Williams_Lecture 2
DOM Stability, Lability and Resistance
1/12
STRUGGLING WITH THE ILL-DEFINED NOTIONS OF ORGANIC STABILITY,
RESISTANCE & LABILITY
BACKGROUND
THE KIRCHMAN MODEL
Conceptual model
of DOM depth
distribution
proposed by
Kirchman et al.
1993
We learn more by
asking why an
organic compound
is not broken down
than why it is.
The terms labile, semi-labile and resistant are purely empirical and provide no insight into the
mechanism giving rise to these properties
. It is far easier to address stability than lability.
The figure, taken from
Linton and Watson (2000),
shows the flows and
feedbacks; negative
feedbacks, which are
essential to the stabilisation
of an ecosystem are shown
as dotted lines,
characteristically are
associated with the
accumulation of material.
Williams_Lecture 2
DOM Stability, Lability and Resistance
2/12
PART I: A CONCEPTUAL BASIS FOR PERSISTENCE
It is useful to recognise two bases for persistence:
A) Stability – simply will not break down: in principle this is an absolute property and an innate
property of the molecule – sometimes also referred to as recalcitrance.
B) Resistance – not inclined to breakdown. This has a variety of potential causes
Inbuilt resistance, a property of the molecule
Circumstantial resistance, an ephemeral property set by the prevailing circumstances
A) Molecular stability:
1) Massive molecular monsters
For stearic reasons certain molecular structures are resistant to microbial decomposition. The classical
view of resistant DOM was that they were large, complex structures, e.g.
It’s bit like trying to eat
a bowl of spaghetti,
when you are only
allowed to nibble away
at the ends
Ron Benner’s work effectively blew that away and we now need more subtle concepts
2) The concept of xenophoric structures
In simple terms, there are a variety of substitutions into linear carbon chain that enhance resistance to
microbial decomposition – notable the tertiary carbon structure and the elements Cl, Br, nitrate and
sulphate, interestingly all molecules present in seawater.
C
*
C
C
* Cl, Br, NO3, SO4, CH3
Note:
1) It is not clear whether this
confers stability or simplicity
enhances resistance
2) There is an interesting
asymmetry here between
chemical and microbial
reactivity – we’ll return to this
later.
Williams_Lecture 2
DOM Stability, Lability and Resistance
3/12
It is axiomatic that all biologically produced molecules are susceptible to biological decomposition, so
how are xenophoric structures produced
Mmmm - – now there’s an
interesting problem to
ponder over
B) Resistance
We identified two forms of resistance:
1) Inbuilt resistance, a property of the molecule
2) Circumstantial resistance, a property set by the prevailing circumstances
1) Inbuilt Resistance
We are aware that certain molecules tend to persist, where others are prone to rapid decomposition.
Thermodynamic stability - This is associated with the energy released by the
decomposition of the molecule – the Gibbs free energy. As the Gibbs free
energy is determined by the reaction, the reactants, other than the molecule,
play a role, so a grey area exists between this and circumstantial resistance.
i.e. thermodynamics
cannot predict
kinetics – interesting!
However, it’s axiomatic that whereas thermodynamics can predict whether or not a molecule will
persist in a particular circumstance it can give no insight into the rate of decomposition
Glucose and its polymers are a nice example to consider. Glucose and cellulose will have much the
same Gibbs free energy (see left column below) so their decomposition will release a similar amount of
energy. However, cellulose is much more resistant than glucose (middle column). This, in isolation,
could lead to the conclusion that molecular weight is a main determinant of resistance. BUT as soon as
we introduce the second polymer of glucose – starch – this explanation collapses. Over everything but
very short timescales, starch is every bit as readily decomposable as glucose. The figure (right and
middle columns) shows the solubility generally correlated with reactivity and in the case of glucose and
its polymers it may be a major determinant – but it cannot be a universal explanation – indeed one
would be unwise to search for single universal controls on reactivity.
Williams_Lecture 2
DOM Stability, Lability and Resistance
High
4/12
High
Gibbs free energy
High Biological
Reactivity
Solubility
POLYNUCLEAR
AROMATIC
GLUCOSE
GLUCOSE
SUBSTITUTED
AROMATIC
ALIPHATIC
ALIPHATIC
HUMIC ACID
STARCH
STARCH
ALIPHATIC
SUBSTITUTED
AROMATIC
SUBSTITUTED
AROMATIC
CELLULOSE
HUMIC ACID
HUMIC ACID
POLYNUCLEAR
AROMATIC
POLYNUCLEAR
AROMATIC
CELLULOSE
Low Biological
Reactivity
Solubility
CELLULOSE/
GLUCOSE/
STARCH
Gibbs free energy
Low
Low
2) Circumstantial Resistance
It is becoming realised that the reactivity of molecules may vary with the environmental circumstances.
Let’s see an example (again it’s glucose):
0
2
4
6
8
10 12 14 16 18 20 22
60
Glucose Conc ( M)
Observed total monosaccharide conc.
40
Glucose resists
decomposition
30
Glucose undergoes
decomposition
20
4
6
8
10 12 14 16 18 20 22
40
A
Cumulative glucose addition
50
2
0
Glucose Conc ( M)
Exp Day
Glucose added
B
Cumulative glucose addition
30
Observed total monosaccharide conc.
20
Glucose undergoes
decomposition
10
10
0
0
0
2
4
6
8
10
12
Sampling Day
14
16
18
20
22
0
2
4
6
8
10 12
Sampling Day
14
16
18
20
22
There are four circumstantial mechanisms that may effect control on reactivity
i) Stoichiometric consequences
ii) Trophic dynamics
iii) Kinetic controls
iv) Thermodynamic controls
Initially,
glucose, a labile
molecule, when
added on day-5
accumulates,
then is removed
But, in a replicate
system, where the
addition of glucose
is delayed until
day 14, it id
removed
immediately
i) Stoichiometric consequences)
Box 1. Stoichiometric Argument
The problem we wish to analyse is: how do the nitrogen requirements of the cell affect the uptake of organic substrates
The bacterial C/N quota is low (~ 4.5) and has to be met
Consider the quotient of the Cell Quota (QC/N) and the Carbon Growth Yield (YC )
Then, (QC/N)/(YC ) gives the maximum C/N of a satisfying substrate
Assume QC/N = 4.5
and YC = 0.25 (below YC = 0.5, the exact value used is not critical)
Then maximum C/N substrate = 4.5/0.25 = 18
Thus, substrates with higher C/N ratios can only be assimilated, with the associated assimilation of inorganic nitrogen
Williams_Lecture 2
DOM Stability, Lability and Resistance
5/12
Thus, organic compounds with C/N ratios above 18, will leave the cell N-deficient, whereas C/N ratios
below 18 out to result in inorganic nitrogen excretion.
C/N ratios of nitrogen-containing biochemicals vary from 2-9 (e.g. glycine and tyrosine) then infinity
(e.g. glucose). Thus, no single biochemical has a C/N ratio in the range >9 to infinity. If no organic
nitrogen containing compounds are available then the micro-organism needs to resort to assimilating
inorganic nitrogen, typically ammonia.
In the euphotic zone, assimilation inorganic nitrogen brings the bacteria into competition with the
photosynthetic organisms. Thus, stoichiometric control can lead on to a form of trophic control – so
let’s have a look at that
ii) Trophic dynamics
A simple trophic network of the plankton, comprising
two size groups of algae (the autotrophic
nanoflagellates and the diatoms), two of protozoa: the
herbivorous microzooplankton (e.g. the ciliates) and
the heterotrophic nanoflagellates and at the top of the
food web the mesozooplankton (e.g. the copepods)
and the bacteria at the bottom of the heterotrophs. All
but the mesozooplankton organisms have similar
growth rates so are tightly locked together.
The bacteria are predated upon by the heterotrophic
nanoflagellates and when using non-nitrogenous
organic substrates they compete with the diatoms but
mainly the autotrophic nanoflagellates for inorganic
nitrogen. Thus, when assimilating non-nitrogenous
organic compounds they are fighting on two fronts.
Top down control, by the mesozooplankton can be
seen to determine the success of their attempt to
assimilate non-nitrogenous compounds.
Inorganic Nutrients
This has been an area explored by Thingstadt. The diagram and the following is a summary of his
atguments.
Diatoms
Mesozooplankton
Autotrophic flagellates
Herbivorous
microzooplankton
Bacteria
Heterotrophic
nanoflagellates
Nitrogen containing
organic
Williams_Lecture 2
DOM Stability, Lability and Resistance
6/12
Diatoms
Mesozooplankton
Autotrophic flagellates
Herbivorous
microzooplankton
Bacteria
Heterotrophic
nanoflagellates
Inorganic Nutrients
B
Inorganic Nutrients
A
Non-nitrogen
containing organic
Diatoms
Mesozooplankton
Autotrophic flagellates
Herbivorous
microzooplankton
Bacteria
Heterotrophic
nanoflagellates
Non-nitrogen
containing organic
A flow diagram of the passage of material through the microbial network during two phases. The solid
arrows indicate the scale of flow, the open arrows indicate major sites of top down control.
The sequence A to B is hypothesised to be the basis of the phases of glucose non-lability and lability
shown previously
60
Glucose resists
decomposition
30
Glucose Conc ( M)
Glucose Conc ( M)
Observed total monosaccharide conc.
40
40
A
Cumulative glucose addition
50
Glucose undergoes
decomposition
20
B
Cumulative glucose addition
30
Observed total monosaccharide conc.
20
Glucose undergoes
decomposition
10
10
0
0
0
2
4
6
8
10
12
Sampling Day
14
16
18
20
22
The theory goes; initially the mesozooplankton, slow growers are
sparse, thus there is little control on
the fast growing microzooplankton, so they flourish, crop
down the bacteria, so no removal of
glucose.
Given ti
0
2
4
6
8
10 12
Sampling Day
Given time, the mesozooplankton increase in
numbers, crop down the
micro-zooplankton and so
allow the bacteria to
floueish, who remove
the glucose – that’s the
theory anyway
14
16
18
20
22
Williams_Lecture 2
DOM Stability, Lability and Resistance
7/12
Kinetic and Thermodynamic Constraints to Microbial Utilisation
Two other causes for persistence have been raised:
iii) Kinetic mechanisms: on occasions: that the concentrations in seawater become so low that their
rate of arrival at the cell surface is to slow to sustain some critical rate of growth
iv) Thermodynamic mechanisms: that taking up the substrate is not economic as it requires more
energy to take it up than is gained from its subsequent respiration (
These propositions can be analysed physico-chemically, and
neither turn out to have any likely basis. They are discussed here
mainly to eliminate them.
Maybe, it the dilute
organic environment in
the sea, molecules
simple don’t arrive fast
enough at the cell
surface.
iii) Kinetic constraint to acquisition
The calculation (given in Box 2) asks the question at what concentration does the arrival of molecules
limit growth.
Box 2. Kinetic constraint to acquisition
The problem we wish to analyse is: do molecules arrive at a sufficient rate to sustain
metabolism
Consider bacteria at a concentration of 12mg (i.e. 1 m) /dm3, growing at 1 div/day
with a growth efficiency of 25%.
This would require 1 x 4 x 10-6 x 6x1023 molecules /day = 24 x1015 molecules/day =
2.78 x 1013 per second
The bacterial surface area is approximately 1dm2/dm3
The calculated minimum arrival must be 3 x1013 molecules/sec.cm2
Such a collision rate would be sustained by 1010 molecules/dm3; i.e. ~ 10-14M
Assume the successful collision was 1 in 103 then the minimum concentration of the
food substrate would need to be 10-11M, i.e. 10510 times greater than the calculated
thermodynamic constraint (see Box 3) but not a long way from the 10-8 M observed
The calculation, other than the required arrival rate at the surface, is fairly routine. The major
uncertainly other than the collision rates in solution is the fraction of successful collisions. I have taken
a cautious value of 1 in 1000. Given this concentrations would need to fall to 10-11 molar before they
would limit growth.
Conclusion: presently our analytical methods for hexoses and free amino acids are probable limited to
concentrations no lower that 10-8 molar, this in principal if we can detect a molecule to be present it
will be arriving at a sufficient frequency to sustain microbial growth
Williams_Lecture 2
DOM Stability, Lability and Resistance
8/12
iv) Thermodynamic constraint to acquisition
Work has to be done to take up a molecule against a diffusion gradient
and the energy for this work has to be acquired from the respiration of
the molecule. This is a different issue to the work for the transport of
the molecule across the membrane. If it takes more work to acquire it
than is gained from its metabolism,then the organism perhaps would
not take it up.
Its like the
question- why is
there is still gold in
the Welsh hills..
The calculation has as much, if not more, entertainment value as scientific.
Box 3. Thermodynamic Constraints to Acquisition
The problem: at what concentration does it become uneconomic to take up organic compounds; i.e. when
does the energy required to concentrate the material exceed that obtained from its respiration
This can be calculated by reversing the Heat of Dilution Calculation
Heat of Dilution:- F = RT ln (C1/C2); when -F = -G, then there is no surplus energy
-G for the oxidation of glucose is 2870 kJ/mole
Then: 2870 x 103 = 8.3 x 290 x 2.303 x log(C1/C2)= 5.5 x 103 x log(C1/C2
log(C1/C2)= 2870 x 103 / 5.5 x 103 = 517
Thus : (C1/C2)= 10517; if C1 = 10-4 M, then C2 = 10-521 M
Take Avogadro’s number as 6 x 1023 and the Oceans as 1021litres
Then one molecule in the whole of the Oceans would have a concentration of ~10 -45 M, i.e. 10476
time greater than the minimum concentration!!!!
Thus it would economic to acquire one molecule of glucose from 10 476 oceans, although the airfreight charge
would be excessive.
Note also that as the mass of the oceans is of the order 10 24g and the universe of the order 1055g, we could
only have a mere 1031 oceans
The outcome of the calculation is that the work required to concentrate material against a concentration
gradient is disappearing small.
It’s counterintuitive, but
that’s not good grounds for
rejecting it – we may have
used the wrong physical
chemical argument of course.
Williams_Lecture 2
DOM Stability, Lability and Resistance
9/12
PART II: CONCEPTUAL MODELS OF OCEANIC DOC PROFILES
The second section of this talk is largely drawn from Williams (2000)
I want to look at the fundamental structure of models used to describe DOC persistence through the
water column.
Three modes of decomposition were considered by Williams (2000):
A) Continuous decomposition
B) Periodic decomposition
C) Intermittent decomposition (I probably won’t cover this one, certainly in any detail as it takes too
much time)
The problem is to have models, with realistic biological parameterisation that can:
DOC concentration (M)
0
50
0
Age: c. 4,000 yrs
3
Conc: c.80m-molC/m
1,000
Depth (m)
Cycle time
c.1,500
2,000
3,000
Age: c. 6,000 yrs
Conc: c.50m-molC/m3
4,000
5,000
We need a model that
behaves broadly like this
That can explain profiles
that look like this
100
Williams_Lecture 2
DOM Stability, Lability and Resistance
10/12
Essentially we have to get two properties right:
the typical concentrations of surface and deep waters (c.80m-mol C/m3 and c.50 m-mol
C/m3)
and the observed 14C-determined ages of the surface and deep ages (c.4,000 and 6,000
years)
i)
ii)
A) Continuous decomposition
Early water quantity work used models (first order models) in which the rate of decomposition of a
substrate is dependent upon its concentration, i.e.
dC/dt = kC
when Ct = Coe-kt
It is easy, using “depth for time” models to write equations that give varyingly accurate adequate
representations of DOC profiles.
1) Single component models (DOCt = DOCoe-kt) – this is shown in A below – we can explain the
rapid phase, but not the tail. We can either model the DOC age profile or its concentration
profile, but not both (see Table below)
2) We can add on some refractory component (R) such as the monster molecule shown earlier
giving (DOCt = DOCoe-kt + R), shown in B. However although this looks nice and has a warm
feel about it, in truth it’s overly simplistic, and inventing resistant organic material to enable the
equation to fit the profile the profile doesn’t get us very far. The reason being that molecule
needs to be everlasting and so has an infinite age, and so the age of our deepwater material
would exceed the observed 6,000 years
3) We can use a multi-component model, comprising labile (fast decomposing) and resistant (slow
decomposing) decay constants (ΣDOCt = [DOC-labile]t + [DOC-resistant]t = [DOC-labile]oe-k1t
+ [DOC-resistant]oe-2kt. This (C below) gives quite good representations and looks fine and
dandy but there is a major problem. You can tune the model to give good representations of
both the age of the DOC and its concentration. The rate constants for the decay process require
very slow rates – requiring the decomposition to go on over thousands of years
DOC concentration (m-mol/m3)
50
DOC concentration (m-mol/m3)
100
0
0
0
1,000
1,000
2,000
2,000
3,000
Depth (m)
Depth (m)
0
B
3,000
50
C
DOC concentration (m-mol/m3)
100
0
50
0
+
1,000
Depth (m)
A
2,000
Fast decay
Slow decay
Combined
3,000
4,000
4,000
4,000
5,000
5,000
5,000
100
Williams_Lecture 2
DOM Stability, Lability and Resistance
11/12
SINGLE EXPONENTIAL EQUATION
Solved for age
Annual organic input
(m-mol/m3 a)
Solved for
concentration
DUAL EXPONENTIAL EQUATION
Fast rate
constant
Slow rate
constant
16.7
16.7
16.7
0.007
0.00017
0.203
10
0.00011
Half-life
4,080
years
3 years
25 days
6,300 years
Modelled age
(a)
6000
6
6000
50,000
50
48
Decay rate constant
(a-1)
Modelled
conc.(moles/m3)
The problem we face with this type of model is that it requires continued metabolism over very long
periods, i.e. very low rates. Although chemical reactions, e.g. radioactive decay, may occur over
millennia, in the case of biological reactions the organism has to keep the metabolic engine in working
order, against the forces of entropy. This in principle sets some limit to the slowness of biological
reactions – the through put of energy is not sufficient to keep the “plant” in order. Consider the
analogy that, whereas one can operate a factory producing 10 cars a day, producing the same number
over 10 years simply is not economic
B) Periodic decomposition
This dilemma is circumvented if decomposition is periodically
switched on and off. Williams (2000) discusses various
switches, the only one that stood up to close analysis was the
switching on and off of photochemical reactions. This type of
model of decomposition has built into it the notion of
recalcitrant molecules (i.e. xenophoric structures) that
susceptible only to photochemical attack.
Here the asymmetry between chemical and bacterial reactivity,
noted earlier is critical. The concept is that whereas microbial
reactions can in principal occur all the time during the cycling
of the oceans, photochemical can only occur during their brief
period (say 5-10% of the time) they are in the illuminated zone
– this is illustrated in the two figures below. Implicit in the
concept is that photochemical reactions are relatively slow,
due to the low flux of photons. Anderson and Williams (1999)
worked with a model of this type (see figure below) and were
able to produce profiles of DOC concentration that matched
the classical profiles, furthermore, with no fitting of constants,
they obtained an age of 2300 years for the deepwater DOC –
which is good for a start.
DOC production
Labile
DOC
DOC
Xenophoric
DOC
Microbial
Decomposition
Photochemical
Decomposition
Labile
DOC
Microbial
Decomposition
Williams_Lecture 2
DOM Stability, Lability and Resistance
12/12
Residence time c. 50-100 years
Microbial + Photochemical
Time Course of Periodic Decomposition
Residence time c. 1,500years
Only Microbial processes
80
Remaining material
Cycle time
c.1,500
60
40
20
0
0
1500
3000
4500
6000
7500
Time (years)
Figure 10. Steady state solution of model (01000 m):
modelled DOC divided into
refractory (stippled), semilabile material
originating from the mixed layer (shaded),
semilabile organics originating from detrital
hydrolysis below the mixed layer (striped),
and labile (unshaded, barely visible on
graph); origins of the semilabile material
were derived by splitting S into two state
variables and running to steady state; data
(total DOC) are Sargasso Sea (solid circles),
Sargasso Sea (solid diamonds),, tropical
Atlantic (solid crosses) North Central Pacific
(open crosses and diamonds) and equatorial
Pacific (open circles)
USEFUL READING MATERIAL
Williams P J L (2000) Heterotrophic Bacteria and the Dynamics of Dissolved Organic Material pp153200 in Kirchman Microbial Ecology of the Oceans John Wiley NY.
Carlson C A (2002) Production and Removal Processes. Chapter 4 pp. 91-151 in Hansell & Carlson
Biogeochemistry of Dissolved Organic Matter Academic Press San Diego