Methods for

1
Methods for Estimating New Primary Production In
Upwelling Systems James J. Bisagni University
of Massachusetts, Dartmouth Physics Department
School for Marine Science Technology New
Bedford, Massachusetts, USA
2
Talk Outline I. List Some Goals II. Provide
Working Definitions III. Summary of Model
Types IV. Model Descriptions V. Discussion
3
Tentative Project Goals I. Describe the size of
New Primary Production (NPP) in the Peru/Chile
upwelling system II. Describe the mean seasonal
cycle of NPP in the Peru/Chile upwelling
system III. Describe interannual variability of
NPP in the Peru/Chile upwelling system IV.
Achieve goals I-III with a model-based approach,
using both satellite and in-situ data
4
Total Primary Production Total primary
production in the ocean may be divided into new
and regenerated production based on the source
of the nitrogen which is utilized. (Dugdale
Goering, 1967 Eppley Peterson, 1979).
Allocthonous nitrogen or nitrate (NO3) is input
into the euphhotic zone from horizontal and/or
vertical advection and diffusion. Autocthonous
nitrogen or ammonium (NH4) is input in the
euphotic zone from metabolic recycling caused by
biota within the water column and sediments.
5
Importance of New Primary Production It is
reasonable to state that in the absence of an
allocthonous nitrogen supply, any given marine
ecosystem will eventually become non-sustainable
due to export of nitrogen through sinking of
biogenic material and harvesting activities such
as fishing, and predation from migratory pelagic
species. (Platt et al., 1989) Thus, it is clear
from the standpoints of the global oceans
ability to sequester atmospheric CO2, along with
a local regions ability to maintain a
sustainable ecosystem, that new primary
production, rather than total primary production,
is the key quantity.
6
Measurements of New Production Assimilation of
15N-labeled compounds allows instantaneous
estimates of uptake rates of the available
nitrogen sources by phytoplankton (Dugdale and
Goering, 1967) and an estimate of the so-called
f-ratio of new production to total production,
where
(Eppley and Peterson, 1979)
However, estimates of regional or global new
production must be done using other techniques
such as using the relationship between new and
total production and remotely-sensed data in
order to discern how the presence of nitrogen
species within the euphotic zone varies in space
and time.
7
New Production Models 1) Nitrate Uptake
(Shift-Up) Models Utilize the inverse
relationship between temperature and nitrate and
the physiological response of phytoplankton
within the euphotic zone. gt Largely
kinetics-based Examples include Dugdale et
al., 1989, Northwest Africa Kudela Dugdale,
1996, California Dugdale et al., 1997,
California 2) Nitrate Bulk Models Utilize the
inverse relationship between temperature and
nitrate and a nitrate budget to account for the
amount of nitrate within and entering the
euphotic zone. gt Largely physics-based Examp
les include Waldron Probyn, 1992,
Benguela Townsend, 1998, Gulf of Maine Bisagni,
in-press, Gulf of Maine
8
Nitrate Uptake (Shift-Up) Model
(After Dugdale et al., 1989)
9
AVHRR (Only) Model (Left-Half) For Each Pixel
Satellite SST Provides NO3
10
AVHRR (Only) Model (Left-Half) For Each Pixel
Satellite SST Provides NO3
Satellite DSST Provides Time-Base
11
AVHRR (Only) Model (Left-Half) For Each Pixel
Satellite SST Provides NO3
Satellite DSST Provides Time-Base
Maximum measured specific NO3 uptake (Vmax NO3)
at timet assuming shift-up
Shift-Up Vmax NO3(t) VNO3 (i) A(t) x t
Measured acceleration of NO3 uptake, d/dt(VNO3)
12
AVHRR (Only) Model (Left-Half) For Each Pixel
Satellite SST Provides NO3
Satellite DSST Provides Time-Base
Maximum measured specific NO3 uptake (Vmax NO3)
at timet assuming shift-up
Predicted VNO3 at timet assuming shift-up
Michaelis-Menton kinetics
Shift-Up Vmax NO3(t) VNO3 (i) A(t) x
t Michaelis-Menton Kinetics VNO3(t) Vmax
NO3(t) x NO3/(Ks NO3)
Measured half-saturation constant
13
AVHRR (Only) Model (Left-Half) For Each Pixel
Satellite SST Provides NO3
Satellite DSST Provides Time-Base
Maximum measured specific NO3 uptake (Vmax NO3)
at timet assuming shift-up
Variation in VNO3 versus irradiance depth of
euphotic zone measurements (or from ocean color)
Predicted VNO3 at timet assuming shift-up
Michaelis-Menton kinetics
Shift-Up Vmax NO3(t) VNO3 (i) A(t) x
t Michaelis-Menton Kinetics VNO3(t) Vmax
NO3(t) x NO3/(Ks NO3)
14
AVHRR (Only) Model (Left-Half) For Each Pixel
Satellite SST Provides NO3
Satellite DSST Provides Time-Base
Maximum measured specific NO3 uptake (Vmax NO3)
at timet assuming shift-up
Variation in VNO3 versus irradiance depth of
euphotic zone measurements (or from ocean color)
Predicted VNO3 at timet assuming shift-up
Michaelis-Menton kinetics
Integrate VNO3 over the euphotic zone and
multiply by PON to yield NPP
Shift-Up Vmax NO3(t) VNO3 (i) A(t) x
t Michaelis-Menton Kinetics VNO3(t) Vmax
NO3(t) x NO3/(Ks NO3)
15
AVHRR (Only) Model (Left-Half) For Each Pixel
Satellite SST Provides NO3
Satellite DSST Provides Time-Base
Maximum measured specific NO3 uptake (Vmax NO3)
at timet assuming shift-up
Variation in VNO3 versus irradiance depth of
euphotic zone measurements (or from ocean color)
Predicted VNO3 at timet assuming shift-up
Michaelis-Menton kinetics
Integrate VNO3 over the euphotic zone and
multiply by PON to yield NPP
Compute f-ratio
Shift-Up Vmax NO3(t) VNO3 (i) A(t) x
t Michaelis-Menton Kinetics VNO3(t) Vmax
NO3(t) x NO3/(Ks NO3)
16
Nitrate Bulk Models
NPP is estimated using a bulk quantity termed
potential new production (PNP) as a proxy
upper limit for NPP where PNP has been defined
in a variety of ways For each Benguela
upwelling event
z 0
Stored NO3
z -Ze
Then sum over n upwelling events
(Waldron Probyn, 1992)
17
In the Gulf of Maine
gt PNP NO3 Flux x Redfield ratio
z 0
z -Ze
NO3 Flux
(Townsend, 1998)
18
In the Gulf of Maine
gt PNP d/dtStored NO3 - NO3 Flux x
Redfield ratio
z 0
Stored NO3
z -Ze
NO3 Flux
(Bisagni, in press)
19
Model Differences
Nitrate Uptake (Shift-Up) Models In addition to
standard hydrographic measurements of
temperature, NO3, and PON, such kinetics-based
NPP models require rate measurements of A(t),
KS, and KE, the half saturation constant for NO3
uptake as a function of irradiance (depth).
Moreover, measured 15N incubations, if available
would allow verification of modeled VNO3(t)
values. Models are sensitive to slope of
temperature- NO3 regression and the applied
constant heating rate (may not be constant!).
Nitrate Bulk Models Utilize standard
hydrographic measurements of temperature and NO3
but require estimates of KZ and upwelling
velocity to measure the proxy PNP. Models are
sensitive to slope of temperature- NO3
regression, KZ and upwelling velocity.
20
Summary Conclusions
Models exist which are able to estimate
NPP Nitrate uptake (shift-up) models are able to
make pixel-by-pixel estimates of NPP, but make
some strong assumptions (heating rate) and
require many in-situ rate measurements. Nitrate
Bulk Models are able to make pixel-by-pixel
estimates of the upper limit of NPP through
estimation of PNP as a proxy for NPP, but require
some physical oceanographic quantities (KZ and
w). Choosing a model depends largely on the
questions being asked and available data.
21
Summary Conclusions
Models exist which are able to estimate
NPP Nitrate uptake (shift-up) models are able to
make pixel-by-pixel estimates of NPP, but make
some strong assumptions (heating rate) and
require many in-situ rate measurements. Nitrate
Bulk Models are able to make pixel-by-pixel
estimates of the upper limit of NPP through
estimation of PNP as a proxy for NPP, but require
some physical oceanographic quantities (KZ and
w). Choosing a model depends largely on the
questions being asked and available data. What
are the available data?