Ryther 1963

1
Impact of Pacific Climate Variability on Ocean
Circulation, Marine Ecosystems Living
Resources
Francisco Chavez MBARI
Lead PI Dick
Barber, Duke University
Co PI Fei Chai,
University of Maine
Co PI Yi Chao,
JPL of Calif. Inst. of Tech.
Co PI David Foley,
NOAA SW Fish. Center
Collaborating PI

Supported by Earth Science Research
Program Carbon Cycle and Ecosystems Focus
Area Biodiversity and Ecological Forecasting
Section.
2
Usual justification of ocean ecosystem
development - better management of living
resources - helping to achieve sustainability,
and - optimizing societal investment in fishery
infrastructure But
3
For example, a proposal funded by NSF in May
1975 said The goal of the Coastal Upwelling
Ecosystems Analysis Program (CUEA) is to
understand the coastal upwelling ecosystem well
enough to predict its response far enough in
advance to be useful to mankind.
4
What was CUEA? A big, multi-institutional,
multi-disciplinary, multi-agency, long-term
project Big 14 mil from 1972 to 1980 24
PIs, 14 institutions, four agencies
(NSF/NOAA/ONR/NASA) Peru field work in 1976/1977
with 4 US ships, an NCAR plane, a NASA
radiometer, several shore-based met
stations plus a lot support from Peru in people,
logistics and diplomacy CUEA was a successful
interdisciplinary basic research project
but did not deliver any useful to mankind
product. Why?
5
three sources of systemic model error theory -
understanding, equations resolution -
time/space realism initialization - initial
state realism
6
In the 70s there were Two specific biological
theory deficiencies food web structure
Fe, Two specific physical theory deficiencies
remote forcing decadal variability Plus 3
fatal technical constraints computing power,
observing power (satellites) information
handling infrastructure
Tech constraints unconceivable, at the time
Absolutely fatal Orders of magnitude
changes required to fix
7
In 2008 we think this is the status of ocean
forecast modeling Two specific biological
theory deficiencies FIXED food web
structure Fe Two specific physical theory
deficiencies ½ FIXED remote forcing
decadal variability Plus 3 fatal technical
constraints ALL FIXED computing power,

observing power (satellites
and moorings) information infrastructure
8
computational power revolution. The potential
consequences of Moores Law for operational
forecast modeling are impressive increased
time and space resolution, new concepts (i.e.,
assimilation), scale convergence, scale
expansion, spatial nesting, reanalysis, model
complexity, near real-time modeling,
retrospective modeling, etc., etc., etc.
9
(No Transcript)
10
Model complexity re need for diatom and
picophytoplankton response to perturbations
pico-micro steady-state rates shift-up, but with
small biomass change
Diatoms bloom, crash and export rates and
biomass change (first , then --)
picophytoplankton
Chl (mg m-3)
diatoms
-1 0 1 2 3 4 5
6 7 8 9 10 11
Days since first Fe Addition
From Barber and Hiscock, GBC, 2006
11
Remote Forcing El Ninos influence California
Current System
(J. Ryan, MBARI)
12
(No Transcript)
13
Sardine Landings
Japan
California
Peru
South Africa
1925 1950 1975
Chavez et al. Science (2003)
1925 1950 1975
2000
14
Scale convergence of eddy kinetic energy of
model and observations in a coastal upwelling
system
Eddy kinetic energy (cm2s-2)
Drifter
Model
Resolution (km)
15
Different physical and biological response to the
same initial physical perturbation
16
A Nine Month Forecast of Peru Coastal Chl
that is PDG for the first 5 months!
17
The same nine month forecast embedded in a longer
record
18
European Centre for Medium-Range Weather
Forecasts (ECMWF)
19
Conclusion The deficiencies in theory and
technology that prevented useful operational
marine ecosystem forecasts for resource
management in the past are now being
surmounted. Its time to test this new, and still
primitive, forecasting capability in
ecosystem-based management of living
resources. Peru is the best place to start.