5. The Measurement of Primary Production in Aquatic Ecosystems
77
ward measuring the net community production
(NCP) of the plankton community. Note, however,
that in many ecosystems at many points in time,
NCP can be a negative number; although rarer, negative rates of NPP also occur. The 14C method,
however, always shows a positive rate of uptake.
Note also that long incubations generally are not
desirable since they aggravate what are called "bottle effects" (see below).
Some researchers have attempted to correct for
the algal respiration that occurs during 14C incubation (Wofsy 1983; Cole et al. 1992). However,
the relationship between photosynthesis and respiration in phytoplankton is not easily predictable or
measured. Respiration rates for phytoplankton vary
from 5 to 50% of the rate of light-saturated photosynthesis (Raven and BeardalI1981). This makes
it difficult to determine an empirical relationship
between 14C-Iabeled DIC assimilation and rates of
GPP or NPP. Since the 14C technique does not truly
measure GPP (or NPP, NEP, or NCP), results are
most commonly reported as just "primary production" or " 14 C production." Despite problems with
interpretation, using such data in a comparative
manner across ecosystems has proven quite useful.
For instance, Nixon et al. (1996) have demonstrated
that 14C production is well correlated with nitrogen
inputs to estuarine and marine ecosystems across a
wide range of inputs. However, one cannot use estimates of 14C production to prepare carbon or nutrient budgets without making assumptions about
how these rates relate to GPP and NPP (Howarth
et al. 1996).
Problems and Challenges
with Light and Dark Bottle
and Carbon-14 Techniques
Photosynthesis is generally a nonlinear function of
light intensity (irradiance). Production rates increase linearly with irradiance at low light levels.
At some moderately high light level, photosynthetic rates saturate and then actually decrease at
higher levels of irradiance due to photoinhibition
processes. This relationship between light and photosynthesis is commonly called the "P versus I
curve" (for photosynthesis vs. irradiance), and the
relationship varies among different ecosystems,
presumably because phytoplankton are well
adapted or acclimated to the in situ light levels of
the ecosystem where they occur (Gallegos and Platt
1981; Falkowski and LaRoche 1991). Therefore,
when measuring production by either the light and
dark bottle method or the 14C technique, one ideally
measures production rates over a range of light intensities that encompasses the spectrum seen by the
plants. By using short-term in situ light measurements over depth in the ecosystem and the empirically determined P versus I curves, one can scale
the rate measurements and estimate the in situ rate
of production per square meter (Gallegos and Platt
1981). The use of an incubation chamber with a
well-characterized gradient of light intensities simplifies this approach (Fee 1973, 1980), but P versus
I curves can also be determined from incubating
bottles in situ in the ecosystem if bottles are incubated at different depths (light intensity decreases
with depth) and the irradiance is measured at each
depth (Bower et al. 1987). Alternatively, one can
incubate the samples for the entire diel period and
trade off an increase in the accuracy of the light
field for a longer incubation (with the concomitant
increase in the potential for bottle artifacts).
In water masses where phytoplankton are continually buoyed in well-lit surface waters, the 14C
technique works fairly well when rates are extrapolated by use of a P versus I curve, as shown by
good agreement with rates so derived and rates determined by whole-lake additions of 14C-labeled
DIC (Hesslein et al. 1980; Bower et al. 1987).
However, simple integration of 14C uptake rates
measured in a series of bottles incubated in situ
over depth can underestimate rates compared with
whole-lake 14C-Iabeled DIC additions (Bower et al.
1987). In some ecosystems, phytoplankton get
mixed in the water column to depths below the
compensation depth (the depth at which photosynthesis equals respiration by phytoplankton). This
commonly occurs in deep, turbid rivers and, in the
ocean, during deep winter mixing at high latitudes.
To estimate 14C production in such systems, one
must be able to estimate what percentage of time
phytoplankton spend at different light levels while
being mixed through the water column (Lewis
1988; Cole et al. 1991, 1992). In such systems, 14C
production can be quite difficult to interpret, as
phytoplankton respiration is such a great percentage of GPP due to the high percentage of time spent
77
ward measuring the net community production
(NCP) of the plankton community. Note, however,
that in many ecosystems at many points in time,
NCP can be a negative number; although rarer, negative rates of NPP also occur. The 14C method,
however, always shows a positive rate of uptake.
Note also that long incubations generally are not
desirable since they aggravate what are called "bottle effects" (see below).
Some researchers have attempted to correct for
the algal respiration that occurs during 14C incubation (Wofsy 1983; Cole et al. 1992). However,
the relationship between photosynthesis and respiration in phytoplankton is not easily predictable or
measured. Respiration rates for phytoplankton vary
from 5 to 50% of the rate of light-saturated photosynthesis (Raven and BeardalI1981). This makes
it difficult to determine an empirical relationship
between 14C-Iabeled DIC assimilation and rates of
GPP or NPP. Since the 14C technique does not truly
measure GPP (or NPP, NEP, or NCP), results are
most commonly reported as just "primary production" or " 14 C production." Despite problems with
interpretation, using such data in a comparative
manner across ecosystems has proven quite useful.
For instance, Nixon et al. (1996) have demonstrated
that 14C production is well correlated with nitrogen
inputs to estuarine and marine ecosystems across a
wide range of inputs. However, one cannot use estimates of 14C production to prepare carbon or nutrient budgets without making assumptions about
how these rates relate to GPP and NPP (Howarth
et al. 1996).
Problems and Challenges
with Light and Dark Bottle
and Carbon-14 Techniques
Photosynthesis is generally a nonlinear function of
light intensity (irradiance). Production rates increase linearly with irradiance at low light levels.
At some moderately high light level, photosynthetic rates saturate and then actually decrease at
higher levels of irradiance due to photoinhibition
processes. This relationship between light and photosynthesis is commonly called the "P versus I
curve" (for photosynthesis vs. irradiance), and the
relationship varies among different ecosystems,
presumably because phytoplankton are well
adapted or acclimated to the in situ light levels of
the ecosystem where they occur (Gallegos and Platt
1981; Falkowski and LaRoche 1991). Therefore,
when measuring production by either the light and
dark bottle method or the 14C technique, one ideally
measures production rates over a range of light intensities that encompasses the spectrum seen by the
plants. By using short-term in situ light measurements over depth in the ecosystem and the empirically determined P versus I curves, one can scale
the rate measurements and estimate the in situ rate
of production per square meter (Gallegos and Platt
1981). The use of an incubation chamber with a
well-characterized gradient of light intensities simplifies this approach (Fee 1973, 1980), but P versus
I curves can also be determined from incubating
bottles in situ in the ecosystem if bottles are incubated at different depths (light intensity decreases
with depth) and the irradiance is measured at each
depth (Bower et al. 1987). Alternatively, one can
incubate the samples for the entire diel period and
trade off an increase in the accuracy of the light
field for a longer incubation (with the concomitant
increase in the potential for bottle artifacts).
In water masses where phytoplankton are continually buoyed in well-lit surface waters, the 14C
technique works fairly well when rates are extrapolated by use of a P versus I curve, as shown by
good agreement with rates so derived and rates determined by whole-lake additions of 14C-labeled
DIC (Hesslein et al. 1980; Bower et al. 1987).
However, simple integration of 14C uptake rates
measured in a series of bottles incubated in situ
over depth can underestimate rates compared with
whole-lake 14C-Iabeled DIC additions (Bower et al.
1987). In some ecosystems, phytoplankton get
mixed in the water column to depths below the
compensation depth (the depth at which photosynthesis equals respiration by phytoplankton). This
commonly occurs in deep, turbid rivers and, in the
ocean, during deep winter mixing at high latitudes.
To estimate 14C production in such systems, one
must be able to estimate what percentage of time
phytoplankton spend at different light levels while
being mixed through the water column (Lewis
1988; Cole et al. 1991, 1992). In such systems, 14C
production can be quite difficult to interpret, as
phytoplankton respiration is such a great percentage of GPP due to the high percentage of time spent