SeaWiFS pictures will bring
more answers, and they just cant wait.
|