Sustaining and Mining the Time Series: Days to Decades

T
he terrestrial landscape has been modified extensively by the clearing
of forests to build homes, grow crops, and raise livestock. These delib-
erate intrusions have greatly modified the abundance of native species but
have been considered necessary for human survival. The uncomfortable realiza-
tion that human activities on land are changing the atmosphere and ocean in

unforeseen ways has been brought home by two noteworthy by the National Science Foundation and focus on blue-water
examples: the growth of the ozone hole in response to the oceanography. The third time series, privately funded at
release of synthetic chlorofluorocarbons used in refrigera- MBARI by the David and Lucile Packard Foundation, has
tion systems, and ocean acidification and the potential been documenting daily-to-seasonal-to-decadal variability
for global warming caused by the increase in atmospheric for more than 16 years and focuses on the green, coastal
carbon dioxide due to our burning of fossil fuels. Such ocean off central California.
global-scale processes may change climate and ecological
Under the leadership of Francisco Chavez, MBARI’s
systems in unforeseen ways that could affect political and
approach to time-series collection is multidisciplinary,
economic stability if we are not suitably prepared, so it is
imperative that we learn as much as possible about the
causes and effects of such change. This becomes a diffi-
cult task, as over the past several decades researchers have
found that natural climate variability is much more intense
than previously imagined, as evidenced by recent strong
El Niños and discovery of extended, natural climate cycles
such as the Pacific Decadal Oscillation. It turns out that
the magnitude of the natural fluctuations is similar to or
larger than that predicted to result, initially, from human-
induced perturbations to global climate, making it difficult
to clearly identify the sources of ecosystem variability.
While scientists continue to debate whether observed
changes in climate and ecosystems are human-induced or
part of the natural environment, all agree that long-term
records are required to accurately characterize and eventu-
ally understand year-to-year changes. And all agree that
such records are severely lacking for the ocean. So why are
these long time series so rare? It takes many years before a
time series pays off by accumulating a long enough record
so that signal can be extracted from noise. During those
years before payoff, substantial effort and investment must
be maintained, sometimes for decades. Such long-term
commitment is difficult to sustain, both personally for Figure 16: The R/V Point Lobos (nearshore day expeditions) and
scientists and institutionally for funding agencies. Despite the R/V Western Flyer (offshore overnight expeditions) routinely
these obstacles three pioneering ocean time series, all begun collect information about Monterey Bay and the contiguous waters
in the late 1980s, continue to provide records of the ocean’s of the California Current System. The picture shows the launch of
a profiling conductivity, temperature, and depth (CTD) recorder
changing physical, chemical, and biological character. Two
with a rosette carrying bottles that collect water at depths chosen
of them, the Hawaiian Ocean Time-Series (HOT) and the by the operator.
Bermuda Atlantic Time-Series Study (BATS), are supported

2004 Annual Report | 15

 Sustaining and Mining the Time Series: Days to Decades

multiplatform, observational, and experimental. Begin- of thin biological layers, natural iron fertilization of the
ning in 1989, shipboard measurements of physical, chemical, ocean, and the development of harmful algal blooms use
and biological parameters were made during time-series the MBARI time-series data as the starting point for under-
cruises in Monterey Bay (Figure 16). These observations were standing the environmental background for experiments
later extended offshore into the California Current (1992) and observations.
and into the equatorial Pacific (1997). These high-quality
The data sets collected to date, augmented with informa-
data sets have been augmented by sensor and platform
tion from satellites, now span many time and space scales
development to automate data collections. The first techni-
and permit construction of climatologies against which
cal development was the OASIS mooring, which has been
change can be measured. In the early years, the seasonal
producing high-frequency time-series data in Monterey Bay
and spatial pattern of the physics, nutrient chemistry, and
since 1992. Most recently, autonomous underwater vehicles
primary production provided a detailed understanding of
(AUVs) have been collecting high-resolution spatial data
upper water column dynamics in Monterey Bay. In the
in Monterey Bay since 2003 (Figure 17). MBARI realized
1990s the time series documented the impacts of El Niño
early on that the only realistic avenue for sustained time
and La Niña in central California and the equatorial Pacific,
series was to minimize the human factor and automate
leading to the realization that global climate fluctuations
collection, processing, and analysis of information. At the
cause dramatic changes in local ecosystems. And finally,
same time we recognized that shorter-term, high-visibility
following the 1997-98 El Niño, time-series data indicated
process studies can amplify results from the time series,
that Monterey Bay had cooled—only slightly—but still
and researchers have been attracted to the Monterey Bay
area to conduct such specific-issue studies partly
as a result of the data background that the time
series provide.
The main goal of the MBARI time-series program
is to understand phytoplankton primary produc-
tion in coastal and open-ocean upwelling systems
in relation to climate perturbations. Phytoplank-
ton occupy a central role in oceanic ecosystems,
as their nutrient uptake, growth, and sinking
mediate the biogeochemical fluxes of carbon
and other elements between the atmosphere,
surface ocean, and deep ocean. Phytoplankton
convert inorganic nutrients and sunlight to the
organic compounds that support life in the seas.
Because the atmosphere and oceans are physi-
cally coupled, climate exerts strong effects on the
marine biogeochemistry and food web mediated
by phytoplankton.
Figure 17: Ship tracks for the R/V Point Lobos and the R/V Zephyr for April 28,
A secondary but equally important goal of 2005 plotted on a bathymetric map of Monterey Bay, California, showing the
MBARI’s time-series effort is to provide basic locations of moorings M0, M1 and M2. The Point Lobos visits the three moor-
ings and an additional site at the mouth of Monterey Canyon where it collects
physical, chemical, and biological context for a
information on the long-term state of the ocean (see Figure 18) as well as cali-
number of other studies conducted in Monterey brate the moored instrumentation. The Zephyr currently follows an autonomous
Bay. Investigations including such diverse topics underwater vehicle (AUV) as it undulates along the path of the Point Lobos.
as energy flow in midwater and benthic food webs, Eventually the AUV will be launched from shore and carry out the survey indepen-
microbial recycling of nutrients, the formation dently or will be docked in the bay and commanded to operate on demand.

16 | Monterey Bay Aquarium Research Institute

 Sustaining and Mining the Time Series: Days to Decades

enough to significantly affect local ecosystem dynamics.

(Figure 18) Discussions with colleagues indicated that this
cooling was associated with large-scale changes in the Pacific
and may have been linked to longer term cycles associated
with the Pacific Decadal Oscillation (PDO). Recent data
support the idea that the eastern Pacific is well entrained
in La Vieja, the cool phase of this phenomenon which
lasts on order of 25 to 30 years. The analysis indicates that
the central California environment follows the Pacific
basin rather closely. For example, the Monterey Bay time
series covers the same period as the Hawaii Ocean Time-
Series. Monterey Bay and HOT are contrasting but linked
and complementary sites, as both ecosystems are regu-
lated by the north Pacific climate and gyral circulation.
Tiny picoplankton (ranging from .2 to 2 microns in size)
dominate at the nutrient-deficient HOT site, while larger
diatoms dominate the nutrient-rich Monterey Bay waters.
Parameters at these sites seem to oscillate out of phase
with each other. Productivity at HOT increases during El
Niño and the warm phase of the PDO, while it decreases
in Monterey Bay.
The sustained and broad-based data collections in Mon-
terey Bay undertaken by the time-series group at MBARI Figure 18: Anomaly plot of selected Monterey Bay time-series
and supported by the talented technicians in MBARI’s data from 1989 to 2004. The plots are produced by removing the
average annual cycle so that deviations from normal are displayed,
Marine Operations Division over the past 16 years have
with positive differences in red and negative differences in blue. For
produced a deep and continually updated data library from example in the top panel warm anomalies are associated with the
which fundamental insights emerge. Based on this data set, 1992-93 and 1997-98 El Niños. A long cool period followed after
state-of-the-art high-resolution (on a scale of 1 kilometer) 1998 and temperatures have been close to average since 2002.
coupled physical-biological models are being developed. Salinity, nitrate, chlorophyll, and primary production all show the
The long-term goal of the modeling effort is to be able to opposite pattern through 2002, which is expected. In 2003-04,
however, normal surface temperature coincided with low salinity
accurately predict changes in the quality and quantity of
and high chlorophyll and primary production, which remains unex-
primary production over time. With a renewed emphasis plained. Such relationships between variables determine linkages
on sensor development, the suite of measurements that and assess changing conditions in the northeast Pacific.
can be made autonomously in Monterey Bay can now be
expanded. As has been the case in the past, technological
developments will be exported from the natural labora- challenges. First, it will be necessary to bring together the
tory in Monterey Bay to other U.S. and international pro- massive volumes of data produced by the different observing
grams. As an example, carbon dioxide sensors developed methods so that scientists can see the connections between,
at MBARI are now populating moorings of the Tropical for example, anomalous oceanographic conditions and
Atmosphere Ocean (TAO) array in the equatorial Pacific. simultaneous increases in toxic algae. Probably the only
Bio-optical instrumentation vendors are also marketing way do this is by development of coupled data and modeling
their products using bio-fouling techniques pioneered systems for visualization, interpretation, and prediction.
under this program. These data-synthesis challenges extend beyond sensor or
platform development and will require significant improve-
The proposed expansion of coastal observatories for ments in our scientific concepts and more than likely the
improved stewardship of the coastal ocean is providing new development of a new generation of Earth scientists.

2004 Annual Report | 17