Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 55
Critical Processes Affecting the
Distribution of Chemical Species
BIOLOGICAL AND SURFACE SOURCES
BY. R. CICERONE, C. C. DEEWICHE, R. HARRISS, AND R. DICKINSON
THE IMPORTANCE OF BIOLOGICAL SOURCES
In recent years we have seen increasingly clear mani-
festations of the important, even dominant, roles of bio-
logical processes as sources of atmospheric chemicals.
While it has long been recognized that the sheer variety
and adaptivity of the biosphere permit a wide range of
phenomena, the quantitative size of biological influ-
ences on the atmosphere (examples below) has amazed
many atmospheric chemists and climate experts. More
biologically oriented scientists have long recognized the
great potential of atmosphere-biosphere exchange proc-
esses; it has even been proposed that much of our chemi-
cal and physical environment is under biological con-
trol. For example, Lovelock has postulated the existence
of an encompassing living feedback system through
which the biosphere regulates the physical environment
by and for itself as external stimuli change-the Gala
hypothesis.
There are both general and specific indications for the
importance of biology in affecting or controlling the
chemical composition ofthe atmosphere. Rough indica-
tions may be obtained from examining the state of dis-
equilibrium of the atmosphere's chemical concen-
trations of gases that would exist if the earth's oceans and
atmosphere were in perfect thermodynamic equilib-
rium (TE) or with only external inputs of solar, galactic,
and electrical energy. With calculations such as these,
Lovelock and Margulis found concentrations of N2, 02,
CH4, N2O, NH3, and CH3I many orders of magnitude
lower than in the actual atmosphere. Large departures
from pure TE or an analogous abiological system in the
atmosphere's composition are one indication of the in-
fluences of the biosphere. From a more empirical con-
sideration, a comparison with Venus and Mars, planets
with no life, suggests that there would be several orders
of magnitude less O2 on earth without life.
The importance of biological sources is indicated
strongly by other empirical data and analyses. For ex-
ample, the ~4C content of atmospheric CH4 was mea-
sured to be about 80 percent of that in recent wood.
These data, mostly from W. F. Libby, were gathered in
1949-1960, before nuclear explosive devices were tested
widely in the atmosphere, and they show that at least 80
percent of atmospheric CH4 then was derived from re-
cent organic material and not from old carbon, e.g.,
primordial CH4 or fossil fuels. Although the burning of
biomass could also yield CH4 high in t4C, it is likely that
ruminating animals and termites, rice paddies, and
shallow inland waters are the principal contemporary
CH4 sources. Also, from a geological and geochemical
analysis the mostly biological origin of atmospheric O2
and the important biological role in controlling CO2
levels can be deduced.
Other strong evidence exists for the importance of
biological processes in the cycles of many elements
55
OCR for page 56
56
whose volatile forms pass through and affect the atmo-
sphere. In the nitrogen cycle, biological and industrial
fixation of N2 from the atmosphere leads to subsequent
return of NH3, N2O, N2, and possibly NO to the atmo-
sphere. The importance of biogenic sulfur compounds
to the atmosphere and to the global sulfur cycle was
recognized long ago. Principal compounds of interest
are dimethyl sulfide, dimethyl disulfide, H2S, COS,
and possibly others. Methylated materials of several
other elements also appear to constitute globally impor-
tant biological sources. These include CH3Cl, CH3I,
CH3Br, and several methyl-metal compounds, e.g.,
those of mercury, arsenic, and probably others. While
biomass burning is a source of CH3Cl, the oceans ap-
pear to be a more important source. The physical
Cl--I- exchange reaction in seawater postulated in the
early 1970s may not be compatible with recent measure-
ments indicating the simultaneous existence of CH3I
and CH3C1 in seawater. In any case, CH3C1 is the only
significant natural source of chlorine atoms to the strato-
sphere, and its origin, probably biogenic, requires
study. Over land, much of the water transferred from
the surface passes through vegetation.
Regardless of whether the biosphere acts as an inte-
grated, almost purposeful, system in its release and reg-
ulation of atmospheric gases, or whether individual spe-
cies and blames act independently with inadvertent
results, the biologically released materials are important
in atmospheric chemistry and climate. Available data
show that biogenic sulfur compounds can make appre-
ciable contributions to sulfate deposition measured in
certain regions, and probably to dry SO: deposition; it is
important to define these contributions and regions
more-clearly. Similarly, field survey data and studies on
individual plants show that' the natural emissions of veg-
etation represent potentially significant hydrocarbon
sources for the atmosphere. Often these are isoprene
and terpenes (comprised largely of isoprene-like units).
These biogenic hydrocarbons react photochemically
and along with NOX gases can lead to photochemical
production of O3. There is also evidence that the direct
emission of natural hydrocarbons and the burning of
biomass can generate significant amounts of organic
acids in rainfall; tropical forest areas are especially inter-
esting in this regard.
Processes and raw materials that produce and/or con-
trol O3 concentrations in the background troposphere
are not only central to tropospheric chemistry but are
also of importance to climate. In the relatively clean,
unpolluted troposphere, O3 iS a major source of reactive
free radicals as well as a key reactant itself. Its climatic
role arises from its 9. 6-pm band absorption (in the atmo-
spheric wavelength window); in the troposphere, this
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
band is considerably pressure-broadened. In the photo-
chemical control oftropospheric 03, biogenic gases such
as CH4, CO, many hydrocarbons, N2O, and possibly
NH3 are important players, N2O through its strato-
spheric production of odd-nitrogen oxides that flow
downward into the upper troposphere, and NH3 as a
possible NOX source and as an NOX sink.
From a climatic point of view, the potential warming
effects of several biogenic and anthropogenic trace gases
are startling. Upward trends in the concentrations of
tropospheric O3, CH4, N2O, CH3CCl3, CCl2F2, and
CC13F are of particular concern, although other species
also need attention. The effects of these gases will add to
those of CO2. For O3, the extent of influence of biogenic
input, summarized above, is a key question. For CH4
and N2O, biological sources are known to dominate the
global budgets. For CH4 and CH3CCl3, whose primary
sinks are tropospheric OH reactions, any understand-
ing and ability to predict future trends will require
knowledge of the natural and human-controlled sources
and of global patterns of OH and their controlling proc-
esses. For the fluorocarbons CC12F2 and CCl3F, it will be
important to understand how tropospheric chemistry
will respond to the stratospheric changes they cause,
e.g., more ultraviolet radiation reaching the tropopause
and higher O3 concentrations in the lower stratosphere
and upper troposphere.
Finally, there is another large role for biological proc-
esses in atmospheric chemistry, that of surface receptors
for depositing gases and particles. There is evidence that
vegetated continental areas are major sinks for O3,
HNO3, SO2, and possibly NH3, for example. Further,
the effectiveness of these surface sinks is largely con-
trolled by dynamic responses of the involved plants. In
certain regions and seasons, the deposition of atmo-
spheric gases and particles delivers important nutrients
and, at other places and times, pollutants and toxins for
plant life and soils.
There has been an increasing awareness by the scien-
tific community of the critical role played by biosphere-
atmosphere interactions in biogeochemical cycles in gen-
eral and in tropospheric chemical cycles in particular. A
number of recent documents discuss these interactions
and their implications in considerable detail. Examples
include the National Research Council reportAtmosphere-
Biosph~re Interactiorl. Toward a Better Urlderstarlding of the Eco-
logical Cor~sequences of Fossil Fuel Combustion (~1981~; the
NASA Technical Memorandum 85629, Global Biology
Research Progress Program Plan (1983~; and several publi-
cations by the Scientific Committee on Problems of the
Environment (SCOPE), including Some Perspectives of the
May'or Biogeochern~'cal Cycles (1981), and The May'or Biogeo-
chernical Cycles and TheirInteractions (1983~.
OCR for page 57
CRITICAL PROCESSES
THE NATURE OF BIOLOGICAL SOURCES
Those atmospheric constituents to which biological
sources are major contributors (nitrogen, oxygen, CO2,
N2O, compounds of sulfur, halogens, and others) are
products ofthe energy reactions of one or another organ-
ism. The process of photosynthesis and its reversal (res-
piration) are the most familiar expression ofthis fact, but
the close similarity of processes yielding the other con-
stituents is not always appreciated. An examination of
the interrelationships between these biological oxida-
tions and reductions from the viewpoint of their driving
forces and modulating influences is in order.
First, although no one has yet offered a completely
convincing explanation of the driving force behind the
phenomenon called "life, " there is much evidence of its
potency. Virtually any chemical reaction that can take
place in an aqueous system, that yields energy in excess
of 40 kilojoules (kl) per mole for two-electron transfer,
and for which the required reactants have been available
in reasonable abundance over the evolutionary period of
the earth has been exploited. Aside from photosynthe-
sis, no physical source of energy has been so utilized.
The importance of this concept from the standpoint of
the atmospheric chemist is that it implies that the present
composition of the atmosphere is closely coupled to the
biological system, and that any alteration by physical
phenomena or human activity may be countered by a
biological response giving the appearance of "resist-
ance" to the change. As a single example from many
possible, the annual fixation of CO2 in photosynthesis
(and the equivalent release of CO2 in oxidative reac-
tions) is about 10 percent of the atmospheric carbon
pool. This speaks for an exceedingly tightly coupled
cyclic exchange.
Any increase in atmospheric CO2 levels would be
expected to result in a countering increase in photosyn-
thesis with the fixation of more CO2. This is known to
happen, but because there are so many other factors
influencing photosynthetic production, the relationship
. , .
Is not linear.
Of the elements from biological sources appearing in
the atmosphere, the ones most actively cycled are those
having several oxidation states within the range of stabil-
ity of water, and for which at least one reasonably stable
gaseous form exists. Under the reducing conditions of
anoxic environments (reducing because some biologi-
cally oxidizable compound is present and the influx of
atmospheric oxygen is limited by a diffusion barrier or
other means), compounds of these elements serve as
"electron acceptors" for the biological oxidation of
other, more reduced, compounds. In the process, gas
57
ecus compounds are released, some of them reaching
the atmosphere.
Typical reactions are as follows:
1. Denitrification:
a. [HCHO] + 0.8 NO3 + 0.8 H+-CO2 +
1.4H2O + 0.4N2
or
b. [HCHO] + NO3 + H+-CO2 +
1.5 H2O + 0.5 N2O
2. Sulfate reduction:
tHCHO] + 0.5 SO4 + H+-CO2 + H2O +
0.5 H2S
3. Hydrogen production:
tHCHO] + H2O ~ CO2 + 2 H2
4. Methane production:
tHCHO]-0.5 CO2 + 0.5 CH4
In all the above, tHCHO] represents carbohydrates,
although the organic substrate can be any of a variety of
materials. The reactions are simplified in their represen-
tation, with none of the intermediates shown. The point
is that each reaction yields one or more volatile com-
pounds to the atmosphere.
A cursory treatment of the subject, as given here, is
sufficient to demonstrate the variety of reactions possi-
ble and to suggest some of the implications discussed
below.
The operation of the various biomes contributing to
the whole of this biological process is dynamic, influ-
enced by all ofthe parameters that drive it. Most ofthese
ecosystems (indeed, all from an absolute standpoint) are
limited by one or more of their constituents. Available
organic substrate, as we have seen, is a major limitation,
but in most systems, one or more mineral elements
required for life are also limiting. Compounds of nitro-
gen and sulfur are among the more notorious products
of modern industry, and, although one tends to classify
these emissions as "pollutants," they probably also are
"fertilizers" for some species and in some locations.
Unquestionably, human activities have altered the bio-
sphere, but it is difficult to evaluate that alteration as
quantitatively as desired.
Concentration of atmospheric constituents on an ar-
eal basis is part of the question. Immediately downwind
of a point source, the concentration of an element or
compound can be lethal for some organisms. On the
basis of the considerations offered above, one can as-
sume that any alteration of the concentration of a partic-
ular element can only result in an alteration of the associ
OCR for page 58
58
ated biological population. The significance of this
alteration is more difficult to interpret.
The excess of CO2 injected into the atmosphere over
the rate at which it can be sequestered by various carbon
sinks is a good example of the complexities of this sort of
question. Concern over the possible effects that in-
creased atmospheric CO2 levels may have on climate is
tempered by the uncertainty regarding the conse-
quences ofthese effects. It is not known what would have
been "normal" secular trends in the absence of this
excess CO2. Intuitively, it is often assumed that any
change is undesirable, but firmer information is re-
quired for planning purposes.
An interesting feature of the atmospheric carbon cy-
cle is the role of CH4. Available data indicate that there
has been a significant increase in the concentration of
CH4 in the atmosphere since the industrial revolution,
and i4C data imply that 80 percent or more of atmo-
spheric CH4 was recently in living matter. The concen-
tration increase of CO2 is better documented. The fact
that the increase in atmospheric concentration of CO2 is
less than would be expected on the basis of known proc-
esses for the removal of carbon from the atmosphere has
led to some debate. The carbon of atmospheric CH4
contains less ~4C than that of atmospheric CO2. Biologi-
cal sources are largely "modern," and as noted above,
they are large. The quantity of fossil carbon from natu-
rat venting and fossil fuel burning does not appear to be
sufficient to explain the deficit of i4C in atmospheric
CH4. There is a discrimination against i4C in the CH4
formation reaction, but this can explain only part of the
deficiency in TIC.
Recent suggestions that large quantities of CH4 are
coming from magmatic sources could explain some of
this "fossil" CH4 but not all. Thus there are debates
within the debates, all of them emphasizing the need for
. ,% .
more mtormatlon.
The biological production of CH4is a marginal thing
from an energetic standpoint. The rather elegant expo-
sure of details of the process by H. A. Barker in 1941 has
proved to be even more complex. What was thought to
be the conversion of ethanol to acetate, with a concomi-
tant reduction of bicarbonate ion to CH4, has turned out
to be a coupling of reactions by two interdependent
organisms, one oxidizing ethanol to acetate with the
production of hydrogen, the other forming CH4 from
hydrogen and the bicarbonate. Although there are two
separate organisms involved, the removal of hydrogen
by the methanogen utilizing hydrogen is required to
provide the energy gradient for life support ofthe hydro-
gen producer.
Because of the close constraints placed on energy yield
(and therefore growth) by the concentration of hydrogen
gas in reducing systems, anything that reduces hydro
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
gen concentration will accelerate oxidation of available
organic substrate. On the other hand, hydrogen con-
sumption (and CH4 production) depends on CO2 con-
centration. Thus an increase in atmospheric CO2 levels
will correspondingly affect diffusion rates from CO2
sources (anoxic environments) and could stimulate CH4
production. This, in turn, could accelerate the oxidation
of available carbon compounds (some ofthem fossil) and
result in an increased production of CH4, some of it
deficient in i4C relative to atmospheric CO2. These en-
ergy relationships are shown below.
Fermentation of ethanol to acetate and hydrogen:
CH3CH2OH + H2O-CH3COO- + H+ + 2 H2.
At pH 7 and with other reactant concentrations stan-
dard, this reaction yields only 5.3 kJ of energy, insuffi-
cient to support life. With the partial pressure of hydro-
gen at 1.0 x 10-3 atm, the energy yield is about 39.4 kJ,
adequate forlife support if properly coupled to synthetic
reactions.
Oxidation of hydrogen with CO2 as an electron ac-
ceptor:
4H2 + CO2 2H2O + CH4.
The standard free energy for this reaction as written is
about-140 kJ. By talking the 1 x 10-3 atm concentra-
tion of hydrogen suggested by the former reaction, a
partial pressure for CO2 of 3.2 x 10-4 aim, and a partial
pressure for CH4 of 1.6 x 10-6 atm, the energy yield
becomes 85.0 kJ.
As written above, there are four molecules of hydro-
gen involved. The exact pathway of the reaction is not
known, so it is not possible to identify the point at which
energy is extracted. For each mole of water produced (a
two-electron process), there is a yield of 42.5 kJ of en-
ergy, and it is difficult to visualize any other energy-
coupling reactions. Because there is no room in the en-
ergy figures for increase in the CH4CO2 ratio, the
generalization probably is permitted that atmospheric
CO2 concentrations should influence biological CH4
production. The close parallel in their rates of increase
in the atmosphere may well be related to this interdepen-
dence of biological processes. The increase in CH4 in the
atmosphere would then be explained at least in part by
the mobilization of organic matter in anoxic zones,
some of which is "fossil" on the comparatively short
time scale (thousands of years) of carbon half-life.
The data to test the significance of processes such as
this are lacking, and the extent to which a concentration
feedback such as this can explain present inconsistencies
in the data is not known. The example does serve to
demonstrate the complex interaction of biological and
other factors in establishing atmospheric composition,
and the challenge to unravel the processes at play.
OCR for page 59
CRITICAL PROCESSES
GLOBALLY IMPORTANT BIOMES
Given the discussions above on the apparent impor-
tance of biological sources and on the nature of the bio-
logical processes that release materials to the atmo-
sphere, it is now necessary to review some of the
characteristics of world biomes and to formulate criteria
for evaluating their potential importance. Before exam-
ining data on various distinct biomes, we present the
following criteria that permit identification of biomes
that require research relative to their potential role as
significant sources of atmospheric chemical species:
1. Biomes coveringlarge geographic areas;
2. Biomes with high gross primary productivity
rates;
3. Biomes with fast cycling rates for nutrients;
4. Biomes with anoxic sites;
5. Biomes with fast rates of change of the local popu-
~at~on compared to the time scale for natural succession
(30 to 70 years);
6. Biomes where processes can trigger irreversible
changes (e.g., desertification or climatic change);
7. Biomes of special importance to human life (e. g.,
agricultural areas);
8. Biomes with unique characteristics (due, for ex-
ample, to toxicological, meteorological, or successional
considerations);
9. Biomes or processes that are poorly understood
and that satisfy some of the criteria above.
We will refer to these criteria frequently in the discussion
that follows.
Schemes to classify world biomes vary somewhat de-
pending on the purpose of the classification, need for
detail, and other reasons. Many of the available data
and compilations have grown from research on the
global carbon cycle and from needs of individual re-
searchers to extrapolate data from isolated, in situ mea-
surements into regional or global estimates. A further
complication results from the distinction between gross
and net primary productivity of the biomes. The former
is the rate of photosynthetic carbon fixation; the latter is
this rate minus the rate of respiration, i.e., the rate of
carbon storage. Table 5.1 presents one compilation of
data on terrestrial biomes, their sizes, net primary pro-
ductivities, and phytomasses. This particular compila-
tion accounts for nearly all terrestrial surfaces, or about
30 percent ofthe total global surface. For our purposes at
present, the most important entries in Table 5.1 are the
geographical areas covered by the individual biomes.
To identify biomes of special interest in atmospheric
chemistry, Table 5.2 lists about twenty specific, although
informally classified, biomes and one process, biomass
burning, as the rows of a matrix. The columns of this
~.
59
matrix are biogenic gases, individual species such as
CH4, and groups of species such as methylated metals
(CH3M) and organohalides (RX). A measure of the
scientific interest in the emissions of the listed biogenic
gases from the listed biomes is assigned, considering the
criteria outlined above and the available data. An "X"
indicates reason to expect significant emissions, and a
circled "X" indicates strong reason (or directly applica-
ble, available data) to expect a particularly significant
biome-emission relationship. In the remainder of this
section, we focus attention on several biomes and proc-
esses that are potentially significant as sources for tro-
pospheric chemical species.
T'~n~lra and Other Northern Environments
The "tundra" biome and the boreal forest at a lower
latitude cover about 14 percent of the land area of the
globe, most of it in the northern hemisphere. Total pho-
tosynthetic productivity of this area, although less than
many environments on an area basis, still is large (an
estimated 10 percent of all land area, based on the fig-
ures of Whitaker). Underlain by permafrost, much of it
poorly drained, this area could be a large contributor to
the reduced compounds delivered to the atmosphere by
the biosphere (CH4, reduced sulfur compounds, and the
products of denitrification).
Because of its secondary economic interest and its
inaccessibility, this area has not been studied intensively,
and its significance in the budget of atmospheric constit-
uents is poorly known.
A research program to obtain needed information on
fluxes from tundra regions should be flexible, starting
with exploratory studies. The results of these prelimi-
nary investigations will then guide further program de-
velopment. Because of the two-phased nature of the
tundra research, the exploratory studies should be initi-
ated as early as possible on a modest scale, with the
extent and nature of future studies left flexible. Aside
from the physical (logistic) problems involved in investi-
gations of this environment, there are geopolitical con-
straints. Initial studies can be performed in North
America and in the Scandinavian countries, but be-
cause of the large area involved on the Eurasian conti-
nent, cooperative participation by Soviet scientists
should be sought.
Because these frequently water-logged environments
are expected to yield significant quantities of reduced
carbon, nitrogen, sulfur, and in some cases, halides, the
species of interest will be the products of denitrification
and sulfate reduction: CH4, NH3, halides (in the vicin-
ity of the ocean), various hydrocarbons, and other re-
duced carbon species in forested areas. Sulfur may be a
OCR for page 60
60
limiting nutrient in many of these areas, so the sulfur
components may be low or absent except within the
range of ocean spray delivery. Analysis of precipitation
may be desirable in later phases, but because of costs and
logistic problems, precipitation sampling should not be
attempted during the first 2 years except where facilities
of cooperating institutions provide the opportunity.
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
Initial investigations at three sites are proposed. Site
selection is based upon a compromise of factors includ-
ing the likelihood that these sites will yield information
representative of significantly large areas, factors of cost
and logistics, and the probable availability of cooperat-
ing individuals and institutions. The suggested sites in-
clude:
TABLE 5. 1 Surface Areas, Net Primary Productivity, and Phytomass of Terrestrial Ecosystems of
the Biospherea
Ecosystem Type
1. Forests
Tropical humid
Tropical seasonal
Mangrove
Temperate evergreen/coniferous
Temperate deciduous/mixed
Boreal coniferous (closed)
Boreal coniferous (open)
Forest plantations
2. Temperate woodlands (various)
3. Chaparral, maquis, brushland
4. Savanna
Low tree/shrub savanna
Grass-dominated savanna
Dry savanna thorn forest
Dry thorn shrubs
5. Temperate grasslands
Temperate moist grassland
Temperate dry grassland
6. Tundra arctic/alpine
Polar desert
High arctic/alpine
Low arctic/alpine
7. Desert and semidesert shrub
Scrub dominated
Irreversible degraded
8. Extreme deserts
Sandy hot and dry
Sandy cold and dry
9. Perpetual ice
10. Lakes and streams
11. Swamps and marshes
. temperate
Tropical
12. Bogs, unexploited peatlands
13. Cultivated land
Temperate annuals
Temperate perennials
Tropical annuals
, ~ . . .
roplca. . perennla. .s
14. Human area
Total
aAnnual average values.
hOfwhich 40 percent (or 0.8 x 1 on m2) productive.
SOURCE: Adapted from Ajtay et al. (1979).
Surface
Area
(1012 m2)
31.3
10
4.5
0.3
3
3
6.5
2.5
1.5
2
2.5
22.5
6
6
3.5
7
12.5
5
7.5
9.5
1.5
3.6
4.4
21
9
12
9
8
1
15.5
2
2
0.5
1.5
1.5
16
6
0.5
9
0.5
2b
149.3
NPP
DM
(g2/yr)
2300
1600
1000
1500
1300
850
650
1750
1500
800
2100
2300
1300
1200
1200
500
25
150
350
200
100
10
50
o
400
2500
4000
1000
1200
1500
700
1600
500
895
.
Total
Production
DM
(lOls g)
48.68
23
7.2
0.3
4.5
3.9
5.53
1.63
2.62
3
2
. 39.35
12.6
13.8
4.55
8.4
9.75
6
3.75
2.12
0.04
0.54
1.54
3
1.8
1.2
0.13
0.08
0.05
o
0.8
7.25
1.25
6
1.5
15.05
7.2
0.75
6.3
0.8
0.4
133.0
L. .
1vlng
Phytomass
DM
(10 g/m )
42
25
30
30
28
25
20
18
7
7.5
2.2
15
5
2.1
1.3
0.15
0.75
2.3
1.1
0.55
0.06
0.3
o
0.02
7.5
15
5
0.1
5
0.06
6
4
3.75
OCR for page 61
CRITICAL PROCESSES
TABLE 5.2 Twenty-two Biomes, Sites, and Processes and Twelve Gaseous Species or Groups
61
NOx RON NH3 N2O CH4 CO RS H2S COS RX CH3M NMHC
Sterile ocean
Productive ocean
Tropic wet
Tropic dry
Desert 1
Desert 2
Desert 3 productive
Wet subarctic
Dry subarctic
Tundra
Tropic agriculture
Temperate agriculture
Rice agriculture
Temperate evergreen
Temperate mixed
Temperate grassland
Wetlands
Inland waters
Sewage sources
Feedlots
Coastal shelf
Biomass burning
X X
X
X X
X ' X
X X
X
X
X
~X
X
X
X
X X
X
X
X
X
X
X
X X
X
~X
X
X
X X
X
X
X
X 1
X
X
X
X
X
X
X
X
X
X
X
~ ~X
X X X X
X
X
X
X
X X
X
X
X
X
X
X
X
X
X
X
X
X X
X X
~X X
X X X
X X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
~ X
X
NOTES: An "X" indicates that there is some reason to expect a significant source; a circled X indicates especially strong interest or evidence. The symbol
"R" represents an organic group, RX means a methyl halide, and CH3M means a methylated metal. " 1 " includes termites and ruminants.
1. Alaskan arctic seaboard, vicinity of Point Barrow
or Prudhoe Bay;
2. Interior Alaska, vicinity of Fairbanks;
3. Hudson Bay area, vicinity of Churchill.
Species to be measured in the initial studies should
include CH4, N2O, H2S, dimethyl sulfide [(CHINS],
CO, and volatile halides. In later studies, such species as
NH3 and volatile metals (e.g., mercury, arsenic, and
selenium) should be measured.
Ideally, gradient measurements should be made for
flux determinations. During this initial phase, portable
equipment with sufficient sensitivity will probably not
be available, so bulk samples should be collected for
analysis at cooperating laboratories.
Samples of opportunity should be collected, prefera-
bly during the spring ice breakup and during midsum-
mer. Where possible, samples should be collected on a
regular schedule throughout the year.
Temperate Forests
Observations on temperate forests will be of greatest
value if accomplished at sites where ongoing research
and monitoring programs will provide supportive infor-
mation. A number of these are available within the con-
tiguous United States. They provide representative sites
for the forest types of interest.
Forest Type
coniferous
Sierran mixed
conifer
Southern
.,
coniferous
Mixed hardwood
Possible Sites
Pacific Northwest Oregon State collaboration with OSU
School of Forestry
Sequoia National Park
(California) collaboration with UCSB
and UCB scientists
Tennessee collaboration with ORNL
scientists
Hubbard Brook (New Hampshire)
Analytical Protocol
This portion of the study could be done at different
levels of intensity, but for the information needed, a
rather elaborate and detailed (including micrometeoro-
logical information) approach is most desirable. This
approach would provide boundary layer gradient infor-
mation on volatiles of interest, which, in turn, would
make possible the estimation of emission and absorption
rates. Instruments of the resolution and sensitivity
needed for such a study exist, but they have not been
applied specifically to any study ofthis sort. Initial appli-
cation will emphasize the development of appropriate
procedures and the calibration of instruments and pro-
cedures, possibly at only one or two of the possible sites.
As procedures are refined and important data are ob-
tained, the direction and intensity of program develop-
ment will become known.
OCR for page 62
62
Both laboratory and field studies of processes are de-
sirable, and they involve both the canopy trapping and
the gradient measurements cited above. Procedures for
canopy trapping and bench analysis are at a reasonable
state of development.
Molecular Species of Interest
NH3, CO, and NMHCs (including terpenes and
other hydrocarbons) are of most interest. N2O,
NOx, CH4, and reduced sulfur compounds probably are
not present in significant quantities, but should be ex-
amined.
Topical Areas (Forests and Savannas)
Tropical continental areas, both wet forested regions
and drier sites such as savannas, are probably major
sources of a variety of trace gases important in tropo-
spheric chemistry. The potential importance of these
regions stems from their geographical size, biological
productivity, anoxic environments (in wet areas), and
high turnover rates (largely by insects in dry areas).
According to the criteria we have adopted in identifying
biomes of potential importance, measurements of bio-
logical emissions from both tropical wet and dry areas
deserve a high priority.
In tropical forests, investigations are needed on the
fluxes of various nonmethane hydrocarbons, CH4, CO,
N2O, and volatile sulfur-containing gases. Volatile spe-
cies containing metals (probably methylated) should
also be sought, and the potential for regional CO pro-
duction from hydrocarbon oxidation must be explored
through measurements and photochemical modeling. A
variety of specialized approaches must be utilized: air-
borne instruments must be deployed to determine the
concentrations and fluxes of these gases above the forest
canopy, sampling of emitted species from individual
trees and plants is required, and airborne studies of the
photochemical and cloud-mediated transformations in
the forest plume must be undertaken. The latter studies
would address questions related to the production and
destruction of photochemical oxidants, e.g., 03, perox-
ides, NOx, and CO. Similarly, investigations focused on
cloud-water chemistry in these regions will result in a
more complete understanding of the origins of a variety
of organics and acids in precipitation collected in heavily
forested areas remote from populated or industrialized
centers. Because biological systems such as forests can
act as sinks as well as sources for trace species in the
troposphere, a final recommendation is for measure-
ments of deposition of these species to tropical forests
particularly of potential nutrients such as NH3 or am-
monium ion (NH4 ), NOx or wet nitrate ion (NO3 ), SO2
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
or wet sulfate ion (SOT), 03, CO2 (to deduce exchange
rates), and trace metals.
The measurement of chemical fluxes from and to the
tropical forests will be difficult. Base facilities must be
established and local scientists with similar interests
must be involved. Before a large, coordinated expedi-
tion is undertaken in the tropics, methods for measuring
concentrations and fluxes should be tested in more ac-
cessible forests, for example, in North America. Be-
cause qualitatively different emissions are likely to dis-
tinguish tropical from temperate forests, investigations
in both regions will be required.
Dry tropical areas also display high cycling rates for
nutrients. Although they store less material in their
shrubs and grasses than is found in the wood of tropical
forests, the biological material of savannas has a shorter
lifetime and has higher nitrogen/carbon and sulfur/car-
bon ratios than hardwood. From the rapid turnover
rates, the chemical composition of the material, and the
present data that suggest a large role for herbivorous
insects, we conclude that significant volatile emissions
are likely from certain dry tropical areas. In particular,
there is potential for large emissions of CH4, CO, CO2,
nonmethane hydrocarbons, N2O, and possibly methyl-
ated metals. Initial measurements should focus on sites
and processes that concentrate nutrients (termite colo-
nies, for example), but the large land areas involved
leave room for significant emissions from lower intensity
sources distributed over large areas.
The most difficult task will be to estimate total emis-
sion rates from tropical forests. The general methods
mentioned above and the use of meteorological towers
are envisioned. About six full-time scientists and techni-
cians would be needed. Measurement of chemical depo-
sition to the tropical forests would require sustained ob-
servations over a period of at least several months, both
near the top of the forest canopy and at ground level for
useful interpretation. In the dry tropical areas, wet ver-
sus dry season differences may be marked. Two separate
seasonal investigations or one longer investigation
stretching through the wet and dry seasons would be
required. Four full-time scientists would be needed for
this study.
Topical Areas
Two distinct kinds of tropical blames are identified for
concentrated research on biological sources important
in tropospheric chemistry. These will be termed "wet
tropics" and "dry tropics. " The potential importance of
each of these environments and relevant investigations
in each are outlined below. Separate discussions are
presented for biomass burning and rice agriculture be-
low.
OCR for page 63
CRITICAL PROCESSES
Wet TropicalAreas
Wet tropical areas are usually covered by tropical for-
ests. These are evergreen or partly evergreen, they are
frostfree and have average temperatures of 24°C or
higher and rainfall rates of 100 mm or more per month
for 2 of every 3 years. About 1.6 x 107 km2 of the earth's
surface enjoys this climatic range, and (0.9 to 1 .1) x 107
km2 is actually covered by tropical forests now. These
tropical forests contain about 60 percent of the global
biomass, but 1 to 3 percent of it is permanently defor-
ested annually, usually for agriculture, timber harvests,
cattle grazing, and firewood. There are perhaps several
million species of life in these forests, but only about
50O, 000 have been described. Barring irreversible cli-
matic changes, it is possible that similar floristic species
could be regenerated in perhaps 50 years after cutting.
Approximately 80 percent of this biome occurs in only
nine nations, five in South America and four in South-
east Asia. These forests cycle nutrients rapidly through
microrhizial (root) systems, and there are few inorganic
reserves in their highly leached soils. Other indicators of
the potential importance of tropical forests include their
high net primary productivity, their potential~trigger
effects of large changes, scientists' relative ignorance
about them, and the existence of many anoxic sites.
Several climatic effects could be triggered by changes in
tropical forests because they mediate regional hydro-
logic cycles. Cleared areas are more susceptible to pro-
longed droughts and to much more erosion and flooding
in wet periods than forested areas. A1SO7 these forests
store CO2 and present a darker surface to sunlight than
cleared areas.
Potentially large emissions of gaseous hydrocarbons,
N2O, CH4, reduced sulfur compounds, and possibly
methylated metals and CO (see below) can be released
from tropical forests. This potential arises from the prev-
alence of anoxic sites in saturated soils and, during noc-
turnal hours, in shallow surface waters. The largest po-
tential source of CO could be the oxidation of biogenic
hydrocarbons like isoprene (C5He), although several
steps in the gas-phase oxidation of this and other hydro-
carbons are not dear at this time. Effects of these emis-
sions on the acidity of regional watersheds and rainfall,
principally through oxidation of organics to formic and
acetic acids, are likely and need investigation.
D?:y TropicalAreas
In dry tropical areas there are likely to be globally
important biological sources of key gases. For example,
in tropical savannas there is generally high net primary
productivity although little of the fixed carbon enters
into long-lived tissue (wood). The indigenous grasses
63
and shrubs are short-lived and are recycled by animals,
largely insects. The chief species among these are ter-
mites and ants; in some areas, termite mass densities per
unit area exceed the highest mass densities of grazing
animals anywhere. The potential for large emissions of
CH4, CO2, CO, and perhaps other species from ter-
mites, whose digestion is fermentative, is quite large.
A1SO7 more speculatively, emissions of N2O and NOX
gases are possible because nitrogen-fixing bacteria are
known to live in termite guts. The roles of these crea-
tures and of the relatively unexciting but extensive dry
tropical areas in nature's atmospheric chemistry are ripe
~ . . .
tor ~nvest~gat~on.
Coastal Marsh, Estuary, ancl Continental Shelf
Environments
Coastal ecosystems are characterized by high biologi-
cal productivity, active chemical and physical exchange,
and transport driven by freshwater runoff, tidal forces,
and wind-driven circulation. Existing data on emission
of reduced sulfur gases, CH4, and N2O indicate that
specific habitats in the coastal environment may be in-
tense sources. Thus, though the areal extent of these
sources may be relatively small, their contributions to
the atmospheric budgets of certain reduced gases may
be considerable.
An additional consideration for placing priority on
these environments is that higher fluxes and concentra-
tions of many chemical species of interest place less de-
mand on instruments (detection limits, response time,
etc.~. For gases such as N2O, CH4, CO, COS, CS2,
(CH3~2S, SO2, and a few others, it is reasonable to pro-
pose initiating immediately a f~eld research program
emphasizing basic processes of gas production from
these source areas and their exchange with the atmo-
sphere. For more reactive chemical species such as NH3,
NO, and volatile metals, existing technology is probably
inadequate for quantitative biosphere-atmosphere ex-
change studies even in intense source areas. For almost
all reduced gas species, measurement technology is cur-
rently inadequate for studying very low-level sources
and sinks.
The primary objective ofthese studies is to develop an
improved quantitative understanding of the processes
that control the production and consumption of bio-
genic gases in coastal wetland and aquatic environments
and their exchange with the lower troposphere. When
this information is available, it should be possible to
extrapolate to regional and global flux estimates with
supporting data from remote sensing and other geo-
graphical and meteorological data bases.
The proposed studies will require measurement of a
wide range of biological, physical, and meteorological
OCR for page 64
64
variables at a selected set of coastal habitats. Five specific
habitats of importance are (1) salt marshes, (2) man-
grove swamps, (3) sea grass, (4) kelp, and (5) exposed
mud and soil surfaces. Variables that can influence trace
gas production and consumption in soils and sediments
and their exchange with the atmosphere include biologi-
cal community composition and dynamics, nutrient
quantities, inputs of energy from different sources, re-
moval of wastes by tidal forces and river runoff, temper-
ature, salinity, pH-Eh, sediment physical properties,
and wind stress on water and soil surfaces.
Previous studies have demonstrated high variability
in time and space. A quasi-continuous research pro-
gram at each coastal habitat of interest will be required
to accomplish the objectives of this experiment. One
efficient way of conducting studies on biogenic gas
sources in coastal environments would be to use NSF-
LTER (Long-Term Ecological Research) sites for the
proposed measurement program. Ongoing activities at
these sites would provide supporting biological, mete-
orological, and geochemical data required for the com-
prehensive gas production and exchange studies pro-
posed here.
Point Sources
There is a class of potentially important biological and
surface sources that are either intense and spatially
small, or qualitatively distinctive in their emissions. Ex-
amples include lightning, industrial emissions such as
combustion or waste plumes, volcanoes, and animal
feedlots. A further source is biomass burning, an activ-
ity that is widespread in certain regions at sites whose
exact locations vary from year to year.
Biomass Burning
Although not a biome but a process, biomass burning
is included in the present discussion because of its great
potential importance and distinctive characteristics.
Qualitatively, biomass burning may be regarded as a
type of nonindustrial pollution. Many types of biomass
burning combine to yield a large total of biomass burned
annually. For example, biomass burning is used to clear
tropical forests for agriculture, to prepare forested areas
for settlements, and to dispose of agricultural wastes
(e.g., sugar cane). Large quantities of biomass are
burned as fuel in industries, for individual human
needs, and in wildfires. Recent estimates of the annual
global area involved in biomass burning range from 3 to
7 x 1 o6 km2, with estimates of the total biomass burned
ranging from 4400 to 7000 Tg/yr. Although total bio-
mass burning quantities are probably uncertain to
within a factor of 2 or 3 and vary from year to year, they
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
are almost certainly significant in the global atmo-
spheric carbon cycle and probably in other cycles as
well, e.g., oxygen and nitrogen. Ecologically, pro-
nounced changes accompany deforestation through bio-
mass burning, e.g., in flora, soil structure, and surface
hydrology. Much biomass burning for deforesting oc-
curs in areas that are not well characterized ecologically.
Surface albedo values, surface winds, and turbulence
are also affected. Certain types of biomass burning also
produce charcoal, thus effectively constituting a CO2
sink.
From a physical and chemical point of view, biomass
burning is a high-temperature process that is dramatic
both in quality and in quantity. Large amounts of mate-
rial are transformed, mobilized, and volatilized quickly.
Partially combusted particles become airborne, and a
wide spectrum of gases are produced in the flames and
through the process of smoldering. Gases containing
carbon, hydrogen, oxygen, nitrogen, sulfur, halogens,
phosphorus, and trace metals are involved. Many of the
gaseous species so produced are highly reactive photo-
chemically, but stable gases like CO2, CH4, N2O, and
CH3C1 are also generated. The photochemically active
species are known to give rise to rapid O3 production
and probably yield other photochemical smoglike spe-
cies, e.g., peroxyacetylnitrate(CH3COO2NO2 (or
PAN)) Not surprisingly, a number of carcinogenic sub-
stances are also produced in biomass burning; gases
include benzene (C6H6) and toluene (C7He), and re-
lated airborne solids are certain to be found. Further, it
now appears that several oxygenated hydrocarbons
from biomass burning yield organic acids in sufficient
quantity to acidify precipitation and groundwaters re-
gionally.
There is a great deal of fundamental research to be
done on the atmospheric chemistry effects of biomass
burning. The full spectrum of compounds injected into
the atmosphere in this way needs description. Quantita-
tive production rates of C H4, C2H6, and other alkanes,
N2O, C H3C1, aldehydes, ketones, nitrogen oxides,
N H3, HCN,C H3CN, oxides of sulfur, CO, CO2, H2,
and several other species must be determined. Methods
need to be developed to quantify the various production
and atmospheric injection rates; simply ratioing each
species to CO2 and deducing CO2 emission rates might
not suffice. Fortunately, some existing data suggest that
the relative yields of some key gases (alkanes, alkenes,
aldehydes, ketones) in temperate and tropical forest fires
are not terribly different, so we can reasonably propose
to concentrate initial research in temperate latitude ar-
eas where logistics are less of a concern. For example,
there are controlled (or prescribed) forest fires managed
by experts in Georgia and Oregon that might be suitable
for several studies. Obviously, certain distinctive fea
OCR for page 65
CRITICAL PROCESSES
lures of tropical forests (e.g., smoldering in damp ar-
eas), tropical agricultural products (e.g., sugar), and
tropical grasses (e.g., the high cyanide content of sor-
ghum) will eventually require targeted research expedi-
tions. Global considerations such as for long-lived gases
(CH4, CO2, N2O, CH3Cl, and COS) will also demand
that the size of areas (biomass) being burned annually be
quantified region by region.
Lightning
The most dramatic types of atmospheric electrical
discharge undoubtedly produce certain gases and de-
stroy others in their intense pulses of thermal and optical
energy. Spectroscopic and chemical analyses have de-
tected many interesting products of lightning. The sin-
gle most pressing question today concerns the amount of
fixed nitrogen (as NO) that is produced annually by
atmospheric electrical discharges. Estimates of this
quantity vary widely; they cover the range from major,
i.e., comparable to combustion production of NOX, to
smaller but significant fractions of this quantity. Key
uncertainties are whether laboratory discharges simu-
late the full range of atmospheric electrical phenomena
adequately and the actual frequency of atmospheric dis-
charges. Microdischarges near pointed biological sur-
faces (e. g., pine needles) might also have important con-
sequences.
Volcanoes
By mass fraction, the principal emissions from volca-
noes are H2O and CO2. A variety of other gaseous and
particulate substances is also emitted in quantities that
are potentially significant to the regional and global at-
mosphere. These include SO2, several volatile heavy
metals, and ash particles. Through large explosive
events, volcanic inputs and impacts can be enormous if
short-lived. By their nature, these effects defy predic-
tion, and once they occur, they defy averaging that is,
it is not easy or particularly meaningful to compare
volcanic emissions to annual averages from other
sources.
Animal Feedlots
Volatile losses of feed nitrogen (principally as NH3,
R-NH2, and possibly N2O and NO) and organic sulfur
compounds from animal feedlots are appreciable, at
least when expressed as a fraction of the feedlot's animal
waste and possibly in absolute terms. In the United
States, perhaps 40 percent of all fertilizer nitrogen is
consumed by cattle, and the average length of time that
cattle spend in feedlots or other concentrated popula
65
lions is high. It is likely that regional impacts of these
emissions are significant, especially for levels of gaseous
NH3, particulate NH4, and NH4 in precipitation. If so,
the atmospheric transport of NH3 represents an air-
borne fertilizer distribution system.
Industrial Emissions
This group includes combustion products, wastes
from chemical production processes, mineral, gas and
oil exploration and refining, burning or processing of
wastes, and losses of solvents. As sources of atmospheric
gases and particles, these processes can be distinctive
both in kind and size, for example, by emitting unnatu-
ral substances or natural ones in quantities that are
somehow comparable to natural cycling rates. For sev-
eral clearly important gases like NOX species and SO2,
emission inventories are available for most industrial-
ized countries, and some of these have been prepared
with good spatial resolution (100 km x 100 km).
Changing industrial intensities, processes, and practices
dictate that these emission inventories will require up-
dating.
Rice Agriculture
The potential importance of rice paddies as sources of
atmospheric chemicals might not seem obvious. Several
~ - ~ . . .
lines ot 0 Elective reasoning and preliminary field data
combine to argue strongly in this direction, however.
First, a number of our criteria (expressed above) indi-
cate that rice agriculture represents a globally significant
biome. These include physical area, reasonably high
primary productivity rates, and the anoxic character of
rice paddy soils. As a principal staple in world food diets,
rice is extremely important in world agriculture. Ac-
cordingly, large areas are cultivated- 1.3 x lo6 km2 in
the late 1960s and 1.45 x 106 km2 in 1979. Further, in
many countries new emphasis has been placed on multi-
ple cropping through improved irrigation. Two crops
per year is becoming the norm in tropical areas. One
practical result of this is that rice paddy soils are under-
water for perhaps 8 months instead of 4 months annu-
ally, and the more negative redox potential of water-
covered soils allows more reduced gases like CH4, NH3,
and N2O to form. Indeed, the rice paddy environment is
ideal for the evolution of a number of volatile species
containing nutrient elements. The soils are oxygen-poor
(strongly reducing), and they are nutrient-rich, often
through fertilization.
Direct indications of the impact of rice paddies on the
global atmosphere have centered on CH4. In the early
1960s, a Japanese scientist showed that paddy soils,
when cultured in the laboratory, released copious quan
OCR for page 83
CRITICAL PROCESSES
radical is believed to be reaction with O2 to form a
peroxy radical or, alternatively, SO3 and HO2, i.e.,
o
H-O-S + O
o
o
S = 0 + HO2
o
The peroxy species may react with NO or HO2 or per-
haps hydrate as a result of collisions with H2O. The only
certainty in this chemistry now appears to be that, like
SO3, the final product is some form of sulfuric acid. The
latter species is rapidly removed from the gas phase by
various heterogeneous processes.
In each of the above systems (e.g., NMHC, NH3,
and the sulfur compounds), the mechanism following
the initiating step leading to the formation of a final
oxidized product is unknown. The absence of this ki-
netic information reflects, to a large degree, the absence
of adequate methodologies to study the kinetics of poly-
atomic free radical species. New kinetic information will
be essential to achieving an acceptable level of under-
standing of these important oxidative atmospheric path
ways.
HOMOGENEOUS AQUEOUS-PHASE
TRANSFORMATIONS
The aqueous phase is most frequently considered in
the context of physical removal processes. Like the gas
phase, however, this medium also encompasses exten-
sive chemical transformations And, like the gas phase,
these chemical transformations are oxidative in their
chemical nature and involve some of the same reactive
agents i.e., OH, HO2, and 03, although the aqueous
phase is far more complex in its chemistry than the gas
phase. Not only are there a large number of single-step
elementary-type reactions to contend with, but there are
also numerous fast equilibria. Furthermore, this chem-
istry involves the reactions of neutral free radicals, free
radical ions, On-free-radical ions, and nonradical, non-
ionic, reactive species such as H202 and O3.
Making this chemistry still more complex is the fact
that the aqueous phase is distributed in the atmosphere
in the form of a broad spectrum of aqueous aerosols.
The two most general classes of liquid aerosols may be
identified as (1) those found in clouds or fogs, and (2)
those present under clear air conditions. In the first case,
the most important size range is 2 to 80 ~m, although
rain droplets up to a few millimeters can be found. The
second category encompasses particles ranging from the
size of critical clusters (10 angstroms) up to a few mi-
crometers. The number density of aqueous aerosols as a
function of size is also highly variable, being critically
83
influenced by exact environmental conditions. How the
chemistry of aqueous aerosols differs as a function of size
is currently one of many poorly understood characteris-
tics of these species.
Traditionally, the approach taken in unraveling this
chemistry has involved studies of closed chemical reac-
tor systems in which only two or perhaps three major
chemical species are added to solution reactors. (In
many respects, these studies have their analogue in gas-
phase smog chamber investigations.) This has been par-
ticularly true of studies designed to elucidate the oxida-
tion pathways of SO2 and nitrogen oxides. In one of the
most extensively investigated systems, involving aque-
ous SO2 mixtures, added oxidizing agents have in-
cluded O2-saturated solutions with and without added
metal ion catalysts, O3-saturated solutions, and H2O2
solutions. The qualitative as well as semiquantitative
data generated from these investigations have shown
that each reaction system could potentially be important
in the aqueous-phase oxidation of S(IV) to S(VI). All
show some pH dependence, but of these the O3 system
appears to have a particularly high sensitivity to changes
in pH level (see Figure 5.10~.
For the most part, mechanistic details on the S(IV) to
~o-6
1010 ~
10~1 ~
4 4.5
1 1 1
5 5.5 6
pH
7~
O2 /
Without /
Catalysts /
-
6.5 7 7.5
FIGURE 5.10 Conversion of S(IV) to S(VI). The pH depend-
ence of the reaction rate is for the systems H2O2, 03, and O2.
OCR for page 84
84
S(VI) aqueous chemistry have been lacking. Thus it is
unclear how the bulk chemical conversion rates mea-
sured for the 03, O2/metal ion catalyst, and H2O2 oxi-
dizing systems may be combined for the case of a real
aqueous aerosol environment. There is growing evi-
dence, in fact, that many ofthe same reactive intermedi-
ates (especially free radicals) exist in all three systems.
More recently, an integrated modeling approach has
been attempted on these aqueous-phase systems. In this
approach, as in modeling studies of homogeneous gas-
phase chemistry, the entire reaction system is built up
from a large number of elementary reactions. Sulfur,
nitrogen, and/or carbon chemistries are taken to occur
simultaneously and continuously in time. Illustrative of
the latter approach is the chemical scheme shown in
Figure 5.11. This chemical scheme portrays aqueous-
phase chemistry as being made up of both numerous
very fast equilibria and individual rate controlling ele-
mentary reactions. Chemical intermediates include
both ionic and free-radical-type species. HxO' oxidizing
agents in the system may result from aerosol scavenging
of H202, 03, OH, and HO2 or by the in situ liquid-
phase photolysis of the species H2O2 and O3. The aque-
ous aerosol model may also be expanded to include the
chemistries of NO2 and NO3 as well as soluble carbon
species such as formaldehyde (CH2O).
The use of building block elementary reactions to
construct the complex chemistry of aqueous aerosols
now appears to offer considerable potential. Facilitating
this approach is the availability of a large volume of rate
_~
~ ~ ~ i ~
Cl
FIGURE 5.11 Primary chemical pathways for a cloud droplet
containing reduced sulfur, carbonate, and C1- ion, and a source of
reactive HxOy The symbol ~ indicates fast equilibria, an ~
either an elementary reaction or multistep fast aqueous-phase
process, and the dotted enclosures indicate various types of micro-
chemistries taking place within the larger overall aqueous system.
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
coefficients for elementary solution reactions. Most of
these have been generated over the past 15 years by
radiation chemists using, in particular, pulse radiolysis
techniques. Even so, there remain numerous reactions
of possible importance to this chemistry that still are
without rate constants. Others, which have been mea-
sured, need to be reexamined with more advanced ki-
netic tools, especially with regard to establishing their
temperature dependence.
In all cases, the question may be raised: Are rate
coefficients measured in bulk liquid phase applicable to
the broad spectrum of aqueous aerosols in the environ-
ment? Certainly, there would appear to be a need to
investigate this chemistry under conditions where indi-
vidual aerosol species could be studied as a function of
time. Such studies will challenge the best technology, but
must be viewed as a critical step in advancing the under-
standing of this science.
-
HETEROGENEOUS PROCESSES
Normally, a heterogeneous reaction implies one oc-
curring at an interface between two phases, e.g., gas-
liquid, gas-solid, or liquid-solid. Several interfaces may
be involved in an overall process. Of course, solid-solid
and liquid-liquid interfaces are also possible, but are not
likely to be of great importance in atmospheric chemis-
try.
The study of interfaces, with their interesting and
significant physical and chemical processes, is an old
discipline that has recently reawakened. To appreciate
the role of interfaces in many phenomena, one need only
recognize that, in any multiphase system, communica-
tion between the bulk phases occurs through the surfaces
that connect them. Even when these surfaces make up a
small fraction of the total volume as is usually the case
for particles suspended in ambient air- they may have a
dominant effect. A classical example can be seen in the
phase transition of supercooled water droplets to frozen
droplets (contact freezing) or ice crystals (sublimation
freezing). This fundamental precipitation-producing
process (Bergeron process) is believed to be induced and
controlled to a significant extent by particles with very
specific surface characteristics (ice nuclei).
As in all chemistry and solid-state physics, measure-
ments at the atomic and molecular level lie at the heart of
a satisfying description of surface structure and compo-
sition. Processes that occur at surfaces are described in
terms of the time evolution of reactant, product, and
intermediate structures. Without definition of surface
structures in terms of equilibrium bond lengths and
bond angles, as well as the potential energy functions
that describe their variations, an adequate description of
reactions at the molecular level is impossible. Because
OCR for page 85
CRITICAL PROCESSES
even the simplest heterogeneous reactions are very com-
plex at the molecular level, understanding at this level
requires the application of many complementary exper-
imental and theoretical tools. As a result, attempts to
resolve heterogeneous processes of importance to atmo-
spheric chemistry are still quite rudimentary. An at-
tempt to present a comprehensive picture of heteroge-
neous atmospheric chemistry can be found in various
workshop documents listed in the bibliography.
Most reactions that occur on such surfaces are
thought to be noncatalytic. These include chemical re-
actions in which both phases participate as consumable
reactants, or physical processes involving either trans-
port or growth, or both. The process of absorption or
adsorption is a heterogeneous process. Heterogeneous
catalytic processes normally imply the "conserved" par-
ticipation of the interface material or a species adsorbed
on it. Heterogeneous catalysis requires, among other
things, the demonstration of "turnover" numbers 'far in
excess of unity. The turnover number is essentially the
number of repeated reactions conducted per unit time at
a catalytic site.
Reactions can be heterogeneous overall but locally
homogeneous, as represented by reaction within the
bulb of an aerosol particle where reactants are trans-
ported in from the gas phase. These reactions might
better be termed multiphase rather than heterogeneous
because the reactants react in one phase, although some
originate from another phase.
One special class of heterogeneous reactions is that
Direct Emission or
Produced by Gas
BOX 1
Water Vapor,
Atmospheric
Trace Gases
(NH3, NOx,
H2S, SO2'03'
Unsaturated
Hydrocarbons)
Depletion by
Gas-Phase
Reactions and
Other Sinks
85
referred to as gas-to-particle conversion. These reac-
tions cause the transfer of a chemical species from the gas
phase to an aerosol or liquid droplet suspended in the
atmosphere, or they may cause formation of new parti-
cles.
Figure 5.12 shows a box diagram for gas-to-particle
conversion processes including the following:
1. Homogeneous, homomolecular nucleation (the
formation of a new stable liquid or solid ultrafine particle
from a gas involving one gaseous species only);
2. Homogeneous, heteromolecular nucleation (for-
mation of a new particle involving two or more gaseous
species);
3. Heterogeneous heteromolecular condensation
(growth of preexisting particles due to deposition of mol-
ecules from the gas phase).
The coexistence of homogeneous and heterogeneous
reaction paths governing the distribution of key chemi-
cal species is shown in Figure 5 .13 .
Of the many gas-to-particle conversion processes be-
lieved to occur in the atmosphere, such as those depicted
in Figure 5. 13, one of the most interesting is the conver-
sion of gas-phase SO2 to sulfate. Because this process
generates two hydrogen ions, it often is responsible for
producing acid rain in regions containing high levels of
so2.
Heterogeneous reactions may also be of importance
in aqueous-phase atmospheric chemistry. The potential
for transition metals, commonly found in atmospheric
Radiation and
Other Energy Input
Constant or Time
Dependent
Rate of
Production
Heterogeneous Processes
Coagulation
Sorption
Relative Humidity
Constant or Vary-
ing with Time
Relative Humidity
Constant or Vary-
ing with Time
1
~. Heteromolecular
BOX 2 Heteromolecular BOX 3 . . Condensation
Products of Low Volatility Nucleation Rate . BOX 4
(Formed by Gas- of Production Embryon~c Embryonic
Phase Chemical ~Critical Size Brownian Droplets of
Reactions, Hydrated Gas-to-Particle Coagulation LargerSize
Species Included) Formation)
Mixing and
Coagu ration
BOX 5
Preexisting and Newly Produced Aerosol,
Liquid or Solid (Size Variation
Possible Due to Fluctuations of
Relative Humidity, Coagulation,
Deposition of Trace Gas Molecules, etc.)
FIGURE 5.12 Box diagram for gas-to-particle conversion. The boxes contain the substance, and the arrows describe the process.
OCR for page 86
86
FIGURE 5.13 Gas-phase constituents
and major reaction pathways (solid lines).
Interactions between chemical families are
indicated by dashed lines. Heavy (double)
arrows show key heterogeneous pathways
involving aerosols (A) and precipitation (P)
(Turco et al., 19824.
aerosols and cloud droplets, to function as heteroge-
neous catalysts is relatively well known. Based on their
abundance, chemical form, stable oxidation states, and
bonding properties, one can speculate that iron, manga-
nese, and perhaps copper are most likely to function in
this way (especially in the case of rural or urban atmo-
spheric environments). For an atmospheric reaction
heterogeneously catalyzed by a transition metal, the
rate-determining step is likely to be either reaction at the
catalyst surface or permeation of the reactant through
the organic film that is sometimes observed on aqueous
atmospheric aerosols. Heterogeneous catalysis involv-
ing transition metals is likely to be of little consequence
for species with rapid gas-phase or homogeneous liquid-
phase reaction pathways, but may be significant for
slower processes such as the aqueous-phase oxidation of
so2.
Soot-catalyzed SO2 oxidation may be another impor-
tant mechanism for sulfate formation in the atmo-
sphere. Soot is synonymous with primary carbonaceous
particulate material. It appears that this material is
present not only in urban atmospheres, but also in re-
mote regions such as the Arctic. It is a chemically com-
plex material consisting of an organic component and a
component variously referred to as elemental, graphitic,
or black carbon. Soot has properties similar to those of
activated carbon, which is well known to be a catalytic
surface active material.
The above discussion makes it quite apparent that the
inclusion of heterogeneous processes is essential to
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
A
HO2
~ HN024 p
NO ~
~a' l i ~ H NO3 ~ p, A
H2O2-OHMS 1~ __ NO2 ~ HO2 NO2
A, P I ~ ~-`
----_ A' ,>~` \ NO3 ~ ~ N2O5
! \ O3 ~-O ~ /
I '. ~ L _'' A, P
\~ I'---------______
C H4 CH3 O2 NO2 _ _- ---
CH3:CH3 0 ~ CH2O ~ P
~ CIO
P. A
(CH3 )2S
S_SO-- _~SO2 ' HSO3
C1CS CS HIS ~SO3
CS2 H2S H2SO4
at,
A, P
achieving a complete understanding of the tropospheric
cycles of sulfur, nitrogen, chlorine, carbon, and so on.
However, because ofthe complexity ofthis chemistry, its
quantification in existing models has not yet been satis
factorily accomplished. Thus both extensive laboratory
and extensive field studies are needed.
BIBLIOGRAPHY
Homogeneous Gas-Phase Transformations
Atkinson, R., K. R. Darnall, A. C. Lloyd, A. M. Winer, and i. N.
Pitts, Jr. (1979~. Kinetics and mechanism of the reaction of the
hydroxyl radical with organic compounds in the gas phase. Adv.
Photochem. 11 :375-488.
Chameides, W., and J. Walker (1973~. A photochemical theory of
tropospheric ozone. J. Geophys. Res. 78:8751.
Crutzen, P. J. (1983~. Atmospheric interactions homogeneous
gas reactions of C, N. and S containing compounds. Chapter 3 in
The Major Biogeochemical Cycles and Their Interactions. SCOPE 2 1,
B. Bolin and R. Cook, eds. Wiley, New York, pp. 67-112.
Demerjian, K. L., and J. G. Calvert (1974~. The mechanism of
photochemical smog formation. Adv. Environ. Sci. Technol. 4:1-
262.
Fishman, i., S. Solomon, and P. Crutzen (1980~. Observational
and theoretical evidence in support of a significant in situ photo-
chemical source of tropospheric ozone. Tellus31: 432.
Levy, II, H. (1974~. Photochemistry ofthe troposphere. Adv. Pholo-
chem. 9:5325-5332.
Liu, S. C., D. Kley, M. McFarland, J. D. Mahlman, and H. Levy,
II (1980~. On the origin of tropospheric ozone._. Geophys. Res.
85: 7546-7552.
Logan, J. A., M. J. Prather, S. C. Wofsy, and M. B. McElroy
OCR for page 87
CRITICAL PROCESSES
(1981~. Tropospheric chemistry: a global perspective. I. Geophys.
Res. 86:7210-7254.
Seller, W. (19741. The cycle of atmospheric CO. Tellus 26: 1 16.
Seinfeld, I. H. (Chairman) (1981~. Report on the NASA Working
Group on Tropospheric Program Planning. NASA Reference Publica-
tion 1062.
Wofsy, S. C. `1976~. Interactions of CH4 and CO in earth's atmo-
sphere. Annul Rev. Earth Planet. Sci. 4:441-469.
Homogeneous Aqueous-Phase Transformations
Chameides, W., and D. D. Davis `1982~. The free radical chemis-
try of cloud droplets and its impact upon the composition of rain.
I. Geophys. Res. 87 4863-4877.
Farhataziz, and A. B. Ross `1977~. Selected specific rates of reac-
tions oftransients from water in aqueous solution. III, Hydroxyl
radical and perhydroxyl radical and their radical ions.
NSRDS NBS 59, special publication. Department of Com-
merce, National Bureau of Standards.
Graedel, T. E., and C. I. Weschler (1981~. Chemistry within aque-
ous atmospheric aerosols and raindrops. Rev. Geophys. Space Phys.
19:505-539.
Heiko, B. G., A. L. Lazrus, G. L. Kok, S. M. Kunen, B. W.
Grandrud, S. N. Gitlin, and P. D. Sperry (1982~. Evidence for
aqueous phase hydrogen peroxide synthesis in the troposphere.
J. Geophys. Res. 87:3045-3051.
Junge, C. E., and T. A. Ryan (1958~. Study ofthe SO2 oxidation in
solution and its role in atmospheric chemistry. Quart. I. Roy.
Meteorol. Soc. 84:46-55.
Penkett, S. A., B. M. R. {ones, K. A. Brice, and A. E. i. Eggleton
(1979~. The importance of atmospheric ozone and hydrogen
peroxide in oxidizing sulfur dioxide in cloud and rainwater. At-
mos. Environ. 13:123-13 7.
Pruppacher, H. R., and I. D. Klett (1978~. Microphysics of Clouds and
Precipitation. Reidel, Boston, Mass., pp. 1-714.
87
Ross, A. B., and P. Neta (1979~. Rate constants for reactions of
inorganic radicals in aqueous solution. NSRDS NBS 65. De-
partment of Commerce, National Bureau of Standards, pp. 1-
55.
Scott, W. D., and P. V. Hobbs (1967~. The formation of sulphate in
wafer droplets.~. Atmos. Sci. 24:54-57.
Stedman, D. H., W. L. Chameides, and R. i. Cicerone (1975~.
The vertical distribution of soluble gases in the troposphere.
Geophys. Res. Lett. 2:333-336.
Taube, H., and W. C. Bray (1940~. Chain reactions in aqueous
solutions containing ozone, hydrogen peroxide and acid. I.
Amer. Chern. Soc. 62:3357-3375.
Heterogeneous Processes
fames, D. E., ed. (1979~. U.S. NationalReport, 1975-1978, Seven-
teenth General Assembly International Union of Geodesy and
Geophysics, Canberra, Australia, December 2-15, American
Geophysical Union, Washington, D.C.
Kiang, C. S., D. Stauffer, V. A. Mohnen, I. Bricard, and D. Vigla
(1973~. Heteromolecular nucleation theory applied to gas-to-
particle conversion. A tmos. Environ. 7: 1279- 1283.
Schryer, David R., ed. (1982~. Heterogeneous Atmospheric Chemistry.
Geophysical Monograph 26. American Geophysical Union,
Washington, D.C.
Turco, R. P., O. B. Toon, R. C. Whitten, R. G. Keesee, and
P. Hamill (1982~. Importance of heterogeneous processes to tro-
pospheric chemistry: studies with a one-dimensional model, in
Heterogeneous Atmospheric Chemistry. Geophysical Monograph 26.
David R. Schryer, ed. American Geophysical Union, Washing-
ton, D.C., pp. 231-240.
Vali, Gabor, ed. (1976~. Proceedings of the Third International Workshop
on Ice Nucleus Measurements, Subcommittee on Nucleation, Inter-
national Commission on Cloud Physics, International Associa-
tion of Meteorology and Atmospheric Physics, International
Union of Geodesy and Geophysics, Laramie, Wyo., ~an. 1976.
OCR for page 88
88
WET AND DRY REMOVAL PROCESSES
BY B. HICKS, D. LENSCHOW, AND V. MOHNEN
In simple terms, the tropospheric concentrations of
many species are determined by their rates of emission
and removal. The species source is not usually a single
term; it typically includes contributions from both natu-
ral and anthropogenic sources, as well as in situ produc-
tion. Likewise, the removal rate is made up of both
transformation and transport terms. However, deposi-
tion to the earth's surface constitutes the major sink for
many tropospheric trace gases and aerosols. This sec-
tion discusses removal at the earth's surface, which in
many cases is the major factor limiting tropospheric
trace gas concentrations.
The residence time of aerosol particles ranges from
the order of a day in the atmospheric boundary layer
(the lowest ~ 1000 m of the troposphere, which is closely
coupled to the surface by convection and mechanical
mixing) to more than a week in the upper troposphere.
These residence times suggest that the physical removal
processes are equivalent to chemical transformation
rates of about ~ percent per hour.
It is convenient to differentiate between wet and dry
deposition processes. The process by which falling hy-
drometeors (e.g., rain, snow, and sleet) carry atmo-
spheric trace constituents to the surface is known as wet
deposition. The processes of gravitational settling of
particles and of turbulent transport (and subsequent
impaction, interception, and absorption to exposed sur-
faces) of particles and gases to the surface are collectively
known as dry deposition. There are several potentially
important processes that do not fit neatly into either
category. These include fog droplet interception, scav-
enging by spray droplets at sea, and processes associated
with dewLall.
In practice it is sometimes not possible to apportion
total deposition between wet and dry components.
Moreover, in some circumstances it is clear that wet
dominates dry, while in other cases the opposite appears
to be true.
Such generalities should be modified according to the
chemical and physical nature of the species under con-
sideration. For example, submicrometer particles are
poorly captured by falling raindrops and are ineff~ci-
ently deposited by dry mechanisms. However, they can
enter into the in-cloud nucleation, coagulation, and coa-
lescence processes that precede precipitation. Soluble
trace gases (such as HNO3 vapor) are easily scavenged
by falling raindrops and are rapidly adsorbed at exposed
surfaces.
WET DEPOSITION
Wet deposition constitutes a very intermittent but
highly efficient mechanism for transforming and even-
tually removing trace gases and aerosol particles from
the troposphere. Aerosol particles act as nuclei for the
condensation of water in warm clouds and for the gener-
ation of ice crystals in supercooled clouds. Subsequent
coalescence and accretion lead to a wide range of droplet
sizes, the largest of which initiate the precipitation proc-
ess (e. g., snow and hail). The droplets collect other par-
ticles and gas as they fall, especially when passing
through urban plumes or through a polluted boundary
layer. The terms rainout and washout are sometimes
used to differentiate between in-cloud and subcloud
scavenging, but their use is dropping from favor.
Airborne particles are removed by falling raindrops
below cloud base by much the same physical processes
as cloud droplets scavenge particles within clouds. Scav-
enging efficiencies are related to particle size and chemi-
cal composition. In-cloud nucleation processes scavenge
soluble, hydroscopic particles more easily than particles
that do not have an affinity for water. Likewise, soluble
and chemically reactive trace gases are more readily
removed than less reactive species.
There has been considerable effort to document and
model cloud scavenging systems. For example, a deep-
rooted convective cell feeds on air from the boundary
layer, which is normally the most polluted portion of the
atmosphere, whereas some stratiform cloud systems
form above the boundary layer and thus exist in a rela-
tively cleaner environment. Scavenging characteristics
of the two kinds of cloud systems will certainly be differ-
ent; futhermore, the trace gases and aerosol particles
accessible to them will differ. Because of the differences
between scavenging within clouds by nucleation (and
related cloud-physical processes) and subcloud scaveng-
ing by impaction and adsorption by falling hydromete-
ors, special care must be taken to interpret correctly the
results of experimental case studies. The results of a
study of particle washout by raindrops falling through a
smokestack plume may not necessarily be applicable to
the case of long-range transport and subsequent precipi-
tation scavenging in remote regions.
The manner in which trace gases and aerosol particles
are scavenged by clouds and by falling precipitation
determines the preferred parameterization for inclusion
in models. Steady precipitation falling through a pol
OCR for page 89
CRITICAL PROCESSES
luted air mass will deplete pollutants at a rate propor-
tional to the instantaneous concentration, so that an
exponential decay of concentration with time will result.
Measurements of the composition of precipitation
through the duration of such a simple precipitation epi-
sode will display the same exponential time decay. Thus,
for some short-term studies an exponential "scavenging
rate" is used to relate precipitation quality to air chemis-
try, analogous in form to the decay constant of radioac-
tive decay. This scavenging rate for small particles is
typically of the order of 10-5 to 10-4 per second.
If average concentrations of chemical species in pre-
cipitation are of concern (with respect either to space or
to time), then it is usual to assume a first-order linear
relationship between concentrations of the species of
interest in precipitation and their concentrations in air.
The "scavenging ratio" defined in this way (i.e., the
precipitation concentration divided by the air concen-
tration) is expressed either on a volumetric or on a mass
basis, and sometimes the precise definition is not made
clear. Further confusion arises from the influence of me-
teorological factors, especially precipitation type and in-
tensity, and the frequent uncertainty concerning the rel-
ative contributions of in-cloud and subcloud processes.
The term "washout ratio" is frequently used synony-
mously with " scavenging ratio. "
Early studies of radioactive fallout showed that in-
cloud mechanisms result in highly efficient scavenging
of many types of airborne gaseous and particulate mate-
rial. Contemporary studies of precipitation acidity have
shown that in-cloud reactions can be rapid, and that
experimental determinations of scavenging ratios can
be strongly affected by these reactions. Ambient SO2
can interact with other chemical constituents in hydro-
meteors (e.g., H2O2 and nitrogen oxides) and can be
deposited as sulfate. The role of clouds as sites for accel-
erated chemical reactions is a major emphasis of the
research program described elsewhere in this report.
Evaporation of falling hydrometeors is sometimes
sufficiently rapid that none of the precipitation leaving
the cloud base reaches the ground. This process (virga)
is a familiar example of cloud-related mechanisms for
transforming material chemically and physically and for
relocating it in the troposphere. The overall effect of
clouds that do not rain is not well understood.
An illustration ofthe uncertainty regarding wet depo-
sition processes is the case of SO2 scavenging. It is
known that SO2 is absorbed in rain droplets at a rate that
is strongly affected by the pH of the droplet. This ab-
sorption causes sulfur scavenging to be dependent on all
other factors that influence precipitation acidity, many
of which are not yet known. Temperature is acknowl-
edged to have a strong influence on the rate at which
89
dissolved SO2 is oxidized; the results of scavenging stud-
ies carried out in winter must be expected to differ from
those obtained in summer. Finally, it is certain that scav-
enging characteristics depend on the physical nature of
the precipitation. Most studies to date have been of rain.
Freezing rain, hail, and snow have yet to receive much
attention.
Research conducted on the relationships between
precipitation chemistry and air quality has often been
hampered by the lack of chemical data at cloud height.
There are obvious difficulties involved in using ground-
level air chemistry observations as a basis for calculating
scavenging ratios. Scavenging ratios for materials of
surface origin are likely to be underestimates if ground-
level air concentrations are used in their derivation, be-
cause air concentrations near the surface will generally
be greater than those characteristic of the air from which
the material is being scavenged by precipitation. Simi-
larly, experimental evaluations of scavenging ratios for
substances with sources in the upper troposphere will
tend to be too high if ground-level air concentration data
are used. Unless this source of error is eliminated by
appropriate use of aircraft sampling or remote probing
to measure chemical concentrations in the air that is
being scavenged, there is little hope of resolving ques-
tions regarding the role of synoptic variables and cloud
chemistry.
The mechanism for generating precipitation clearly
affects precipitation quality. If rain falls through a pol-
luted layer of air beneath cloud level, then a first-order
dilution effect results. Thus the concentration of some
soluble trace gas in rain sampled at ground level would
tend to vary inversely with the amount of rain that fell.
On the other hand, if air from the same polluted layer
were drawn into an active orographic cloud scavenging
material from a constant air stream and depositing it in a
steady rain, then the concentrations in the rain would be
far less influenced by the amount of rain that fell. In
general, the relationship between precipitation chemis-
try and precipitation amount is indeed found to lie be-
tween the extremes corresponding to these two concep-
tual examples.
lust as the quality of rain depends on the quality of the
air from which it falls, the total deposition of chemical
species in precipitation is closely linked to the quantity of
precipitation. Precipitation is a highly variable phenom-
enon that cannot be predicted with accuracy. The net
deposition of chemicals associated with precipitation is
more variable and even more difficult to predict. It
seems unlikely that the capability to predict wet deposi-
tion at a single location on an event basis will ever be
developed, since no organized prediction scheme can
hope to reproduce the details of the random factors asso
OCR for page 90
go
elated with the location and intensity of single storm
cells. These deposition "footprints" have been the sub-
ject of some study; first as a consequence of concern
about radioactive fallout, but most recently under the
aegis of acid rain. Precipitation quality recorded during
a single period of uninterrupted precipitation (an event)
will display features corresponding to a cross section
through the event. As a consequence, interpretation of
the fine structure of time-sequence records observed at a
single station is quite difficult, since it is often not possi-
ble to determine which part of the observed behavior is
due to meteorological or air chemical processes and
which is a result of the vagaries of the sampling cross
section. However, the prediction of average patterns
and of statistical variability (both with time and space)
are achievable goals, provided appropriate information
becomes available on the physical and chemical proc
~.
esses ot Importance.
Because the deposition "footprints" of different
chemical compounds in single precipitation events tend
to look alike, comparisons between deposition records of
different chemical species must be expected to yield high
correlation coefficients. Time records of sulfate deposi-
tion in rain at some specific site should be highly corre-
lated with nitrate, for example, without the need to
, .
Imagine some cause-and-effect relationship between
these two species. In this regard, the determination of a
low correlation coefficient may be as informative as de-
tecting an unusually high value.
Recent emphasis on precipitation acidity has tended
to divert attention from the basic questions of precipita-
tion scavenging of particular trace gases and aerosol
particles. High rainfall acidity does not necessarily
mean very high concentrations of dissolved trace species
in the rain, nor does a pH of 7 mean that the rain is
completely free of dissolved material. Precipitation col-
lected at remote sites is usually somewhat more acidic
than expected solely on the basis of equilibrium with
atmospheric CO2 (pH about 5.6) as a result of back-
ground levels of nitrates and sulfates. The worrying
feature of acid deposition over North America, for ex-
ample, is not only its pH but also the concentrations of
chemicals in the solution being deposited.
In contrast to the case of dry deposition, wet deposi-
tion rates can be monitored with existing techniques.
Wet/dry collectors, which protect precipitation samples
from contamination by dryfall processes during periods
between rain events, became popular during the era of
radioactive fallout studies and are now familiar instru-
ments in most deposition measurement programs. The
use of bulk collection devices is discouraged for studies of
long-term wet deposition, because of the considerable
uncertainty about the effect of dry deposition between
. . .
precipitation events.
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
Collection of precipitation for chemical analysis of
trace constituents, although conceptually simple, is sus-
ceptible to many problems. Many trace species of inter-
est have extremely low concentrations, particularly in
the remote regions that are often areas of concern when
studying global biogeochemical cycles. Precipitation
. . . . .. .
samp es contammg these species are easily contaml-
nated during collection and subsequent sample han-
dling. Problems with wall losses in the collection and
storage vessel, biological activity in the samples, loss of
volatile species, and so on, demand that the greatest care
and preparation be taken before undertaking the seem-
ingly simple task of collecting rain for chemical analysis.
FOG AND DEWFALL
Precipitation collection devices fail to provide repre-
sentative data on deposition via fog interception and
dewfall. Fog droplets can contain relatively high concen-
trations of pollutants; the physical and chemical proc-
esses involved are precisely those that contribute to the
in-cloud component of normal precipitation scaveng-
ing. Iffogforms in polluted air, significant deposition vie
fog droplet interception and deposition is likely. It is not
obvious whether this process best fits under the general
category of wet or dry deposition, and this uncertainty
sometimes causes the process to be overlooked.
Studies ofthe acidity of cloud liquid water have shown
that droplet interception by forest canopies can be a
major route for acid deposition. Exceedingly low pH
values have been reported, presumably in circum-
stances (such as high-altitude, stratiform clouds) in
which there is minimal buffering and negligible access to
the trace metals and NH3 compounds that can serve as
neutralizing agents.
It is clear that even uncontaminated fog droplets will
cause dry deposition rates to be modified by wetting
surfaces. Dewfall (and other processes that cause liquid
water to form on exposed surfaces) will modify dry depo-
sition rates in much the same manner, and for some
chemical species net deposition rates can be significantly
affected.
.
DRY DEPOSITION
Dry deposition rates are influenced strongly by the
nature of the surface and by source characteristics. Sur-
face emissions are held in closer contact with the ground
than emissions released at greater altitudes, so that in the
former case concentration loss by dry deposition would
be expected to be greater. Consequently, dry deposition
fluxes tend to be highest near sources, whereas the high-
est rates of wet deposition ofthe same substances may be
found much further downwind.
OCR for page 91
CRITICAL PROCESSES
Dry deposition rates are intimately related to atmo-
spheric concentrations in the air near the surface A
.
first-order linear relationship is usually assumed. The
coefficient of proportionality between atmospheric con-
centrations and dry deposition rates, which is known as
the deposition velocity, clearly depends on the meteoro-
logical conditions, the chemical nature of the substance
in question, and the nature of the surface on which it is
being deposited.
The term "deposition velocity" suggests an analogy
with gravitational settling that is usually incorrect. In
most instances, deposition through the atmosphere is
accomplished by turbulent mixing to within a very short
distance of the final receptor surface, followed by diffu-
sive transfer across a layer of near laminar flow immedi-
ately next to the surface. Turbulent transfer very near
the surface is possibly influenced by the presence of
small roughness features of the surface, electrostatic
forces, and other mass and energy exchanges that are
occurring. Discussion of the relationship between these
(and other) potentially important factors is simplified by
use of a resistance analogy, in which the inverse of the
deposition velocity is viewed as a resistance to transfer in
direct analogy with electrical resistance as described by
Ohm's law. Individual resistances are associated with
each process contributing to the dry deposition phenom-
enon, and these individual resistances are combined in a
network whose structure reproduces the conceptual
linkage between the various contributing mechanisms.
A total resistance to transfer is then evaluated by using
the electrical analog. The analogy is not perfect, how-
ever, it permits the processes involved in trace gas and
aerosol particle deposition to be compared and com-
bined in a logical manner.
Particles already deposited on a dry surface can be
resuspended by wind gusts exceeding some critical value
related to the size and density of the particle. Soil grains
and particles of surface biological origin can be en-
trained in the lower atmosphere under some conditions.
Suitable circumstances are not necessarily unusual. In
arid regions, a surface saltation layer is frequently visible
in strong winds, and it has been demonstrated that such
aeolian particles can be carried into the upper tropo-
sphere by deep convection and transported horizontally
for considerable distances. The generation of particles as
a result of chemical reactions occurring within vegetated
canopies has been postulated as a cause for the blue haze
phenomenon associated with forests in many parts of the
world. Ocean spray is another well-known example of
surface generation of particles. Resuspended particles
constitute another form of atmosphere-surface interac-
tion, thus sharing many of the features normally associ-
ated with dry deposition.
There is considerable scientific disagreement about
91
the mechanisms involved in dry deposition. Models
(such as the resistance models mentioned above) that
combine knowledge of individual processes to simulate
natural phenomena occasionally omit processes that are
sometimes considered to be important. However, all
such models enable a test to be made of scientists' ability
to simulate nature on the basis of their understanding of
its component parts. For some circumstances and for
some chemical species, the most important factors af-
fecting dry deposition have been formulated well
enough to permit fairly accurate modeling. The sum-
mary of the dry deposition of certain chemical species
that follows is based on a contribution to the Critical
Assessment Review Papers on acid deposition, soon to
be released by the Environmental Protection Agency.
SO2. Uptake by plants is largely via stomates during
daytime, but about 25 percent is apparently via the
epidermis of leaves. At night, stomata! resistance in-
creases substantially. When moisture condenses on the
surface, resistances to transfer should decrease substan-
tially. Deposition to masonry and other mineral surfaces
is strongly influenced by the chemical composition of the
surface material. To water, snow, or ice surfaces, deposi-
tion rates are influenced by the pH of the surface water
and by the presence of liquid films.
O3. Dry deposition to plants is much like SO2, but
with a significant cuticular uptake at night and with the
presence of surface moisture minimizing deposition
rates. Deposition to water surfaces is generally very
slow.
NO2. Similar to O3 for deposition to plants, but with a
somewhat greater resistance to transfer. Even though
NO2 is insoluble in water at low concentrations, deposi-
tion to water surfaces might be quite efficient.
NH3. No direct measurements are yet available, but
a similarity to SO2 appears likely.
Submicrometer particles. Deposition to smooth sur-
faces is a minimum for particles of about 0. 5-pm diame-
ter. Deposition velocities increase as particle size in-
creases, until the terminal settling velocity predicted by
the Stokes-Cunningham formulation is reached. Very
small particles are deposited at rates that are controlled
by Brownian diffusivity across a limiting quasi-laminar
layer in contact with the surface. For rougher surfaces,
deposition velocities tend to increase.
Supermicrometer particles. Turbulence can cause
particles to be deposited by inertial impaction and inter-
ception, with deposition velocities greater than the
Stokes-Cunningham prediction. Particle shape is an im-
portant factor.
OCR for page 92
92
Sulfate particles. A value of 0.1 cm/s is often used for
the deposition velocity for sulfate particles. However,
recent experiments have demonstrated that deposition
velocities for sulfate aerosol vary with the roughness of
the surface. Values less than 0.1 cm/s seem appropriate
for snow and ice, and about 0.2 to 0.3 cm/s for growing
pasture and grassland. There is considerable disagree-
ment concerning forests. Some workers use large depo-
sition velocities (approaching 0.7 cm/s), while others
prefer to continue to use the value 0.1 cm/s used in early
transport and dispersion models. Phenomenological
differences appear likely.
~1
Dry deposition to the oceans remains a major un-
known. Data obtained in laboratory experiments on
trace gas exchange between the atmosphere and water
surfaces indicate that exchange rates are limited by fac-
tors associated with the liquid phase, especially with the
Henry's law constant. The deposition of hydroscopic
particles is known to be influenced by their growth upon
entering the region of very high relative humidity near
the water surface. However, the practical significance of
the effect is still being debated. Of major importance is
the fact that exceedingly little information is available
for dry deposition under typical open ocean conditions.
The average wind speed at sea is about 8 m/s, with a
highly disturbed surface and much spray. In such condi-
tions the relevance of experimental data obtained in
laboratory experiments seems open to question. In
some areas of the world ocean, such as the "Roaring
Forties," the surface is sufficiently agitated that the con-
cept of a distinct, identifiable surface between the air and
the ocean becomes difficult to defend. Rather, there is an
interracial layer with properties somewhat like a gas-
liquid suspension. In such conditions, exchange oftrace
gases and aerosol particles between the atmosphere and
the ocean may be quite rapid but bidirectional. Limiting
processes cannot yet be identified with confidence.
Although detailed knowledge of many of the proc-
esses involved is lacking, the ability exists to measure
dry deposition fluxes in some circumstances, for some
substances. Dry deposition to some surfaces can be mea-
sured directly, e.g., in the cases of accumulation on
snowpacks or ice, or on some mineral and vegetative
surfaces. For very large particles, deposition can be
measured by exposing artificial collection surfaces or
vessels since the detailed nature ofthe surface plays a less
important role. However, until recently there has been
little information on the rate of deposition of small parti-
cles and trace gases to natural surfaces exposed in natu-
ral surroundings. In the last decade, methods developed
for measuring the meteorological fluxes of heat, mois-
ture, and momentum have been extended to 03, CO2,
SO2, nitric acid vapor, nitrogen oxides, and various
PART II ASSESSMENTS OF CURRENT UNDERSTANDING
particulate pollutants, with varying degrees of success.
Some of these experiments have been intensive case
studies, using instrumented meteorological towers, and
were intended to identify and quantify factors control-
ling the deposition. Other studies have used instru-
mented aircraft to measure spatial averages of deposi-
tion fluxes over terrain of special interest. None have yet
demonstrated a capability for routine monitoring.
There are essentially two schools ofthought on moni-
toring dry deposition. The first advocates the use of
collecting surfaces and subsequent careful chemical
analysis of material deposited on them. The second in-
fers deposition rates from routine measurements of air
concentration of the trace gases and aerosol particles of
concern and of relevant atmospheric and surface quanti-
ties. Collecting vessels have been used for generations in
studies of dustfall and gained considerable popularity
following their successful use in studies of radioactive
fallout during the 1950s and 1960s. The inferential
methods assume the eventual availability of accurate
deposition velocities suitable for interpreting concentra-
tion measurements.
In the era of concern about radioactive fallout, dust-
fall buckets were used to obtain estimates of radioactive
deposition, especially of so-called local fallout immedi-
ately downwind of nuclear explosions. It was recognized
that the collection vessels failed to reproduce the micro-
scale roughness features of natural surfaces, but this was
not viewed as a major problem because the emphasis
was on large "hot" particles and the need was to deter-
mine upper limits on their deposition so that possible
hazards could be assessed.
Much further downwind, so-called global fallout was
found to be associated with submicrometer particles
similar to those likely to be of major interest in studies of
global tropospheric chemistry. However, most of the
distant radioactive fallout was transported in the upper
troposphere and lower stratosphere, and its deposition
was mainly by rainfall. The acknowledged inadequacies
of collection buckets for dry deposition collection of
global fallout were of relatively little concern because dry
fallout was a small fraction of the total surface flux.
The acknowledged limitations of surrogate-surface
and collection vessel methods for evaluating dry deposi-
tion have caused an active search for alternative moni-
toring methods. In general, these alternative methods
have been applied to studies of specific pollutants for
which especially accurate and/or rapid response sensors
are available. The philosophy of these experiments has
not been to measure the long-term deposition flux, but
instead to develop formulations suitable for deriving
average deposition rates from other, more easily ob-
tained information such as ambient concentrations,
wind speed, and vegetation characteristics. Neverthe
OCR for page 93
CRITICAL PROCESSES
less, several initiatives are under way to develop micro-
meteorological methods for monitoring the surface
fluxes of particular pollutants. Surrogate surface meth-
ods are also being improved. Although these devices
share many ofthe conceptual problems normally associ-
ated with collection vessels, they appear to have consid-
erable utility in some circumstances. It has been shown
that deposition of small particles to surrogate surfaces is
sometimes similar to that of foliage elements. However,
none of the surrogate-surface or micrometeorolog~cal
methods that have been identified to date has been suc-
cessfully demonstrated to monitor the dry deposition of
a pollutant being slowly deposited.
BIBLIOGRAPHY
Beille, S., and A. l. Alshout (1983~. Acid Deposition. D. Reidel,
Dordrecht, Holland, 250 pp.
Engelmann, R. I. (1968~. The calculation of precipitation scaveng-
ing, in Meteorology and Atomic Energy, D. H. Slade, ed. U.S.
Atomic Energy Commission.
Galloway, I. N., and D. M. Whelpdale (1980~. An atmospheric
sulfur budget for eastern North America. Atmos. Environ.
14:409-41 7.
Galloway, I. N., I. D. Thornton, S. A. Norton, H. L. Volchok, and
R. A. N. McLean (1982~. Trace metals in atmospheric deposi-
tion:areviewandassessment.Atmos.Environ.16:1677-1700.
93
Greenfield, S. M. (1957~. Rain scavenging of radioactive particu-
late matter from the atmosphere.~. Meteorol. 14:115-123.
Hales, i. M. (1972~. Fundamentals of the theory of gas scavenging
by rain. Atmos. Environ. 6:635-659.
Hardy, Jr., E. P., and J. H. Harley, eds. (1958~. Environmental
Contamination from Weapons Tests. Health and Safety Laboratory
Report HASL-42A. U.S. Atomic Energy Commission.
Hicks, B. B., M. L. Wesely, and J. L. Durham (1980~. Cntique of
Methods to Measure Dry Deposition: Workshop Summary. EPA-600/9-
80-050. U.S. Environmental Protection Agency, 69 pp. (NTIS
PB81-126443.)
Lindberg, S. E., R. C. Harriss, and R. R. Turner (1982~. Atmo-
spheric deposition of metals to forest vegetation. Science 215:
1609-1611.
Liss, P. S., end W. G. N. Slinn, eds. (1983).Air-SeaExchangeof Gases
and Particles, NATO ASI Series, Series C. Mathematical and
Physical Sciences No. 108. D. Reidel, Dordrecht, Holland, 561
PP.
Owens, I. S. (1918~. The measurement of atmospheric pollution.
Quart. Hi. Roy. Meteorol. Soc. 44:149-170.
Pruppacher, H. R., R. G. Semonin, and W. G. N. Slinn, eds.
(1983~. Precipitation Scavenging, Dry Deposition, and Resuspension,
Vol. 1, Precipitation Scavenging. Elsevier, New York, 729 pp.
Pruppacher, H. R., R. G. Semonin, and W. G. N. Slinn, eds.
(1983~. Precipitation Scavenging, Dry Deposition, and Resuspension,
Vol. 2, Dry Deposition and Resuspension. Elsevier, New York, 731
PP
Sehmel, G. A. (19803. Particle and gas dry deposition: a review.
Atmos. Environ. 14:983-1012.
Shannon, i. D. (1981~. A model of regional long-term average
sulfur atmospheric pollution, surface removal, and net horizon-
tal flux. Atmos. Environ. 13:1155-1163.
Representative terms from entire chapter:
biomass burning
