vvEPA
United States
Environmental Protection
Agency
National Exposure
Research Laboratory
Athens, GA 30605-2700
Research and Development
EPA/600/S-98/003 May 1998
ENVIRONMENTAL
RESEARCH BRIEF
Impacts of Fire and Solar Ultraviolet Radiation on Trace Gas
Exchange in the Boreal Forest Biome: A Review
Richard G. Zepp and Roger A. Burke1
Abstract
Projections of future climate change suggest that the largest
increases in temperature and other climate variables will occur
in the high-latitude forests and wetlands of the boreal forest
biome. This region also is being exposed to enhanced solar
UV-B radiation caused by recent depletion of stratospheric
ozone. The climate change is expected to have major effects
on the extent and frequency of fires in the boreal forest. The
research described here was part of the Boreal Ecosystem-
Atmosphere Study (BOREAS) that took place during 1994-
1996 in northern Manitoba, Canada. Field studies were con-
ducted to measure soil-atmosphere fluxes of carbon dioxide,
methane, and carbon monoxide, and related ancillary data, for
both upland black spruce (located on poorly drained clay-
textured soils) and jack pine sites (well-drained, sandy soils)
that were in early stages of succession following stand replace-
ment fires that occurred within 7 years of BOREAS. Results of
these studies indicate that locally dependent changes in boreal
fire return intervals linked to global climate change represent a
potential biospheric/physical feedback that is likely to alter the
biosphere-atmosphere exchange of carbon gases. All of the
sites were net sinks of atmospheric methane with median
fluxes ranging from -0.3 to -1.4 mg CH4-C nr2 d-1. Median
fluxes of carbon dioxide from the forest floor to the atmosphere
ranged between 1 and 2 g C nr2 d-1. We estimate that soil CO2
emissions from recently burned boreal forest soils in the north-
ern hemisphere could be on the order of 0.35 Pg C yr1; this
release of carbon may offset up to 50%-60% of the carbon
taken up by the unburned boreal forest. The soil-atmosphere
fluxes of CO, a gas that plays an important role in controlling
1 Ecosystems Research Division, National Exposure Research Laboratory,
Athens, GA 30605-2700.
the oxidizing capacity of the troposphere, were generally posi-
tive at the warmer, sunlit bum sites, but negative (sink activity)
in the shaded, cooler control sites. These results indicate that
interactions between CO released from burned areas and
chemical processes in the troposphere likely have important
regional effects on atmospheric composition over the boreal
forest.
Other studies focused on trace gas biogeochemical processes
that affect organic matter cycling in the oxic zones of wetlands
in the boreal forest biome. Measurements of CO concentra-
tions and fluxes and of ammonium concentrations in selected
beaver impoundments, indicated that solar radiation stimulates
CO and ammonium production in these ecosystems. Labora-
tory studies indicated that UV-induced photodegradation of
dissolved organic matter (DOM) is the primary pathway for the
light-induced production of inorganic carbon and nitrogen. Car-
bon dioxide is the main identifiable photo product and it forms
over an order of magnitude more rapidly than CO. Bioassays
indicate that photodegradation of the DOM enhances its bio-
logical availability through production of inorganic nitrogen and
readily assimilable low-molecular-weight organic compounds.
Introduction
Projections of future climate change suggest that the most
dramatic increases in temperature and other climate variables
will occur in the high-latitude forests and wetlands of the boreal
forest biome. Moreover, the increases in solar UV-B radiation
that have occurred over this region in response to the thinning
of the stratospheric ozone layer are projected to persist well
into the next century.
The boreal forest is a fire-dependent ecosystem that is located
primarily in Canada, the United States (Alaska), and Russia. It
G§S Printed on Recycled Paper
-------
is capable of sustaining very large, high intensity wildfires.
Changes In climatic parameters (e.g., lightning frequency,
drought, temperature) and/or human activities strongly influ-
ence fire frequency and intensity, and future global warming
would likely increase the occurrence of boreal forest fires. The
changes in fire return frequency and resulting impacts on
biosphere/physical climate feedbacks are likely to exhibit a
strong local dependence.
About 700 Pg (1 Pg = 101S g) of carbon are stored in the boreal
forest btome [Apps et al., 1993], which accounts for roughly
30% of the carbon stored in the terrestrial biosphere. Thus,
there is concern that the enhanced warming and shifts in soil
moisture, combined with changes in fire regimes and solar UV
radiation, could cause significant changes in nutrient cycling
and releases of carbon dioxide (CO2), methane (CH4), carbon
monoxide (CO), and other radiatively and chemically important
carbon gases from the boreal forests and wetlands.
Here we report the results of a project that was part of the
Boreal Ecosystem Atmosphere Study (BOREAS), a
multidisciplinary investigation of the boreal forest biome that
Involved research teams from Canadian and U.S. institutions.
BOREAS was a cooperative field and analysis study involving
elements of terrestrial ecology, trace gas biogeochemistry,
modeling, and land surface climatology. The goal of BOREAS
was to understand the interactions between the boreal forest
btome and the atmosphere in order to clarify their roles in
global change. An overview of BOREAS and the results of
BOREAS studies can be viewed at the BOREAS Worldwide
Web site (URL is http://boreas.gsfc.nasa.gov).
The research described here focused on the exchange (net
uptake or emission) of carbon gases (CO2, CH4, and CO)
between the soils and atmosphere or water and atmosphere at
selected sites located in or near the BOREAS Northern Study
Area (NSA) [Sellers et al. 1995]. The NSA was located close to
Thompson, Manitoba, within a few hundred kilometers of the
northern boundary of the boreal forest. These experiments
were conducted in soils of upland black spruce and jack pine
stands that have recently experienced stand-replacement, crown
fires and in nearby controls that had not been burned for at
least 80 years. Other studies of trace gas exchange and
nutrient cycling were conducted in beaver impoundments lo-
cated in the NSA. Results of these studies indicate that the
locally dependent changes in boreal fire return intervals that
are linked to global climate change represent an important
biospheric/physical feedback that is likely to alter the bio-
sphere-atmosphere exchange of CO and CO2. The results
further indicate that solar UV radiation has significant effects
on the emissions of CO and production of ammonium in wet-
lands of the boreal biome.
Experimental
Forest Site Description
As shown on Figure 1, studies of the trace gas exchange were
conducted at both upland black spruce (Picea mariana) and
jack pine (Plnus bankslana) sites that had experienced intense,
stand-replacement fires within 7 years of the 1994 BOREAS
field campaigns (BOREAS '94). Four black spruce sites were
selected near the road to Gillam, Manitoba, about 100 km from
Thompson, Manitoba. Three of the sites had distinctly different
recent bum histories. The fourth site served as a control and
had not burned for at least 80 years. Sites are denoted as CBS
(control Wack spruce) and 94BS, 92BS, and 87BS (stands
burned in 1994, 1992, and 1987, respectively). All four black
spruce sites were within 5 km of each other and were exposed
to very similar climatic conditions. The jack pine site (89JP)
was located in a large burned forest (115,643 ha; summer,
1989) that is west of the NSA. A nearby jack pine stand (CJP),
unburned for at least 80 years, served as the control for the
89JP site.
The black spruce stands were located on an orthic grey luvisol
soil with high water holding capacity. Biomass densities of the
black spruce stands ranged from 40,000 to 180,000 kg/ha with
most sites at the lower end of this range. The forest floor of the
CBS site was predominantly covered by a thick layer of
feathermosses. The combined organic layers at this unburned
black spruce stand ranged to between 20 and 25 cm with
much thinner layers representing the areas covered by lichens
only. The burned sites had all experienced fires with intense
torching and crowning that resulted in 100% overstory and
ground cover mortality. Most of the surface was covered with
charred moss, other debris, and a variety of plants that are
characteristic of boreal forest regeneration. Young black spruce
seedlings were just starting to emerge at the 1992 and 1987
bum sites; the largest were generally less than 15 cm tall.
The forest floor of the control jack pine site was covered
mainly by a thin layer of reindeer lichen (Cladina spp.) with
very little understory. The burned jack pine site was about 5 km
away from the control site. As in the case of the black spruce
sites, all trees at the 89JP site had been killed by fire and most
were still standing during BOREAS '94 with only a thin layer of
charred organic matter found on the forest floor.
Wetland Sites
Most measurements and experiments conducted for this study
were made at a beaver impoundment (NSA Tower Pond) that
was selected for intense studies (tower flux measurements and
characterization of physical, chemical, and biological proper-
ties) during the BOREAS '94 field campaigns. It was located 13
km west of Thompson and approximately 300 m south of
Manitoba Hwy. 391 (55°55'N, 98°01'W). CH4 and CO2 water-to-
atmosphere fluxes were measured and process studies con-
cerning the dynamics of these gases around the open water,
vegetated zones and periphery of the pond are reported else-
where [Roulet et al., 1997].
Soil-Atmosphere Flux Methods
We used a static chamber technique [e.g., Whalen and
Reeburgh, 1988] to estimate soil/atmosphere exchange. The
chamber consisted of a permanently deployed circular alumi-
num collar, with a water seal and skirt (~10, 25, or 30 cm deep
in areas with significant moss, lichen, or burned vegetation
cover, ~5 cm deep in areas in which the vegetation had burned
to the mineral layer), and cover. For the CO2 and CH4 studies,
the cover was constructed of aluminum and was equipped with
a septum for syringe sampling and a small hole to equalize
pressure. A transparent, borosilicate cover was used for the
CO flux studies, so that the effects of solar radiation on CO
exchange could be investigated.
Gas flux samples (5 to 7 samples per flux measurement) were
collected at 5-7 minute time intervals over the course of 20-30
minutes in polypropylene (for CO2 and CH4 samples) or glass
(for CO samples) syringes and the gas samples in the syringes
were analyzed by gas chromatography after return from the
field sites to the laboratory.
-------
Stephen's Lake
Figure 1. Map of northern Manitoba showing locations of sites included in the studies reviewed in this report. The jack pine and black spruce
sites were located in upland stands that had experienced intense stand-replacement fires during the decade prior to the Boreal
Ecosystem-Atmosphere Study (BOREAS). The wetland sites were located at active beaver impoundments, one of which, tower
pond (TP), was a BOREAS tower site.
Wetland Studies
The flux of CO across the water-air interface was estimated
using the static floating chamber method. An inverted cylindri-
cal quartz chamber (cross sectional area = 491 cm2) was
mounted in a floating aluminum base (cross sectional area =
363 cm2) and sealed with distilled water. Gas flux samples
were collected in glass syringes for later analysis. Water samples
also were collected at various depths for measurements of CO
and ammonium concentrations and for subsequent photochemi-
cal experiments in the laboratory. The latter were filtered through
0.22 |j,m membrane filters and stored at 4°C.
Triplicate aliquots of the filtered water samples were sealed
without headspace in 30 mL quartz tubes, mounted and ex-
posed to natural light in an open water section of the NSA
tower pond at various depths. Following exposure, a known
volume of headspace (5-11 mL) was added to each tube by
removing water with a syringe and replacing it contemporane-
ously with CO-free air at atmospheric pressure. The tubes
were shaken for one minute and placed in a water bath. After
equilibration the headspace in each tube was sampled and
analyzed for CO by gas chromatography. The aqueous phase
also was analyzed for ammonium.
Water samples were irradiated using a solar simulator (DSET
Heraeus), in most cases with exposure to the full spectrum.
For some samples, spectral distribution of the radiation was
modified by using Schott glass filters to selectively block out
various wavelengths of UV radiation. Hydrophobic resins were
used to isolate and fractionate the humic substances in the
wetland waters. To evaluate the effects of exposure to solar
radiation on the bioavailability of the organic nitrogen com-
pounds in the wetland water, various bioassay treatments were
conducted in which the growth of a natural bacterioplankton
inoculum was followed over the next 92 hr in an aqueous
solution of humic substances that had been irradiated in the
solar simulator for various time periods. Twelve bioassay treat-
ments were established by additions of inorganic nutrients (N
only, P only, or N+P) and/or a labile carbon source (glucose) to
irradiated or non-irradiated fulvic acid solutions. Bacterial growth
was measured by the uptake of tritiated leucine. Other studies,
conducted in the laboratory, determined the increase in con-
centrations of CO and dissolved inorganic carbon (DIG) in the
water on exposure to simulated sunlight.
Analytical Procedures
A Carle AGC gas chromatograph (GC) equipped with a ther-
mal conductivity detector (TCD) and a flame ionization detector
(FID) was used to analyze for CO2 and CH4 and a Trace
Analytical RGA-3 Reduction Gas Analyzer equipped with mer-
curic oxide detector was used to measure CO concentrations.
The chromatographic responses were carefully calibrated ver-
sus standards provided by the Atmospheric Environment Ser-
vice of Canada (CO2 and CH^ or the National Institute of
Standards and Technology, Gaithersburg, MD (CO). Dissolved
organic carbon (DOC) and DIG in the wetland water samples
were determined using a Dohrmann Model DC85A carbon
analyzer, and ammonium concentrations were determined by
ion exchange chromatography with conductometric detection
and an absorptiometric phenol-hypochlorite method. Total ni-
trogen in the freshwater samples was determined by acid
digestion and chemical oxidation of the organic nitrogen in the
water samples to inorganic nitrogen as described in detail
under Method 351.2 in the FWPCA Methods of Chemical
Analysis of Water and Wastewater (FWPCA Methods, 1979).
-------
UV and visible absorption spectra were obtained using a
Shimadzu Model UV-160-U scanning spectrophotometer.
Results and Discussion
Soil-Atmosphere Exchange of Trace Gases
During Early Post-Fire Succession
Important processes affecting the net ecosystem exchange of
carbon gases are shown in Figure 2. Plants and soils act as
both sources and sinks for carbon gases. Living plants are the
major sink for CO, in the terrestrial biosphere and also are a
direct source of CO to the atmosphere that likely is altered by
the changes In vegetation species and bibmass densities dur-
ing post-fire succession. Past investigations of factors influenc-
ing soil production and consumption of CO2, CH4, and CO,
coupled with known impacts of fires on the boreal forest biome,
suggested to us that soil gas fluxes might well be strongly
altered by boreal fires. Boreal stand replacement fires alter the
thermal regime and moisture content of soil layers owing to
enhanced solar radiation reaching the soil surface. Increased
light exposure should stimulate CO production by surface
photodegradation and the warmer temperatures should en-
hance organic matter decomposition to CO2 and CO. Soil
warming, changes in soil moisture in burned forests, and post-
fire alteration in soil nutrient bioavailability also can alter micro-
bia! transformations in the soil matrix [Conrad, 1995], but the
net effect is not obvious. For example, soil warming in burned
soils tends to stimulate microbial activity, but increased con-
centrations of inorganic nitrogen released by decomposition of
soil organic nitrogen may decrease microbial oxidation of CH4
and CO by nitrifying bacteria. Finally, stand-replacement boreal
fires convert the boreal forest floor from what is generally a
thick living cover of mosses or lichens to a charred, blackened
layer of nonliving carbonaceous substances that are likely to
have quite different biogeochemical properties.
During the spring, summer, and fall of 1994-1996, we mea-
sured soil-atmosphere exchanges of CO., CH4, and CO and
related ancillary data near the BOREAS NSA at upland black
spruce (located on poorly drained clay-textured soils) and jack
pine sites (well-drained sandy soils) that were in early stages
of succession following stand replacement fires that occurred
within 7 years of the 1994 BOREAS field campaigns (Figure 1).
Nearby control stands that had not burned in the past 80 yrs
were studied for comparison. The CO2 and CH, fluxes were
measured using a static chamber technique with an opaque
cover.
Based on our measurements made during 1994 we conclude
that soil/atmosphere exchange of CH4and CO2 are significantly
altered by fire [Burke et al., 1997]. All of the sites were net
sinks of atmospheric CH4 with median fluxes ranging from -0.3
to -1.4 mg CH4-C m* d'1. We observed that sites that have
been recovering from fire for a few years (5 to 7) consume
significantly more methane than sites either very recently dis-
turbed (<1 yr) or sites that have fully recovered from previous
fires. Moreover, our median values indicate that the jack pine
soils tend to consume methane at a slightly greater rate than
the black spruce soils.
We used our soil respiration data, along with the assumption
that initial plant regrowth takes up a negligible amount of
carbon [e.g., Auclair and Carter, 1993] and some information
from the literature, to calculate a rough estimate of the magni-
tude of soil CO2 emissions from recently burned (< decade)
boreal forest [Burke et al., 1997]. For this calculation, we
assumed that 1012 m2 of the boreal forest burns in a decade
[Kasischke et al., 1995], that this area is similar to the burn
sites that we studied with respect to soil respiration rate, and
that the forest burned had a pre-burn net ecosystem carbon
uptake of 50 g C nr2 y1 as estimated for the BOREAS NSA
OBS site [Frolking et al., 1996]. We further assumed that the
rate of soil respiration observed at our burn sites (~1.5 g C rrr2
d'1) occurs for the ~5 months of the year when the soil is not
snow-covered and that the rate of soil respiration during the
rest of the year is equivalent to the 85 g C rrv2 observed for the
BOREAS NSA OBS site during the snow covered period
[Frolking et al., 1996]. These assumptions yield an estimate of
0.35 Pg C yr1 released as post-fire CO2 emissions in the
boreal region. Using similar assumptions regarding area burned,
but basing their CO2 emissions estimates on assumed break-
down rates of dead residual wood and SOM rather than direct
measurements, Auclair and Carter [1993] estimated that simi-
lar amounts of carbon (~0.3 Pg C yr1) would be emitted post-
fire from boreal and temperate forests. By comparison, the
remaining 11 x 1012 m2 of unburned boreal forest would be a
sink for about 0.55 Pg C yr1 if it has the net ecosystem carbon
uptake of 50 g C nr2 y1 that was estimated for a tower site
located in a mature black spruce stand in the BOREAS NSA
[Frolking et al., 1996]. This estimate of the unburned boreal
forest carbon sink is similar to the 0.7 Pg C yr'derived by Apps
et al. [1993] from estimated annual changes in biospheric
carbon pools. The magnitude of these estimates of post-burn
soil CO2 emissions, compared to the boreal biome carbon sink
estimates and within the context of the global carbon cycle
[e.g., Ciais et al., 1995], suggests that further study of post-fire
soil CO2 emissions is needed.
This study provided new evidence that boreal forest fires per-
turb soil-atmosphere exchange of CO fluxes [Zepp et al., 1997].
During BOREAS '94 (June to September 1994) CO was con-
sumed by the forest floors of mature black spruce and jack
pine stands in northern Manitoba that had not burned for close
to a century. In contrast, CO was generally emitted from soils
of forests that had experienced stand-replacement fires within
7 years prior to BOREAS '94.
The consumption of CO in mature black spruce and jack pine
stands on a given day and location was proportional to atmo-
spheric CO concentration. The median deposition velocities
(ratio of CO soil/atmosphere flux to CO atmospheric concentra-
tion) in these unburned forests was 0.008 to 0.015 cm s'1, at
the lower end of the range that has been observed in other
biomes, possibly reflecting the lower temperatures in boreal
forest soils.
Daytime CO fluxes at the bum sites were generally positive
(1011 to 1012 molec cnrv2s-1) and were lowered when solar
irradiance was excluded by covering the transparent chambers
or reduced during periods of cloudiness or smoke. Net fluxes
at the burn sites were controlled by competition between abi-
otic production, mainly at the surface, and microbial oxidation
in the soil. Abiotic production, which was attributable to photo-
production and thermal decomposition of the surface organic
layer and charcoal, strongly correlated with incident solar irra-
diance, with the greatest fluxes during midday. Burned moss
patches were the strongest sources of CO at the sites. Al-
though the soils of the most extensively burned sites were
significantly warmer than the nearby controls, these studies
indicate that photodecomposition of the surface organic matter
was the predominant process that produced CO. Thus, prior
field studies of CO fluxes in open areas using dark chambers
-------
Light Mediated
Direct Emission
Figure 2. Processes affecting the net ecosystem exchange of trace carbon gases and carbon cycling in the boreal forest.
may have greatly underestimated daytime emissions. Evidence
is presented that changes in the competition between soil
uptake and production with smoky air masses present serve to
buffer atmospheric CO concentrations and soil-atmosphere
fluxes. Changes in the soil fluxes of CO caused by fires in the
boreal biome may be a significant local source of atmospheric
CO during summer days (up to 10'* molec cnr2 S'1).
Solar UV Radiation, CO Emissions and Nutrient
Cycles in Selected Boreal Wetlands
Many boreal wetlands contain high concentrations of chro-
mophoric dissolved organic matter (CDOM) and thus are highly
colored. This CDOM, which is derived mainly from decomposi-
tion of terrestrial plant matter from trees and mosses, consists
mainly of high-molecular-weight humic substances that are not
readily assimilated by microorganisms [Bourbonniere et al.,
1995]. Earlier studies have indicated that CDOM from freshwa-
ter and marine environments in other biomes can be
photodegraded by solar UV radiation to various low-molecular-
weight compounds, including carbon monoxide [e.g., Valentine
and Zepp, 1993]. These studies mainly involved laboratory
studies of the effects of changing reaction conditions on rates
of CO photoproduction. Few field studies documenting the role
of sunlight in the production of CO in high-latitude biomes are
available.
Our in situ experiments and measurements of ambient concen-
trations in selected beaver impoundments (TP and another site
shown in Figure 1) confirm that CO production in such ponds
depends on sunlight [Bourbonniere et al., 1997]. The diurnal
changes observed in the flux of CO measured across the air-
water interface also indicate that photoproduction is important
and that the resulting supersaturation of surface water leads to
a net flux out to the atmosphere. The supersaturation factors
noted for this boreal beaver pond are the highest reported from
any aquatic system, five times higher than those for any other
freshwater system studied. The high concentration of CDOM in
this pond, and possibly its terrestrial origin, may be the primary
cause of high supersaturation and deserves further study. The
photoproduction rates of CO observed in quartz containers
positioned at various depths in tower pond were found to agree
closely with rates estimated using a computer model (Figure
3). The model used simulated solar spectral irradiance in the
UV region and the action spectrum for CO photoproduction to
compute the rates at various depths averaged over the full day.
Based on previous studies and the results presented here, the
exchange of CO reflects a dynamic balance between the
photochemical, biological, and physical processes at work here.
Floating quartz chambers seem to reflect near surface (~1 cm)
CO photoproduction rates and these "instantaneous" gas flux
data track changes in solar intensity fairly closely. Flux calcula-
tions based on CO supersaturation may more accurately repre-
sent gas fluxes in the absence of strong sunlight and incorpo-
rate some aspects of mixing. From a mass transfer point of
view, however, total water column CO photoproduction will
reflect the significance of solar irradiance and changes in UV-B
on CO exchange from these dark waters. For this, more com-
plete data on losses from biological consumption and on daily
and seasonal variations in the photochemical and optical prop-
erties of these waters are needed to constrain CO flux esti-
mates over time scales relevant to regional carbon balances.
To better understand the impacts of solar UV radiation on
processes occurring in boreal wetlands, we conducted a vari-
ety of laboratory studies using filtered water samples obtained
at our field sites [Bushaw et al., 1996; Zepp et al., 1995]. The
CDOM also was fractionated using hydrophobic resins to help
determine the response of various components to UV expo-
sure [Bourbonniere et al., 1995; Bushaw et al., 1996]. Labora-
tory studies of the water samples and the fractions were
conducted to determine the kinetics of photochemical changes
in optical properties and the photoproduction rates and quan-
tum yields of carbon dioxide, carbon monoxide, and various
photooxidants [Zepp et al., 1995]. The UV absorbance of the
-------
5 -
10 -
15 -
In [(rate)z/(rate)0]
-4 -3-2-1
Figure 3. Depth dependence for carbon monoxide production at a
beaver impoundment (TP, see Figure 1) located in the
Northern Study Area (NSA) of BOREAS. The terms, (rate)0,
and (rate)2 , refer to the production rates near the surface
and at depth z. Observed rates were determined for filtered
samples of TP water contained in quartz tubes to permit full
exposure to UV-B radiation (Bourbonniere et al., 1997). The
simulated rates were computed using the GCSOLAR
computer program (Valentine andZepp, 1993).
CDOM decreased on exposure to solar UV radiation and this
"fading" was accompanied by photoproduction of dissolved inor-
ganic carbon (DIG) as well as carbon monoxide. As observed
with water samples obtained from water bodies in the southeast-
em United States [Miller and Zepp, 1995], the DIG photoproduc-
tion was 15-20 times more rapid than CO photoproduction. The
hydrophilic acid fraction was significantly more photoreactive
than the hydrophobia acids. Quantum yields at 313 nm for the
hydrophobic acid fraction collected during July were found to be
0.0014 for DIG photoproduction and 4.9 x 10"5 for CO photopro-
duction. The latter value is close to the quantum yields deter-
mined for other wetlands in more temperate regions [Valentine
and Zepp, 1993].
Other studies were conducted to elucidate the possible role of
CDOM photodegradation in carbon and nitrogen cycling in
boreal wetlands. Exposure of filtered pond water, humic sub-
stances and hydrophilic acids isolated from the water indicate
that the CDOM can be degraded by sunlight into a variety of
photoproducts that stimulate the growth and activity of microor-
ganisms In aquatic environments. All biologically labile photo-
products identified to date fall into one of four categories: i) low
molecular weight (LMW) organic compounds (carbonyl com-
pounds with MW < 200); ii) carbon gases (primarily CO); iii)
unidentified bleached organic matter; and iv) nitrogen- and
phosphorus-rich compounds (including NH + and PO/-). A num-
ber 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 [Moran
and Zepp, 1997]. Microbial activity also has been shown to be
enhanced by the photochemical formation of readily assimi-
lable nitrogen compounds. For example, the dissolved organic
nitrogen in water samples and humic substances from the
BOREAS tower pond site were photodegraded to form ammo-
nium and low-molecular-weight organic amines [Bushaw et al.,
1996]. Solar UV radiation was responsible for inducing this
photo-ammonification process (Figure 4). Under N-limiting con-
ditions, the release of nitrogenous photoproducts from aquatic
humic substances was found to significantly increase rates of
bacterial growth [Bushaw et al. 1996]. Nitrogen-rich photoprod-
ucts are likely to be of greatest biological interest in aquatic
ecosystems where plants and plankton are nitrogen limited and
concentrations of light-absorbing DOM are high [Bushaw et al.
1996], e.g., in coastal wetlands and estuaries.
80
Figure 4. Photoproduction of ammonium on exposure of water
samples from the BOREAS tower pond to simulated solar
radiation. The full irradiance impinging on the samples was
close to that observed at mid-afternoon on a clear July day
in the BOREAS Northern Study Area. The rates were
significantly reduced when the radiation was filtered through
Schott glass filters that removed portions of the UV
component. The results indicate that photoreactions of the
dissolved organic nitrogen induced by solar UV radiation
were mainly responsible for the ammonium production.
References
1. Apps, M. J., W. A. Kurz, R. J. Luxmore, L O. Nilsson, R.
A. Sedjo, R. Schmidt, L. G. Simpson, and T. S. Vinson.
Boreal forests and tundra. Water, Air, and Soil Pollution,
70,39-53, 1993.
2. Auclair, A. N. D. and T. B. Carter. Forest wildfires as a
recent source of CO at northern latitudes, Can. J For.
Res., 23, 1528-1536, 1993.
3. Bourbonniere, R. A., L. A. Ziolkowski, S. L. Telford, M. A.
Moran, K. L. Bushaw, W. L. Miller, M. A. Tarr, and R. G.
Zepp. Character and Biogeochemistry of Dissolved Organic
Matter in a Boreal Forest Beaver Pond Near Thompson,
-------
Manitoba, Canada. In Proc. 17th Int. Mtg. Org. Geochem.,
September 1995, 134.
4. Bourbonniere, R. A., W. L. Miller, and R. G. Zepp.
Distribution, flux, and photochemical formation of carbon
monoxide in a boreal beaver impoundment. J. Geophys.
Res., accepted, 1997.
5. Bushaw, K. L, R. G. Zepp, M. A. Tarr, D. Schulz-Jander,
R. A. Bourbonniere, R. E. Hodson, W. L. Miller, D. A.
Bronk, and M. A. Moran. Photochemical release of
biologically labile nitrogen from dissolved organic matter.
Nature, 381: 404-407, 1996.
6. Burke, R. A., M. A. Tarr, W. L Miller, and R. G. Zepp.
Effect of fire on soil atmosphere exchange of carbon
dioxide and methane in a Canadian boreal forest. J.
Geophys Res., 102: 29,289-29,300, 1997.
7. Ciais, P., P. P. Tans, M. Trolier, J. W. C. White, and R. J.
Francey. A large Northern Hemisphere Terrestrial CO2
sink indicated by the 13C/12C ratio of atmospheric CO..
Science, 269: 1098-1102 (1995).
8. Conrad, R. Soil microbial processes and the cycling of
atmospheric trace gases, Phil. Trans. R. Soc. Lond. A,
351:219-230(1995.
9. Frolking, S., M. L. Goulden, S. C. Wofsy, S-M. Fan, D. J.
Sutton, J. W. Munger, A. M. Bazzaz, B. C. Daube, P. M.
Grill, J. D. Aber, L E. Band, X. Wang, K. Savage, T.
Moore, and R. C. Harriss. Modelling temporal variability in
the carbon balance of a spruce/moss boreal forest. Global
Change Biol., 2: 343-366, 1996.
10. Kasischke, E., N. L. Christensen, and B. J. Stocks. Fire,
global warming, and the carbon balance of boreal forests.
Eco/. Appl., 5: 437-451. 1995.
11. Kuhlbusch, T. A. J., R. G. Zepp, W. L Miller, and R. A.
Burke, Jr. CO soil-atmosphere deposition velocities of dif-
ferent soil layers in upland Canadian boreal forests. Tellus,
accepted, 1998.
12. Miller, W.L and R.G. Zepp. Photochemical production of
dissolved inorganic carbon from terrestrial organic matter:
significance to the oceanic carbon cycle. Geophys. Res.
Lett, 22:417-420, 1995.
13. Moran, M. A. and R. G. Zepp. Role of photoreactions in
the formation of biologically labile compounds from dis-
solved organic matter. Limnol. Oceanog., 42: 1307-1316
(1997). '
14. Roulet, N. T., P. Grill, N. Comer, A. Dove, and R.
Bourbonniere. CO2 and CH4 fluxes from a boreal beaver
pond. J. Geophys. Res., accepted, 1997.
15. Sellers, P., F. Hall, H. Margolis, B. Kelly, D. Baldocch, G.
den Hartog, J. Cihlar, M. G. Ryan, B. Goodison, P. Grill, K.
J. Ranson, D. Lettenmaier, and D. E. Wickland. The Bo-
real Ecosystem-Atmosphere Study (BOREAS): An over-
view and early results from the 1994 field year. Bull. Am
Met. Soc., 76: 1549-1577, 1995.
16. Valentine, R. L. and R. G. Zepp. Formation of carbon
monoxide from the photodegradation of terrestrial organic
matter. Environ. Sci. Techno!. 27: 409-412, 1993.
17. Whalen, S. C. and W. S. Reeburgh. A methane flux time
series for tundra environments. Global Biogeochem. Cycles,
2: 399-409, 1988.
18. Zepp, R. G., W. L. Miller, R. A. Bourbonniere, and M. A.
Tarr. Interactions of Changing Solar Ultraviolet Radiation
and Organic Matter Photooxidations in Northern Peatlands.
Preprints of Papers, 210th National Meeting, American
Chemical Society Division of Environmental Chemistry.
35,394,1995.
19. Zepp, R. G., W. L. Miller, M. A. Tarr, and R. A. Burke. Soil-
atmosphere fluxes of carbon monoxide during early stages
of post-fire succession in upland Canadian boreal forests.
J. Geophys Res., 102: 29,301-29,311, 1997.
-------
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
EPA/600/S-98/003
------- |