LIMNOLOGY
AND
September 1997
Volume 42
OCEANOGRAPHY
Limnol Ocermogr., 42(6), 1997, 1307-1316
0 1997, by the American Society of Limnology
and Oceanography,
Number 6
Inc
Invited Review
Role of photoreactions in the formation of biologically
compounds from dissolved organic matter
labile
Mary Ann Moran
Department of Marine Sciences, University
of Georgia, Athens, Georgia 30602-3636
Richard G. Zepp
Ecosystems Research Division, National Exposure Research Laboratory,
U.S. Environmental Protection Agency, Athens, Georgia 30605-2700
Abstract
Dissolved organic matter (DOM) can be degraded by sunlight into a variety of photoproducts that stimulate the
growth and activity of microorganisms in aquatic environments. All biologically labile photoproducts identified to
date fall into one of four categories: (1) low-molecular-weight
(MW) organic compounds (carbonyl compounds
with MW of <200); (2) carbon gases (primarily CO); (3) unidentified bleached organic matter; and (4) nitrogenand phosphorus-rich compounds (including NH,+ and POb3-). A number of laboratory studies using bacterial bioassay approaches have shown that the photochemical breakdown of DOM can stimulate biomass production or
activity by 1.5- to 6-fold. Results of photochemical studies, extrapolated to estimate formation rates of biologically
available photoproducts from DOM in surface waters, also predict important biological roles for these compounds.
In a continental shelf system, for example, full exposure of surface seawater to sunlight for one summer day can
produce DOM photodegradation products equivalent to >20% of the bacterial carbon demand. Likewise, 30% of
the bacterial nitrogen demand can be met by photodegradation of the nitrogen components of DOM, a process
likely to be of particular importance in nitrogen-limited systems. When considered on a depth-integrated basis
around the globe, at least 1.0 X lOI g C and 0.15 X 1O1’g N are estimated to be available annually for utilization
by planktonic microorganisms through the conversion of light-absorbing fractions of DOM to more biologically
labile compounds. By comparison, direct photochemical mineralization of DOM is estimated to convert 12-16 X
lOI g C to CO2 annually.
When sunlight is absorbed by dissolved organic matter
(DOM) of either freshwater or marine origin, the average
molecular weight (MW) is reduced and a variety of photoproducts are formed (Zepp et al. 1995). Some of these phoAcknowledgments
Support was provided by NSF LMER (DEB 94-12089) and Ecosystems (DEB 92-22479) Programs. The United States Environmental Protection Agency through its Office of Research and Development also partially funded and collaborated in the research
described here under Cooperative Agreement 821469 to the University of Georgia as part of the Boreal Ecosystem-Atmosphere
Study. We thank R. G. Wetzel and D. Kirchman and two anonymous
reviewers for helpful comments on this manuscript, 0. C. Zafiriou
for sharing information on quantum yields for marine CO photoproduction, and Neil Blough for providing prepublication copies of
manuscripts on the optical properties of coastal DOC and POC.
toproducts are inorganic compounds [e.g. carbon monoxide,
carbon dioxide, and other forms of dissolved inorganic carbon (DIC)] that represent direct photochemical mineralization of the DOM. Other photoproducts are organic molecules
that, although smaller in size than the parent molecules, remain part of the DOM pool. Although chemical characteristics of the bulk DOM clearly are changed by photochemical processes, it is also the case that the biological reactivity
of the DOM is altered. Evidence now indicates that photodegraded DOM can be significantly more biologically active
than is the parent DOM from which it was formed.
Scientists interested in the aesthetics of drinking water
were the first to demonstrate that the color of DOM was
bleached by sunlight (e.g. Whipple 1914). (The term
“bleached” is defined here as a decrease in the absorptivity
of the DOM in the UV and visible spectral regions.) In 1970,
1307
1308
Moran and Zepp
Gjessing and Gjerdahl (1970) estimated that 20% of the reduction in color of a Norwegian lake was due to bleaching
by UV radiation. A few years later, Strome and Miller (1978)
made the first connection between DOM bleaching and biological activity by describing a “priming effect” of sunlight
on the decomposition of natural DOM by bacteria. Almost
a decade later, the connection between photochemical processes and the biological availability of DOM was reintroduced by Geller (1986), and the potential role of photochemistry in affecting the supply of biological substrates became
more widely recognized.
While enhanced activity of microorganisms is by no
means the only important effect of DOM photochemistry in
aquatic environments (others include direct photochemical
mineralization of DOM, increased availability of CO, for
autotrophic fixation, effects on metal cycling and availability, and many more; see Karentz et al. 1994; Wetzel et al.
1995; Zepp et al. 1995), this particular process has considerable implications for aspects of carbon and nutrient cycling
in aquatic environments and substrate availability for microbial growth and metabolism. Here we summarize what is
currently known about biologically available photoproducts
of DOM and examine the potential biogeochemical role of
these compounds in aquatic environments.
Biologically
available
photoproducts
Changes that occur in the characteristics of bulk DOM or
humic subfractions of DOM following exposure to sunlight
are consistent with degradation of larger, photoreactive molecules into smaller, more UV-transparent compounds. Gel
filtration studies of the average MW of bulk DOM indicate
a reduction in molecular size following light exposure (Allard et al. 1994; Amador et al. 1989; DeHaan 1993; Hongve
1994; Strome and Miller 1978). Likewise, sunlight-induced
decreases in both the absorptivity and fluorescence efficiency
of DOM are widely reported (De Haan 1993; Frimmel 1994;
Hongve 1994; Kieber et al. 1990; Kouassi and Zika 1990;
Lindell et al. 1995; Miller and Zepp 1995; Stewart and Wetzel 1981; Strome and Miller 1978; Zepp 1988). That losses
of DOM generally do not keep pace with decreases in absorptivity or fluorescence efficiency suggests that photodegradation results in the formation of compounds that, while
still part of the DOM pool, no longer absorb light to the
same extent as does the parent material (Allard et al. 1994;
Backlund 1992; Strome and Miller 1978; Zepp et al. 1995).
It is within this reservoir of photochemically modified, more
weakly absorbing molecules that biologically available photoproducts likely reside.
Low-MW organic compounds-The
number of biologically available photoproducts identified to date is - 17 (Table
l), 14 of which are low-MW organic molecules. All of the
identified low-MW organic substances are carbonyl compounds, with fatty acids and keto acids being well represented. Most of these contain three or fewer carbon atoms
and have a MW of <loo, although their size ranges up to
a six-carbon molecule (Kieber and Mopper 1987; Kieber et
al. 1989; Kieber et al. 1990; Mopper and Stahovec 1986;
Mopper et al. 1991; Wetzel et al. 1995). Additional carbonyl
Table 1, Biologically available compounds formed via photochemical processes from aquatic DOM.
Compound
MW
Structure
Reference
Low MW Organic Comp wnds
v
Mopper and Stahovec 1986
Mopper et al. 1991
Kieber et al. 1990
?-
Wetzel et al. 1995
DahlCn et al. 1996
Acetaldehyde
44
H3C-C=O
Acetate
59
H3C-C=O
Acetone
f8
Citrate
1t:9
9
Mopper and Stahovec 1986
H3C - C - CH3
51
F-O- 9
9
Wetzel et al. 1995
-O-C-CH2-~-C1J2-C-O0-
Mopper and Stahovec 1986
Mopper ct al. 1991
Kieber et al. 1990
Formaldehyde
:A0
H2C = 0
Formate
4
AC=0
Glyoxal
fi8
HC-CH
Glyoxalate
“3
HC-C-O-
Levulinate
I 5
H3C-C-CHz-CH2-C-O-
Malonate
102
-0-C-CHy6-O-
?-
Wetzel et al. 1995
DahlCn et al. 1996
529
Mopper and Stahovec 1986
Mopper et al. 1991
Kieber and Mopper 1987
Mopper et al. 1991
Kieber ct al. 1990
89
e
9
P
Wetzel et al. 1995
0
Dahlen et al. 1996
99
Mopper and Stahovec 1986
Methylglyoxal
72
Oxalate
:<8
00
-o-E-e-o-
Propanal
.i9
H3C - CH3 - CH
Pyruvate
37
H3C-C-C-O-
$?11
Kieber and Mopper 1987
Kieber et al. 1989
Mopper et al. 199 1
Wetzel et al. 1995
28
co
Mopper et al. 1991
Jones 1991
Valentine and Zepp 1993
Schmidt and Conrad 1993
Miller and Zepp 1995
HC-C-CH3
DahlCn et al. 1996
fi’
Mopper and Stahovcc 1986
Carbon Gases
Carbon Monoxide
Nutrients
Ammonium
17
Phosphate
35
Bushaw et al. 1996
PO43-
Francko and Heath 1982
compounds (e.g. succinate, butanoic acid) have been reported to be formed from DOM under artificial light with emission in the UVC range (Allard 1994; Backlund 1992; Corin
et al. 1996), bul; not yet with natural or simulated sunlight.
Carbon gase:+-Although most of the interest in biologically available photoproducts has focused on the low-MW
organic compounds, at least three other classes of photo-
Labile DOM photoproducts
1309
products may have important effects on biological activity.
The first of these is the inorganic carbon-based gases that
are formed photochemically from DOM, including carbon
monoxide (CO) (Mopper et al. 1991; Valentine and Zepp
1993), CO, and other forms of DIC (Graneli et al. 1996;
Miller and Zepp 1995; Salonen and Vahatalo 1994), and
carbonyl sulfide (COS) (Andreae and Ferek 1992). COS is
not likely to have an important biological role in aquatic
ecosystems because there is no known microbial sink (Najjar
et al. 1995), and will not be discussed further.
CO can be utilized by a group of chemoautotrophic bacteria and thus is potentially of biological importance as a
microbial energy source in sunlit waters. CO is one of the
most abundant photoproducts of DOM. It is produced at
rates several-fold greater than any of the low-MW organic
compounds (Mopper et al. 1991). Concentrations of CO can
reach 50 nM in the surface of freshwater systems on sunny
days (Schmidt and Conrad 1993) and 6 nM in the open
ocean (Conrad et al. 1982; Jones 1991). Uptake by chemoautotrophic bacteria is just one potential route of CO loss
from aquatic environments; the competing route is rapid volatilization to the atmosphere from surface waters, which can
be supersaturated with CO by a factor of 30 or more (Conrad
et al. 1982; Schmidt and Conrad 1993).
DIC formed by photodegradation of DOM can be expected to have a limited impact on biological activity because inorganic carbon is generally nonlimiting to autotrophic growth in aquatic ecosystems; exceptions may include
acidic lakes and other systems with low DIC availability.
Nonetheless, the biogeochemical role of photoproduced DIC
may be substantial. Of all the identified photoproducts of
DOM, DIC is formed most efficiently (Miller and Zepp
1995), and available data indicate that the rates of DIC photoproduction are 15-20 times greater than rates of CO photoproduction from the same water. The first report of this
process in a natural freshwater ecosystem was by Miles and
Brezonik (1983), with recent demonstrations in marine (Miller and Zepp 1995), estuarine (Amon and Benner 1996), and
other freshwater (Graneli et al. 1996; Salonen and Vahatalo
1994) environments.
Nitrogen- and phosphorus-rich
compounds-The
final
group of biologically available photoproducts consists of labile nitrogen and phosphorus compounds. Ammonium has
recently been identified as a photoproduct of the organic
nitrogen component of humic substances and bulk DOM
from a number of freshwater environments (Bushaw et al.
1996). Amino acids are another potential class of nitrogenrich photoproducts of DOM (Bushaw and Moran unpubl.
data); free amino acids have been shown to react readily with
humic substances (forming biologically unavailable nitrogen) and to subsequently be released from humic substances
by exposure to UV light (Amador et al. 1989). Humic substances and other high-MW components of DOM typically
contain up to 2% organic nitrogen by weight (Thurman
1985), about half of which is composed of amino sugars or
is readily converted to ammonia and amino acids by weak
acid hydrolysis (Schnitzer 1985); it is this component of
DOM-associated nitrogen that is hypothesized to be the
source material for nitrogen-rich photoproducts (Bushaw et
al. 1996).
Phosphate is also a demonstrated photoproduct of DOM.
Francko and Heath (1979, 1982) described the release of
phosphate from humic substances by the UV-mediated reduction of iron complexes that bind humic substances and
phosphate together. Photochemically driven phosphate release has been observed in situ, demonstrated as diurnal fluctuations of free phosphate in the surface waters of a bog lake
(Francko and Heath 1982), but is a reversible reaction; in
the dark, iron is reoxidized and phosphate readsorbed to the
humic substances. This process may have important implications as a source of microbial phosphate in some aquatic
environments, particularly low pH freshwater environments
(Francko 1990; Stewart and Wetzel 1981), although evidence
for the importance of DOM photochemistry in influencing
phosphate availability across a range of aquatic systems has
not been forthcoming (Cotner and Heath 1990; Francko
1986; Jones et al. 1988).
Unidentified bleached organic matter-Along
with formation of the identifiable low-MW photoproducts, exposure
to sunlight may also transform DOM into other organic substances that have not been chemically identified but are
nonetheless more susceptible to microbial attack. Miller and
Zepp (1995) found that the conversion rate of DOM to identifiable photoproducts was <20% of the bleaching rate of
the DOM, leaving a large unidentified pool of bleached organic matter that could serve as a source of microbial substrates. Alternatively, as suggested years ago, photoreactions
of marine DOM may play a role in polymerizations that form
marine humic substances (Stuermer and Harvey 1978; Harvey et al. 1983), rendering the DOM less biologically available. Miller and Moran (1997) provided initial evidence that
photochemically modified higher molecular weight components of DOM can indeed be readily assimilated by microorganisms, and that unidentified bleached compounds can in
some cases account for a large proportion of the total biologically available photoproducts.
The clearest demonstration that sunlight enhances microbial activity through alterations of the DOM pool comes
from studies of the transfer of organic carbon to heterotrophic bacteria (Table 2). The approach used in a number of
studies has been to inoculate natural bacterial assemblages
or bacterial isolates into solutions containing DOM or DOM
subfractions, with and without prior exposure to sunlight;
differences in bacterial activity (either biomass accumulation
or growth rate) between irradiated and dark treatments are
used as a measure of the effect of sunlight on the biological
availability of DOM.
In the earliest studies of Strome and Miller (1978), exposure of humic substances isolated from a high-latitude
pond to sunlight resulted in bacterial growth 2- to 3-fold
higher than in unexposed samples. Subsequent studies have
shown higher bacterial activity on photodegraded DOM by
factors of 1.5- to almost 6-fold (Table 2). Variations in the
extent to which the activity of heterotrophic bacteria is enhanced by sunlight are likely due to differences in the length
Effects of photolabile DOM on biological activity
1310
Moran and Zepp
Table 2. Bioassay studies demonstrating effects of photolabile DOM on the activity of heterotrophic bacteria.
DOM
type
Humic substances
Humic substances
Bulk DOM
Pyruvate
Bulk DOM
Humic acids
Humic acids
Plant leachates
Humic substances
Bulk DOM
Humic substances
Exposure
regime*
Activity
index
Enhancement
factor-j
Reference
Artificial
Natural
Natural
Natural
Artificial
Natural
Artificial
Artificial
Natural
Natural
Artificial
Bacterial numbers
Bacterial numbers
Change in [DOC]
Change in [pyruvate]
Bacterial biomass
Bacterial production
Bacterial production
Bacterial production
Bacterial production
Bacterial biomass
Bacterial production
2.1
2.8
1.5
5.8
1.5
2.5
1.5
3
1.8-2.8
3
Strome and Miller 1978
Strome and Miller 1978
Geller 1986
Kieber et al. 1989
Lindell et al. 1995
Wetzel et al. 1995
Wetzel et al. 1995
Wctzcl et al. 1995
Bushaw et al. 1996
Lindell et al. 1996
Miller and Moran 1997
* Exposure to either natural sunlight or to artificial sunlight in a solar simulato:
t Calculated as increase in activity in samples exposed to sunlight relative to dark controls.
and type of light exposure, concentration of DOM, and possibly the source of DOM (Table 2).
Although stimulation of the activity of heterotrophic bacteria by DOM photoproducts is well documented, the effect
of CO photoproduction on chemoautotrophic bacteria is less
clearly understood. A number of different bacterial groups
are known to utilize CO, although we can restrict our consideration to the carboxidotrophs, aerobic bacteria that use
CO as an energy source, as opposed to those that oxidize
CO as a consequence of broad substrate specificity of their
enzymes but are unable to obtain energy; these include ammonia oxidizers and methylotrophs (Jones and Morita 1983;
Morsdorf et al. 1992). Arguments against an important biological role for photochemically produced CO in most
aquatic environments include the facts that cultured carboxidotrophic bacteria are unable to assimilate CO efficiently at
concentrations found in surface waters (Morsdorf et al. 1992;
Schmidt and Conrad 1993) and that the growth of bacterioplankton in lake water was only occasionally stimulated
by even very high additions of CO (Kraffzik and Conrad
199 1). Moreover, microbial turnover times of CO (mediated
both by carboxidotrophs and those bacteria that gain no energy advantage) in various marine and freshwater environments are up to several orders of magnitude longer than the
residence time of CO in surface waters (20-250 h compared
to 3-4 h; Zuo and Jones 1995); thus, bacteria are competing
with rapid volatilization for access to the CO pool. Nonetheless, it has been often demonstrated that a microbial sink
for CO exists in natural waters (Conrad et al. 1983; Johnson
and Bates 1996; Schmidt and Conrad 1993; Zuo and Jones
1995), leaving unresolved the question of the biological role
of photochemically produced CO in aquatic environments.
Microbial activity has also been shown to be enhanced by
the photochemical formation of readily assimilable nitrogen
compounds. Under N-limiting
conditions, the release of nitrogenous photoproducts
from aquatic humic substances was
found to significantly
increase rates of bacterial
growth
(Bushaw et al. 1996). Nitrogen-rich
photoproducts are likely
to be of greatest biological
interest in coastal oceans and
other ecosystems where plankton are N limited and concen-
trations of light-absorbing DOM are high (Bushaw et al.
1996). SecondarIr effects of DOM photoreactions on N cycling that involve interactions of CO photoproduction with
nitrification in the upper ocean also have been postulated
(Jones and Morita 1983).
Photochemistry in complex natural systems
Translating the growing number of laboratory studies
demonstrating enhanced microbial activity into estimates of
the effects of biologically available photoproducts in spatially and temporally complex aquatic environments is a difficult undertakin;. What follows is a description of several
factors that influence the rate of formation of biologically
available photoproducts in situ. We then attempt to model
this process in simplified aquatic systems.
Action spectra-Not all wavelengths of the solar spectrum
are equally efficient at producing photoproducts. The wavelength dependence for DOM photolysis has been quantified
through the determination of “action spectra,” plots of the
number of photoproduct molecules formed per incident photon vs. wavelength. Although reported action spectra for
are sparse, particularly
for marine
DOM photoreactions
DOM, available data indicate that ultraviolet radiation, especially in the UVB region of the spectrum (280-315 nm),
is most effective, with some action in the short-wavelength
(blue) part of the visible region as well. Action spectra for
DOM photoreactions are plots of the cross-product of absorption coefficil=nt times the quantum yield (the number of
molecules of photoproduct formed per photon absorbed) vs.
wavelength. Quantum yields provide a useful unitless gauge
for comparison of photoreactions of DOM from different
natural waters. Quantum yields for photoproduction of hydrogen peroxide (H,O,), a byproduct of the photooxidation
of DOM, have been found to be very similar in both estuarine and open-ocean
waters
(Moore
et al. 1993), and the
wavelength dependence of quantum yields for the photoproduction of CO (Valentine and Zepp 1993) and H,O, are
very similar (Fig. 1). Likewise, quantum yields for the pho-
1311
Labile DOM photoproducts
le-3
4
toproduction
co
slope = 0.020 f 0.002
(? = 0.84)
3
a,
‘5
E le-4
1e-5
300
350
400
450
Wavelength, nm
1e-5
slope = 0.024 f 0.004
(J! = 0.72)
\
300
350
400
of CO from freshwater
DOM
(Valentine
and
Zepp 1993) (Fig. 1) are similar to those that have been observed for marine DOM (Zafiriou pers. comm.). Action spectra for the photoproduction of reactive oxidants from DOM
similarly indicate that UV radiation is the most effective
region of the solar spectrum (Blough and Zepp 1995). Although one action spectrum reported for DOM in coastal
waters exhibited a surprisingly sharp drop-off at wavelengths
>320 nm (Kieber et al. 1989), the wavelength dependence
for quantum yields illustrated in Fig. 1 appears to be representative for most classes of DOM photoproducts.
450
Wavelength, nm
Fig. 1. Plots of quantum yields vs. wavelength for production
of CO (top) and H,O, (bottom) from DOM photoreactions. The
quantum yield is a unitlcss parameter that represents the fraction of
absorbed light that results in a photoreaction. Data for freshwater
arc shown as filled symbols; for seawater, open symbols. Slope of
natural log plots of data vs. wavelength indicate that wavelength
dependence for each data set is the same within experimental error.
Symbols: v, Suwannee River, Georgia; A, Okefenokee Swamp,
Georgia; 0, Houghton Marsh, Michigan; n , Kinoshe Lake, Ontario, Canada; 0, Orinoco River, Site R4/F, Venezuela; + , Orinoco
River, Site RUE;, Venezuela; A, Caribbean Sea, Site L/F-2, 60 m;
0, Caribbean Sea, Site L/F-2, surface. Data are from Valentine and
Zepp (1993) and Moore et al. (1993).
Photoproduct formation in natural waters---To explore
other factors that influence DOM photoreactions, we utilize
an action spectrum based on the CO data in Fig. 1. Translating an action spectrum into in situ rates of photoproduct
formation in aquatic environments requires consideration of
at least two aspects of light availability. First, solar spectral
irradiance is not distributed equally across all wavelengths.
The irradiance in the UVB range is at least lo-fold less than
in the visible region (400-700 nm) (Kirk 1994), and integrated irradiance in the UV region (280-400 nm) accounts
for only 8% of the total radiation outside the earth’s atmosphere and an even smaller fraction at the earth’s surface
(Frederick et al. 1989). Second, penetration of sunlight into
water likewise does not occur equally across the spectrum,
and UVB wavelengths are more effectively absorbed than
are visible wavelengths. The wavelength dependence of the
formation rates of DOM photoproducts in surface waters exposed to natural sunlight therefore looks quite different than
the action spectrum, and the shape of this “potential photoproduction” spectrum changes with depth as the downwelling irradiance becomes more rapidly depleted of the
UVB component relative to the visible component (Fig. 2).
The depth-averaged curve, however, still indicates that UV
radiation is the most effective wavelength region.
Simulations of the rates of DOM photoproduct formation
vs. depth for different types of marine and freshwater environments are shown in Fig. 3, with rates normalized to the
rates at the water surface. The curvatures of the lines are
attributable to the changes in rate vs. wavelength curves
shown in Fig. 2; as longer wavelength radiation makes an
increasingly important contribution to photoreactions with
increasing depth, the attenuation per unit depth gradually
decreases. Note that in open-ocean water, significant DOM
photoreactions occur throughout most of the mixed layer and
the depth at which the rate drops to 1% of its surface value
is close to 80 m. In coastal regions, attenuation is more pronounced with increasing depth, and the depth to 1% of surface photoreaction rates is <20 m.
In wetlands and highly colored rivers and lakes, the rapid
attenuation of sunlight in water dictates that the production
of most biologically available DOM photoproducts is constrained to the top meter of the water column (Fig. 3). In
situ studies indeed show rapid declines in the extent of DIC
photoproduction (Graneli et al. 1996; Salonen and Vahatalo
1994) and DOM photobleaching (De Haan 1993) with depth,
and depth profiles in high-DOM freshwater environments
indicate that most photochemistry occurs in the top 5-20 cm
Moran and Zepp
1312
0
10
20
250 300 350 400 450 500 550 600 650
30
Wavelength, nm
40
-1
We),,,
Fig. 3. Depth profiles of relative rates of formation of biologically available DOM photoproducts for representative marine and
freshwater ecosystc ms. Estimates of diffuse attenuation coefficients
for the boreal ponc sample (A) were based on an absorption spectrum obtained on .I Shimadzu spectrophotometer using a 0.2qm
filtered water sample. Diffuse attenuation coefficients for water
from the Gulf of hlexico (B), moderately productive seawater (C),
and the Sargasso Sea (D) are from Smith and Baker (1979). The
assumed action spectrum is based on that of Valentine and Zepp
( 1993) for CO pho toproduction.
05' _
0.0
I
250 300 350 400 450 500 550 600 650
Wavelength, nm
Fig. 2. Relative rates of formation of biologically available
DOM photoproducts as a function of depth and wavelength in the
open ocean (A) and coastal ocean (B). Rates are the cross-product
of the spectral irradiance for mid-summer, midday at 20”N latitude
and the average response function for CO photoproduction (Valentine and Zepp 1993). Each rate is normalized to the maximum
near-surface rate, which occurs at 330 nm. Scalar irradiance at a
given wavelength and depth was computed from a diffuse attenuation coefficient based on underwater irradiance measurements of
Smith and Baker (1979) for the SargassoSea (A) and coastal Gulf
of Mexico (B). The average rate assumes a completely mixed water
column.
of the water col~unn (De Haan 1993; Graneli et al. 1996;
Salonen and Vahatalo 1994).
Along with DOM, suspended particulate material also intercepts sunlight and can contribute substantially to total
light attenuation in highly turbid or productive environments, a fact that complicates theoretical calculations of light
penetration and photoproduct formation. On the midshelf of
the southeastern United States, for example, plankton and
particulate detritiil material absorb lo-50% of the incident
sunlight in the violet and blue regions (at 412 and 433 nm),
depending on the season (Nelson and Guarda 1995). However, recent measurements of underwater UV radiation in
lakes and coastal regions of North America indicate that solar UV radiation, especially in the UVB region, is predominantly absorbed by the DOM (Degrandpre et al. 1996; Williamson et al. 15196).Thus, although absorption of UV and
blue sunlight by suspended particulates competes with DOM
absorption, the resulting reductions in the formation of biologically labile DOM photoproducts are not predicted to be
large.
Temporal variation in the concentration of photochemitally active DOM may also influence rates of formation of
biologically labile photoproducts. Geller (1986) found measurable photochemical stimulation of heterotrophic bacterial
growth on the DOM from lake water collected in the winter,
Labile DOM photoproducts
but not in the summer. Likewise, a 5-fold decrease in the
rate of photochemical formation of ammonium was observed
for DOM from a high-latitude pond from early to late summer (Bushaw et al. 1996). One successful approach to normalizing for variations in the photoreactivity of DOM is to
express photoproduct formation rates relative to the lightabsorbing characteristics of the DOM. Normalization to the
absorption coefficient at 350 (abs,,,) or 300 nm (abs,,,,) has
been found to eliminate a significant amount of variation in
the rates of photoproduction of low-MW carbonyl compounds (Kieber et al. 1990), CO (Valentine and Zepp 1993),
and ammonium (Bushaw et al. 1996) from natural organic
matter.
Finally, it is not yet known whether different sources of
DOM (e.g. marine vs. terrestrial sources) yield different
types or ratios of photoproducts or whether all photoproducts
are produced in any given sample of DOM regardless of its
source. Although some photoproducts have now been identified independently in several studies of DOM photochemistry (e.g. CO, formaldehyde, and acetate), others have been
reported only once (e.g. levulinate, propanal) (Table 1).
Likewise, different light sources or different intensities of
exposure might cause the formation of different suites of
photoproducts from DOM. Lastly, it has been shown that
some photoproducts are themselves photoreactive and thus
degrade with continued light exposure (Kieber et al. 1990),
a fact that further complicates attempts to predict the concentrations of all possible photoproducts simultaneously.
Full sunlight estimates-As a first step toward predicting
rates of formation of biologically available photoproducts of
DOM in natural environments, we start with a highly simplified model system-a sample of coastal seawater fully
exposed to 1 d of summer sunlight. We assembled available
formation rate data from studies conducted in both marine
and freshwater systems, concentrating as much as possible
on studies conducted with natural sunlight (Table 3), with
the goal of estimating photoproduction of all identified biologically available photoproducts in a single coastal seawater sample. Because intensity of light exposure differed
among the studies, we normalized to average exposure on a
sunny summer day at 32”N latitude (total solar irradiance of
0.063 W cm-*). Likewise, because DOC content and absorption properties varied among these studies, we normalized to a water sample typical of the midshelf of the southeastern United States with a DOC concentration of 83 PM
C and absorption coefficients of 5.5 m-* (abs,,) and 2.7 m-l
(abs,,,) (Nelson and Guarda 1995; Zepp and Schlotzhauer
1981). Given these assumptions, the production of biologically labile photoproducts from DOM in this coastal seawater sample is predicted to be -0.90 p,M C in the form of
low-MW organic compounds, 0.80 FM C in the form of CO,
and 0.12 +M N in the form of ammonium during the course
of a cloudless summer day (Table 3).
Although there are large uncertainties in these calculations, they allow us to begin to place the availability of DOM
photoproducts into a framework for comparison with microbial requirements for carbon and nitrogen. For example, bacterial production in midshelf water averages 2.0 I.LM C d-l
(calculated from Griffith et al. 1990). Assuming 50% bac-
1313
Table 3. Estimated formation of biologically labile photoproducts from DOM in seawater from the continental shelf of the southeastern U.S. during exposure to 1 h of noontime summer sunlight
at 32”N latitude. Calculations were made by normalizing for light
intensity, length of exposure, and light absorption properties of the
DOM among a number of studies. Daily production rates given in
the text were calculated by assuming 7.6 h of noontime sunlight per
day.
Compound
Production
MM CC
or N) h-l)
Low-MW organic compounds
Acetaldehyde
11
Acetate
Formaldehyde
Formate
Glyoxal
5
14
1
6
Glyoxylate
22
Levulinate
Propanal
Pyruvate
46
3
12
Carbon gases
Carbon monoxide
Nutrients
Ammonium
108
17
Reference
Kieber et al. 1990
Mopper et al. 1991
Mopper and Stahovec
Wetzel et al. 1995
Kieber et al. 1990
Mopper and Stahovec
Wetzel et al. 1995
Mopper et al. 1991
Mopper and Stahovec
Kieber et al. 1990
Mopper ct al. 1991
Wetzel ct al. 1995
Mopper and Stahovec
Mopper et al. 1991
1986
1986 .
1986
1986
Mopper et al. 1986
Valentine and Zepp 1993
Miller and Zepp 1995
Bushaw et al. 1996
terial carbon conversion efficiency for low-MW organic
compounds and utilization of 10% of the CO by bacteria
with 16% conversion efficiency (Miirsdorf et al. 1992), biologically labile photoproducts could account for 2 1% of the
bacterial production in near-surface waters. Bacterial nitrogen demand averages -0.4 p,M N d-l, assuming a 5 : 1 C :
N ratio in bacterial biomass (Nagata and Watanabe 1990),
and thus photochemical ammonification could account for
-30% of the bacterial nitrogen requirement. These calculations do not consider potential adverse effects of UV light
on microbial activity due to photoinhibition (Herndl et al.
1993; Zepp et al. 1995), although such an effect would likely
change only the timing and(or) physical location of microbial utilization of the labile photoproducts. These calculations may be revised upward when more becomes known
about the nature and bioavailability of other low-MW photoproducts and bleached organic matter.
Regional to global estimates-For
the previous calculations, all DOM was assumed to be exposed to full intensity
sunlight. However, much of the DOM in aquatic environments resides at depths where light attenuation reduces or
eliminates photochemical processes, making it important to
evaluate the biological importance of DOM photoproducts
on a depth-integrated basis as well. In this section, we make
1314
0’
Moran and Zepp
’
BON
’
60N
I
’
40N
’
20N
’
0
’
20s
’
’
40s
’
60s
h
80s
Latitude
Fig. 4. Estimated depth-integrated photochemical formation of
biologically available carbon photoproducts from DOM and surface
area of the ocean for 10” latitude bands. Photoproduction calculations assumeparticle-free water of sufficient depth for all sunlight
to be absorbed,the action spectrum of Valentine and Zepp (1993),
and cloud-free skies. Ocean area is from Sverdrup et al. (1942).
Annual depth-integrated production for 10” increments in latitude
are based on fluxes that were computed for each month using a
previously described model (Valentine and Zepp 1993) and recent
total ozone data (Harris et al. 1995).
simple estimates of depth-integrated global production of
photoproducts at 10” latitude intervals.
Photoproduction estimates of CO were made using a computer program that has been described elsewhere (Valentine
and Zepp 1993). The program calculates solar spectral irradiance and photodegradation rates at various latitudes, times,
ozone columns, and water depths. Spectral irradiance in the
UVB and UVA regions were computed using equations previously described by Green et al. (1980). Because DOM
photodegradation is affected by changes in UVB radiation,
recent data on ozone depletion (Harris et al. 1995) were used
to increase photoproduction rates accordingly. To estimate
the formation of CO on a depth-integrated basis, average
rates and fluxes were computed assuming completely mixed,
particle-free water of sufficient depth for all the active wavelengths of sunlight to be absorbed. As shown in the plots of
DOM photodegradation rates vs. wavelength (Fig. 2), the
active wavelengths are largely confined to the UV region.
We further assumed that all the solar UV radiation that enters
the ocean is absorbed by DOM, an assumption that likely is
approximately correct for coastal regions (Nelson and Guarda 1995; Degrandpre et al. 1996), although less certain for
the open ocean where chromophoric DOM concentrations
are much lower. Photoproduction rates of low-MW organic
compounds and ammonium were estimated using the ratios
of photoproduction of CO : low MW organic compounds and
CO : ammonium determined for surface waters (Table 3).
This assumption is equivalent to assuming the same wavelength dependence for the quantum yields as that observed
with CO and thus the same attenuation of photoproduction
with depth for these classes of products as computed for CO
(Fig. 3).
Average photoproduction for each 10” latitude band was
multiplied by the area of ocean in that band (Fig. 4) to estimate annual global photoproduction; freshwaters contribute
negligibly to the total water-covered area of the globe. Based
on these calculations, we estimate that at least 1.7 X lOI g
C of biologically available photoproducts are produced annually (0.82 X lOIs as CO and 0.92 X lOIs as low-MW
organic compounds), with the highest production rates in
tropical and subtropical latitudes. Assuming again that 10%
of the CO and all of the low-MW organic compounds are
utilized by bacterioplankton, at least 1.0 X 10’” g C of biologically available photoproducts are estimated to be assimilated by microorganisms annually. By this same method,
total global pholoproduction of ammonium is estimated at
0.15 X 10ls g N y-l. Minimum estimates of the area-weighted photoproduction (i.e. the annual formation of DOM photoproducts per rrerage square meter of aquatic environment
on the globe) arc: 4.8 g C m * yr-I (2.3 g C as CO and 2.5
g C as low-MW organic compounds) and 0.42 g N m * yr-I
(as ammonium).
These estimates of the global production of biologically
available photoFroducts are conservative in some respects
(e.g. with regard to the limited number of photoproducts
considered) but liberal in others (e.g. with regard to the assumption that all solar UV radiation in the ocean is absorbed
by DOM). They are presented as a mechanism to stimulate
interest in the global implications of biologically labile
DOM photoproducts. Our current estimates suggest that
DOM photoproducts can supply <4% of the annual bacterial
carbon demand (assuming annual global primary production
of >50 X 10ls g C and that 60% of primary production
fluxes through bacteria; Hedges 1992; Cole et al. 1988) and
5% of the annual bacterial nitrogen demand (assuming 50%
bacterial growth efficiency and a 5 : 1 C : N ratio in bacterial
biomass). Baseci on arguments that unidentified bleached
photoproducts might be formed at rates many times faster
than CO (Miller and Zepp 1995; Miller and Moran 1997),
the upper estimate of global production of biologically available C photoproducts may be higher.
Photochemical degradation of DOM into more biologically available r:ompounds may be viewed not only as a
potential source of substrates for the microbial loop but also
as a route to the remineralization of DOM with otherwise
extremely long biological turnover times. Our annual global
estimate of photoproduction of biologically labile carbon
compounds (CO plus low-MW organic compounds) is
-0.25% of the oceanic DOC pool (700 X 101sg C; Hedges
1992). Further assuming that DOM photodegradation produces DIC at ral:es 15-20 times higher than CO (Miller and
Zepp 1995), we estimate that 28-46 g C m-* yr-I or 12-16
X 10’” g C yr-I are produced as DIC. Based on the sum of
CO, low-MW crganic compounds, and DIC photoproduction, annual loss. of DOM via photochemical degradation is
2-3% of the oceanic DOC pool and greater than the estimated annual input of terrestrial DOC to the ocean (0.2 X
lOIs; Meybeck 1982). Although these estimates are based on
what may be generous assumptions, they suggest a potential
role for DOM photochemistry in the oceanic carbon cycle.
Regional and gll3bal estimates of the biogeochemical importance of DOM photodegradation will benefit greatly from
future research focusing on the biological availability of unidentified bleached organic matter, the photochemical reactivity of open ocean DOM, and the role of particles in the
absorption of UV light in the open ocean.
Labile DOM photoproducts
References
’
AI~LARD, B., H. B~RBN,
C. PATTERSSON, AND G. ZHANG. 1994.
Degradation of humic substances by UV irradiation. Environ.
Internat. 20: 97-101.
AMADOR, J. A., M. ALEXANDER, AND R. G. ZIKA. 1989. Sequential
photochemical and microbial degradation of organic molecules
bound to humic acid. Appl. Environ. Microbial. 55: 28432849.
ANION, R. M. W., AND R. BENNER. 1996. Photochemical and microbial consumption of dissolved organic carbon and dissolved
oxygen in the Amazon River system. Geochim. Cosmochim.
Acta 60: 1783-1792.
ANDREAE, M. O., AND R. J. FEREK. 1992. Photochemical production of carbonyl sulfide in seawater and its emission to the
atmosphere. Global Biogeochem. Cycles 6: 175-183.
BACKLUND, I? 1992. Degradation of aquatic humic material by ultraviolet light. Chemosphere 25: 1869-1878.
BI,OUGII, N. V., AND R. G. ZEPP. 1995. Reactive oxygen species
in natural waters, p. 280-333. Zn C. S. Foote et al. [eds.],
Active oxygen: Reactive oxygen species in chemistry. Chapman & Hall.
BUSHAW, K. L., AND OTHERS. 1996. Photochemical release of biologically available nitrogen from dissolved organic matter.
Nature 381: 404-407.
COLE, J. J., S. FINDLAY, AND M. L. PACE. 1988. Bacterial production in fresh-and saltwater ecosystems: A cross-system overview. Mar. Ecol. Prog. Ser. 43: l-10.
CONRAD, R., M. AKAGNO, AND W. SEILER. 1983. Production and
consumption of carbon monoxide in a eutrophic lake. Limnol.
Oceanogr. 28: 42-49.
-,
W. SEILER, G. BUNSE, AND H. GIEIIL. 1982. Carbon monoxide in seawater (Atlantic Ocean). J. Geophys. Res. 87: 88398852.
CORIN, N., F? BACKLUND, AND M. KULOVAARA. 1996. Degradation
products formed during UV-irradiation of humic waters. Chcmosphere 33: 245-255.
COTNER, J. B., AND R. T. Hum.
1990. Iron redox effects on photosensitive phosphorus release from dissolved humic materials.
Limnol. Oceanogr. 35: 1175-l 181.
DAHLBN, J., S. BERTILSSON, AND C. PETTERSSON. 1996. Effects of
UV-A irradiation on dissolved organic matter in humic surface
waters. Environ. Internat. 22: 501-506.
DEGRANDPRE, M. D., A. VODACEK, R. E NELSON, B. J. BRUCE, AND
N. V. BLOUGII 1996. Seasonal seawater optical properties of
the U.S. Middle Atlantic Bight. J. Geophys. Rcs. lOl(Cl0):
22,727-22,736.
DE HAAN, H. 1993. Solar UV-light penetration and photodegradation of humic substances in peaty lake water. Limnol. Oceanogr. 38: 1072-1076.
FRANCKO, D. A. 1986. Epilimnetic phosphorus cycling: Influence
of humic material and iron on coexisting major mechanisms.
Can. J. Fisheries Aquat. Sci. 43: 302-310.
-.
1990. Alteration of bioavailability and toxicity by phototransformation of organic acids, p. 167-177. In E. M. Pcrdue
and E. T Gjessing [eds.]. Organic acids in aquatic ecosystems.
Wiley.
AND R. T. HEATH. 1979. Functionally distinct classes of
complex phosphorus compounds in lake water. Limnol. Oceanogr. 24: 463-473.
-,
AND -.
1982. UV-sensitive complex phosphorus:
Association with dissolved humic material and iron in a bog
lake. Limnol. Oceanogr. 27: 564-569.
FREDERICK, J. E., H. E. SCIINELL, AND E. K. HAYWOOD. 1989.
Solar ultraviolet radiation at the earth’s surface. Photochem.
Photobiol. 50: 443-450.
1315
RIMMEI,, E H. 1994. Photochemical aspects related to humic sub-
F
stances. Environ. Internat. 20: 373-385.
1986. Comparison of mechanisms enhancing biodegradability of refractory lake water constituents. Limnol.
Oceanogr 31: 755-764.
GJESSING, E. T., AND T GJERDAHI,. 1970. Influence of ultra-violet
radiation on aquatic humis. Vatten 26: 144-145.
GRANBLI, W., M. LINDELI,, AND L. TRANVIK. 1996. Photooxidative
production of dissolved inorganic carbon in lakes of different
humic content. Limnol. Oceanogr. 41: 698-706.
GREEN, A. E. S., K. R. CROSS, AND L. A. SMIT~I. 1980. Improved
analytic characterization of ultraviolet skylight. Photochem.
Photobiol. 31: 59-65.
GRIFFITI~, P C., D. J. DOUGLAS, AND S. C. WAINRIGHT. 1990. Metabolic activity of size-fractionated microbial plankton in estuarine, nearshore, and continental shelf waters of Georgia. Mar.
Ecol. Progr. Ser. 59: 263-270.
HARRIS, N. R. I?, AND OTIIERS. 1995. Ozone measurements. Zn C.
A. Ennis led.], Scientific assessment of ozone depletion: 1994.
World Meteorological Organization.
HARVEY, G. R., D. A. BORAN, L. A. CHESAL, AND J. M. TOKAR.
1983. The stucture of marine fulvic and humic acids. Mar.
Chcm. 12: 119-132.
HEDGES, J. I. 1992. Global biogeochemical cycles: Progress and
problems. Mar. Chem. 39: 67-93.
HDRNDL, G., G. MULLER-NIKI,AS,
AND J. FRICK. 1993. Major role
of ultraviolet-B in controlling bacterioplankton growth in the
surface layer of the ocean. Nature 361: 717-719.
HONGVE, D. 1994. Sunlight degradation of aquatic humic substances. Acta Hydrochim. Hydrobiol. 3: 117-120.
JOHNSON, J. E., AND T. S. BATES. 1996. Sources and sinks of carbon monoxide in the mixed layer of the tropical South Pacific
Ocean. Global Biogeochem. Cycles 10: 347-359.
JONES, R. D. 1991. Carbon monoxide and methane distribution and
consumption in the photic zone of the Sargasso Sea. Deep-Sea
Res. 38: 625-635.
-,
AND R. Y. MORITA. 1983. Effects of several nitrification
inhibitors on carbon monoxide oxidation and methane oxidation by ammonium oxidizers. Can. J. Microbial. 30: 276-279.
JONES, R. I., K. SALONEN, AND H. DE HAAN. 1988. Phosphorus
transformations in the epilimnion of humic lakes: Abiotic interactions between dissolved humic materials and phosphate.
Freshwater Biol. 19: 357-369.
KARENTZ, D., AND OTHERS. 1994. Impact of UV-B radiation on
pelagic freshwater ecosystems: Report of working group on
bacteria and phytoplankton. Arch. Hydrobiol. Beih. Ergebn.
Limnol. 43: 31-69.
KIERER, D. J., J. MCDANIEL, AND K. MOPPER. 1989. Photochemical
source of biological substrates in sea water: Implications for
carbon cycling. Nature 341: 637-639.
AND K. MOPPER. 1987. Photochcmical formation of glyoxyli; and pyruvic acids in seawater. Mar. Chem. 21: 135-149.
X. ZHOU, AND K. MOPPER. 1990. Formation of carbonyl
compounds from UV-induced photodegradation of humic substances in natural waters: Fate of riverine carbon in the sea.
Limnol. Oceanogr. 35: 1503-l 5 15.
KIRK, J. T 0. 1994. Optics of UV-B radiation in natural waters.
Arch. Hydrobiol. Beih. Ergebn. Limnol. 43: 1-16.
KOUASSI, A. M., AND R. G. ZIKA. 1990. Light-induced alteration
of the photophysical properties of dissolved organic matter in
seawater. Neth. J. Sea Res. 27: 25-32.
KRAFFZIK, B ., AND R. CONRAD. 199 1. Thymidine incorporation
into lake water bacterioplankton and pure cultures of chemolithotrophic (CO, H,) and methanotrophic bacteria FEMS Microbiol. Ecol. 23: 7-14.
LINDEU,, J. J., W. GRAN~I,I, AND L. J. TRANVIK. 1995. Enhanced
ELIER, A.
1316
Moran and Zepp
bacterial growth in response to photochcmical transformation
of dissolved organic matter. Limnol. Oceanogr. 40: 195-199.
-,
AND -.
1996. Effects of sunlight on bacteAa growth in lakes of different humic content. Aquat. Micrab. Ecol. 11: 135-141.
MEYBECK, M. 1982. Carbon, nitrogen, and phosphorus transport
by world rivers. Am. J. Sci. 282: 401-450.
MILES, C. J., AND J? L. BRHZONIK. 1983. Oxygen consumption in
humic-colored waters by a photochemical ferrous-ferric catalytic cycle. Environ. Sci. Technol. 15: 1089-1095.
MILLER, W. L., AND M. A. MORAN. 1997. Interaction of photochemical and microbial processes in the degradation of refractory dissolved organic matter from a coastal marinc environment. Limnol. Oceanogr., 42: 1317-1324.
AND R. G. ZEPP. 1995. Photochemical production of dissoived inorganic carbon from terrestrial organic matter: Significance to the oceanic organic carbon cycle. Geophys. Res.
Lett. 22: 417-420.
MOORE, C. A., C. T. FARMER, AND R. G. ZIKA. 1993. Influence of
the Orinoco River on hydrogen-peroxide distribution and production in the castcrn Caribbean. J. Geophys. Res. 98: 22892298.
MOPPER, K., AND W. L. STAHOVEC. 1986. Sources and sinks of
low molecular weight organic carbonyl compounds in seawater. Mar. Chem. 19: 305-321.
-,
X. ZHOU, R. J. KIEBER, D. J. KIE~~R, R. J. SIKORSKI, AND
R. D. JONES. 1991. Photochemical degradation of dissolved
organic carbon and its impact on the oceanic carbon cycle.
Nature 353: 60-62.
MORSDORF, G., K. FIIUNZKE, D. GADKARI, AND 0. MEYER. 1992.
Microbial growth on carbon monoxide. Biodegradation 3: 6182.
NACJATA, T., AND Y. WATANABE. 1990. Carbon- and nitrogen-tovolume ratios of bacterioplankton grown under different nutritional conditions. Appl. Environ. Microbial. 56: 1303-1309.
NAJJAR, R. G., D. J. ERICKSON, AND S. MADRONICH. 1995. Modeling the air-sea fluxes of gases formed from the dccomposition of dissolved organic matter: carbonyl sulfide and carbon
monoxide, p. 107-132. Zn R. G. Zepp and C. Sonntag [eds.],
The role of nonliving organic matter in the earth’s carbon cycle. Wiley.
NELSON, J. R., AND S. GUARDA. 1995. Particulate and dissolved
spectral absorption on the continental shelf of the southeastern
United States. J. Geophys. Res. 100: 8715-8732.
SALONEN, K., AND A. VAHATALO. 1994. Photochemical mineralization of dissolved organic matter in lake Skjervatjern. Environ. Internat. 20: 307-3 12.
SCMMIDT, U., AND R. CONRAD. 1993. Hydrogen, carbon monoxide,
and methane dynamics in Lake Constance. Limnol. Oceanogr.
38: 1214-1226.
SCHNITZER, M. 1985. Nature of nitrogen in humic substances, p.
303-325. Zn G. R. Aiken, D. M. M&night,
R. L. Wershaw,
and I? MacCarthy Teds.], Humic substances in soil, sediment,
and water. Wiley.
SMITH, R. C., AND K. S. BAKER. 1979. Penetration of UV-B and
biologically effective dose-rates in natural waters. Photochem.
Photobiol. 29: 3 1l-323.
STEWART, A. J., AND R. G. W~TZI~L. 1981. Dissolved humic materials: Photodegradation, sediment effects, and reactivity with
phosphate and calcium carbonate precipitation. Arch. Hydrobiol. 92: 265-286.
STROME, D. J., ARID M. C. MILLER. 1978. Photolytic changes in
dissolved hunic substances. Verh. Internat. Vercin. Limnol. 20:
1248-I 254.
STUERMER, D. H., AND G. R. HARVEY. 1978. Structural studies on
marine humu:,: A new reduction sequence for cabon skeleton
determination Mar. Chem. 6: 55-70.
SVERDRUP, H. U., M. W. JOIINSON, AND R. H. FLEMING. 1942. The
oceans; their physics, chemistry, and general biology. PrenticeHall.
THURMAN, E. M. 1985. Organic geochemistry of natural waters.
Nijhoff/Junk.
VALENTINE, R. L. AND R. G. ZEPP. 1993. Formation of carbon
monoxide from the photodegradation of terrestrial dissolved
organic carbon in natural waters. Environ, Sci. Technol. 27:
409-412.
WETZEL, R. G., I? G. HATCHER, AND T. S. BIANCHI. 1995. Natural
photolysis by ultraviolet irradiance of recalcitrant dissolved organic matter 1.0 simple substrates for rapid bacterial metabolism. Limnol. Oceanogr. 40: 1369-l 380.
WHIPPLE, G. C. 1?14. The microscopy of drinking water. Wiley.
WILLIAMSON, C. E , R. S. STEMBERGER, D. I? MORRIS, T M. FROST,
AND S. G. PAULSEN. 1996. Ultraviolet radiation in North
American lakes: Attenuation estimates from DOC measurements and implication for plankton communities. Limnol.
Oceanogr. 41: 1024-1034.
ZEPP, R. G. 1988. Environmental photoprocesses involving natural
organic matte]; p. 193-214. Zn E H. Frimmel and R. E Christman [eds.], Humic substances and their role in the environment. Wiley.
-,
T V. CAL,AGHAN, AND D. J. ERICKSON. 1995. Effects of
increased solar ultraviolet radiation on biogeochemical cycles.
Ambio 24: 181-187.
AND P. E SC~ILOTZHAUER. 198 1. Comparison of photochemical behavior of various humic substances in water. III.
Spectroscopic properties of humic substances. Chemosphere
10:479-486.
Zuo, Y., AND R. D. JONES. 1995. Formation of carbon monoxide
by photolysis of dissolved marine organic material and its significance in t re carbon cycling of the oceans. Naturwissenschaften 82: 4 74-474.
Received: 8 July 1996
Accepted: 27 November 1996
Amended: 20 August 1997