have occurred) is the NEP, and the decline in oxygen in the
dark bottle represents R. GPP is obtained by adding
the increase in oxygen in the transparent, “light” bottle to
the decline in oxygen in the “dark” bottle. The advantage
of this method is that it partitions the water column production between daytime net production by phytoplankton and respiration by the total community. However,
these measurements only capture P and R of a discreet
location in the water column at a specific time and do
not account for production associated with other organisms such as macroalgae and seagrasses, consumption by
zooplankton and fish, and benthic respiration that occurs
in the sediment. Typically, however, light and dark bottle
measurements are coupled with some sort of additional
measures or estimates of benthic respiration. Some weaknesses of this method are that errors associated with each
measurement are additive, which amplifies the uncertainty, and there is an assumption that the bottle measurements can be scaled up to represent the whole system.
While many of the earlier studies of estuarine metabolism
used this method, the development of high-frequency
data-logging dissolved oxygen (D.O.) and carbon dioxide
(CO 2 ) sensors has led to a reliance on other, less
sample-intensive methods.
Another way to measure metabolism in an estuary is by
measuring either diel (daily cycle) or seasonal changes in
D.O. Unlike the light and dark bottle method, these measurements represent whole water column changes and
interaction with the benthos, integrating all sources of production and respiration, as well as the added complications associated with the physical transport of water
masses and air-sea gas exchange. Whole water column
concentrations of D.O. increase over the course of the
day, as photosynthesis occurs, and decline at night, when
photosynthesis ceases and respiration dominates. While
changes in D.O. concentrations over a 24-h period can
be used to calculate daily NEP, this method is not without
its challenges, particularly in relation to the assumptions
that changes in the dissolved gas are biologically mediated
and entirely capture biological processes. In addition to
accounting for exchange between the water and the atmosphere and across different water masses, some investigators have observed a significant amount of benthic
respiration occurring in portions of the water and sediment
columns that are devoid of O 2 . Thus, the organisms respiring CO 2 are using other elements, such as sulfur, to conduct this respiration. This means that such respiration
would not be captured by changes in O 2 concentrations.
However, the sulfur-mediated respiration can be estimated
in O 2 equivalents (Kemp et al., 1997).
Nutrient budgets
Seasonal or annual nutrient concentrations and ratios have
also been used to successfully calculate TEM. For example, the ratio of dissolved inorganic nitrogen (DIN) to
dissolved inorganic phosphorous (DIP) reflects the
biological uptake and removal processes that use these
nutrients: primary production, respiration, and denitrification. One can calculate the TEM by comparing the
DIN/DIP to nutrient inputs. This technique is well illustrated by Nixon and Pilson (1984) in Narragansett Bay,
Rhode Island. By comparing the annual DIN and DIP
inputs to the mean annual DIN/DIP ratios in the bay, they
were able to calculate that the bay is net autotrophic and
production exceeded consumption by an amount of carbon equal to almost a quarter of that produced by bay phytoplankton. Similar budgeting using single nutrients can
also be used. For example, Smith and Hollibaugh (1997)
used DIP concentrations to estimate the NEP in Tomales
Bay, an estuary in Northern California. They measured
the differences between DIP inputs and outputs, where
the difference between the two was used as an index of
NEP. The DIP concentrations were converted to carbon
units using the 106:1 Redfield ratio of DIC/DIP. In contrast to Narragansett Bay, this US West Coast system
was found to be net heterotrophic. Terrestrial and marine
systems each contributed equally to supporting the excess
NEP in Tomales Bay.
In general, estuaries are thought to be net
heterotrophic. However, on a global scale, increasing
nutrient inputs from upstream may be shifting these systems to autotrophic (Kemp et al., 1997; Smith and
Hollibaugh, 1997). For example, excess nutrient inputs
from sewage was one of the main reasons that Narragansett Bay was autotrophic at the time of the study (Nixon
and Pilson, 1984), but recent upgrades to tertiary sewage
treatment may alter this regime. While Chesapeake Bay
also receives substantial sewage input, most of the nutrients entering the estuary are from agricultural runoff.
Kemp et al. (1997) measured NEP using five different
approaches and found quite similar results in the Chesapeake. They summed all carbon fluxes (including P and
R measurements made using light and dark bottles), developed nutrient mass balances, and compared nutrient ratios.
In addition to strong agreement among methods, they
observed clear spatial variability in the bay where the
upper bay was net heterotrophic, the mid bay in balance,
and the lower bay was net autotrophic. Their measured
values were driven largely by the biology, not physical
processes, and they concluded that the ratio of DIN to
total organic carbon was the controlling factor of NEP in
Chesapeake Bay, where the upper bay received more
organic carbon inputs and the lower bay relied more on
in situ production. Thus, measures of TEM can help to
provide useful insight into the relative importance of
contributions from river inputs versus production within
the estuary.
Summary
Total ecosystem metabolism (TEM), both as discrete measurements and as a theoretical concept, has an important
history in ecosystem ecology, particularly in estuaries.
302
ESTUARINE TOTAL ECOSYSTEM METABOLISM
dark bottle represents R. GPP is obtained by adding
the increase in oxygen in the transparent, “light” bottle to
the decline in oxygen in the “dark” bottle. The advantage
of this method is that it partitions the water column production between daytime net production by phytoplankton and respiration by the total community. However,
these measurements only capture P and R of a discreet
location in the water column at a specific time and do
not account for production associated with other organisms such as macroalgae and seagrasses, consumption by
zooplankton and fish, and benthic respiration that occurs
in the sediment. Typically, however, light and dark bottle
measurements are coupled with some sort of additional
measures or estimates of benthic respiration. Some weaknesses of this method are that errors associated with each
measurement are additive, which amplifies the uncertainty, and there is an assumption that the bottle measurements can be scaled up to represent the whole system.
While many of the earlier studies of estuarine metabolism
used this method, the development of high-frequency
data-logging dissolved oxygen (D.O.) and carbon dioxide
(CO 2 ) sensors has led to a reliance on other, less
sample-intensive methods.
Another way to measure metabolism in an estuary is by
measuring either diel (daily cycle) or seasonal changes in
D.O. Unlike the light and dark bottle method, these measurements represent whole water column changes and
interaction with the benthos, integrating all sources of production and respiration, as well as the added complications associated with the physical transport of water
masses and air-sea gas exchange. Whole water column
concentrations of D.O. increase over the course of the
day, as photosynthesis occurs, and decline at night, when
photosynthesis ceases and respiration dominates. While
changes in D.O. concentrations over a 24-h period can
be used to calculate daily NEP, this method is not without
its challenges, particularly in relation to the assumptions
that changes in the dissolved gas are biologically mediated
and entirely capture biological processes. In addition to
accounting for exchange between the water and the atmosphere and across different water masses, some investigators have observed a significant amount of benthic
respiration occurring in portions of the water and sediment
columns that are devoid of O 2 . Thus, the organisms respiring CO 2 are using other elements, such as sulfur, to conduct this respiration. This means that such respiration
would not be captured by changes in O 2 concentrations.
However, the sulfur-mediated respiration can be estimated
in O 2 equivalents (Kemp et al., 1997).
Nutrient budgets
Seasonal or annual nutrient concentrations and ratios have
also been used to successfully calculate TEM. For example, the ratio of dissolved inorganic nitrogen (DIN) to
dissolved inorganic phosphorous (DIP) reflects the
biological uptake and removal processes that use these
nutrients: primary production, respiration, and denitrification. One can calculate the TEM by comparing the
DIN/DIP to nutrient inputs. This technique is well illustrated by Nixon and Pilson (1984) in Narragansett Bay,
Rhode Island. By comparing the annual DIN and DIP
inputs to the mean annual DIN/DIP ratios in the bay, they
were able to calculate that the bay is net autotrophic and
production exceeded consumption by an amount of carbon equal to almost a quarter of that produced by bay phytoplankton. Similar budgeting using single nutrients can
also be used. For example, Smith and Hollibaugh (1997)
used DIP concentrations to estimate the NEP in Tomales
Bay, an estuary in Northern California. They measured
the differences between DIP inputs and outputs, where
the difference between the two was used as an index of
NEP. The DIP concentrations were converted to carbon
units using the 106:1 Redfield ratio of DIC/DIP. In contrast to Narragansett Bay, this US West Coast system
was found to be net heterotrophic. Terrestrial and marine
systems each contributed equally to supporting the excess
NEP in Tomales Bay.
In general, estuaries are thought to be net
heterotrophic. However, on a global scale, increasing
nutrient inputs from upstream may be shifting these systems to autotrophic (Kemp et al., 1997; Smith and
Hollibaugh, 1997). For example, excess nutrient inputs
from sewage was one of the main reasons that Narragansett Bay was autotrophic at the time of the study (Nixon
and Pilson, 1984), but recent upgrades to tertiary sewage
treatment may alter this regime. While Chesapeake Bay
also receives substantial sewage input, most of the nutrients entering the estuary are from agricultural runoff.
Kemp et al. (1997) measured NEP using five different
approaches and found quite similar results in the Chesapeake. They summed all carbon fluxes (including P and
R measurements made using light and dark bottles), developed nutrient mass balances, and compared nutrient ratios.
In addition to strong agreement among methods, they
observed clear spatial variability in the bay where the
upper bay was net heterotrophic, the mid bay in balance,
and the lower bay was net autotrophic. Their measured
values were driven largely by the biology, not physical
processes, and they concluded that the ratio of DIN to
total organic carbon was the controlling factor of NEP in
Chesapeake Bay, where the upper bay received more
organic carbon inputs and the lower bay relied more on
in situ production. Thus, measures of TEM can help to
provide useful insight into the relative importance of
contributions from river inputs versus production within
the estuary.
Summary
Total ecosystem metabolism (TEM), both as discrete measurements and as a theoretical concept, has an important
history in ecosystem ecology, particularly in estuaries.
302
ESTUARINE TOTAL ECOSYSTEM METABOLISM