1 DC
                     . .J79

A Preliminary
Analysis of
Nitrous Oxide (N2O)
Including A
Materials Balance

This document is available through the National
Technical Information Service (NTIS), Soringfield, VA,
22161, telephone t (703) 557-4650.

                     A PRELIMINARY ANALYSIS OF
                         NITROUS OXIDE(N20)

                            JANUARY 1979
                           IN HOUSE REPORT
                US Environmental Protection Agency
                    Office of Toxic Substances
                           401 M St Sw
                       Washington DC  20460
                        Dr C Richard Cothern
                           Project Officer

This report has been reviewed by the Office
of Toxic Substances, EPA, and approved for
publication.  Approval does not signify that
the contents necessarily reflect the views
and policies of the Environmental Protection
Agency, nor does mention of trade names or
commercial products constitute endorsement
or recommendation for use.


1 . LAND 	
a . REVIEW 	
a . REVIEW 	
3 . OCEANS 	
a . REVIEW 	
a . REVIEW 	
D . SINKS FOR N~ 0 	








                                 - 3 -
Executive Summary

     The Clean Air Act Amendments of 1977 have directed that
the Environmental Protection Agency study the effects of and
possible control techniques of and other details concerning
chemicals that deplete stratospheric ozone.  To understand
how gases reach the ozone layer, their transport through the
troposphere must be determined.  The chemical reactions of a
gas in all the atmosphere must be known in order to determine
the possible effects it may have on stratospheric ozone.

     Nitrous oxide  (N~0), a gas in the family often referred
to as NO , is produced both naturally and anthropogenically
and takes part in the stratospheric chemistry that controls
the ozone level.  Nitrous oxide appears not to enter into any
chemical reactions in the troposphere.  The chemistry of nitrogen
oxides (NO ) in the troposphere appears to be reasonably well
known although more study is indicated for some particular

     The nitrogen cycle requires at one point the fixing of
nitrogen from the atmosphere either naturally or industrially-
Then either through microbiological action or combustion,
the gases NO, NO-, N20 or N~ may be released mainly through
the process of denitrification.

     Nitrous oxide appears to be uniformly distributed in the
troposphere with a concentration of approximately 310 ppb.

                                 - 4 -
Its lifetime in the troposphere that best seems to fit the

models is of the order of 100-150 years.

     A number of sources of N20 are well known but their

relative contribution has not been well quantified due to

problems of measurement and due to the wide diversity and

variability of global sources.  Land sources, particularly

wet, low pH soils appear to be the largest contributers to

atmospheric N20.  The rapidly increasing use of fertilizers

represents a source of N?0 that by the end of the century

may contribute as much N-O as do natural sources.

     The contribution of nitrous oxide from freshwater sources

such as rivers and lakes appears to be small.  The relative

input from oceans is in dispute although it appears that the

smaller estimates are the most likely.  Other anthropogenic

sources such as fossil fuel combustion may be significant

sources but more study seems to be indicated for sources

such as fossel fuel plants.

     The only well known and proven atmosphere sinks for

nitrous oxide are reactions in the stratosphere.  It can

react with 0( D) to form NO which depletes the ozone or it

can be photolyzed to form N2.  Other possible sinks have been

suggested but none have been demonstrated as viable alternatives,

                                 -  5  -
     It is important to realize that N-0 is part of a natural

regulatory cycle involving ozone.  The balance of ozone levels

depends on the natural creation and destruction of N-O.  The

effect of human activity may be to upset this balance.  A

rough idea of the relative contributions of various sources to

the atmospheric burden of N20 can be formed from the data presented

in Table I.

                           Table I

Sources                                         Ranges of Values

                                                 (Mt. of N20/yr)+

Marine                                            5-250

land  (soil)                                      10 - 100

fixed nitrogen including fertilizers*             6-38

man made (direct)                                 2-6

freshwater                                        3-40

Total                                            34 -474 (a)

*Fertilizer, represent approximately 1/2 of fixed nitrogen

+Mt. is a million metric tons and is defined in the Appendix

(a)   This range of values corresponds to a residence time of

     45 - 3.2 yr (53)

                                - 6 -

Stratospheric Reactions                           14 - 45

Trospospheric Reactions                            1-5

Soil                                               0 - 600

                                                  15 - 650(b)

(b)  This range corresponds to a residence time between

     100 and 2.3 years

     As can be seen in Table I the fluxes from sources and

sinks do overlap but the range of possible values is so large

as to prevent any definite conclusions as to whether other

sources (or sinks) exist.

     A number of modelers have predicted the possible effects

of an increase in atmospheric nitrous oxide concentration  (due

to anthropogenic processes) on ozone concentration in the

stratosphere. See table II for a typical list of reactions

used in such calculations.  The models include many

reactions some of whose rate constants are not well known.

Also in some cases the concentrations of the reactants are

not well known.  Model calculations before 1978 predicted that

N20 increases would have depleted stratospheric ozone by

2023 roughly 5%.  However, the present model calculations

predict that if N20 flux is doubled  (achievable in the next

                                - 7  -
50 years) the ozone concentration will only be slightly

affected.  If N20 flux is quadrupoled ozone concentration

will decrease by 5%.  A comparable prediction for the

effect of chlorofluorocarbons  (CFC) on ozone concentration

is a depletion of 20% in the next 25 years at present rates

of production.

     The problem with respect to CFC's was easy to solve

because  (a) sources could be readily identified and quanti-

tatively assessed,  (b) the stratospheric reactions were well

understood, and  (c) the solution was easy to implement.

None of the above is true for the ozone-nitrous oxide cycle.

     It is often argued that if a problem is not a crisis

then its study can be postponed.  Such an attitude leads to

poor planning, inadequate study and understanding of the

mechanism involved and sometimes ineffectual control pro-

cedures.  The N20 problem could well become such a crisis

and with a moderate effort over the next ten years a

possible crisis may be averted.

     To study the effect of N^O on stratospheric ozone requires

researchers from a wide variety of disciplines.  If a coordinated

long term and moderate effort to determine, measure and assemble

the diverse pieces of information concerning N20 and its

relationship to stratospheric ozone is not pursued, the likeli-

hood of a crisis appears good.

                               -  8  -
     A number of recommendations have been made for future

studies.  From the present snapshot in time of this rapidly

moving object it can be concluded that the problem could be

a serious one but it is not an immediate crisis.

                                - 9 -
I.  Introduction

     The Clean Air Act Amendments of 1977 directed, among

other things, that the Environmental Protection Agency should

study the state of knowledge and the adequacy of research

efforts to understand the effects of all substances, practices,

processes, and activities which may affect the stratosphere,

especially ozone in the stratosphere.  Most of the ozone in the

atmosphere is found in the stratosphere and protects the earth

by absorbing much of the incoming solar radiation in the wave

length range 290 to 320 nm that contribute to DUV  (biologically

damaging ultraviolet radiation sometimes referred to as BUV

or erythemal).

     The balance of ozone concentration is affected by the

amount of various gases entering or formed in the stratosphere.

The most important of these gases are the oxides of hydrogen

(HO ), the oxides of chlorine  (CIO ) and the oxides of nitrogen
   j£                              H

(NO ).  The compounds NO  include mainly NO, NO-, N-O,-  N00 and
   2C                    X                      4*   £* 31  ^

HNO., which are produced directly by fossil fuel combustion,

high flying aircraft or by the interaction of microorganisms

in soils, oceans, freshwaters and as a result of the use of

fertilizers.  NO  is sometimes referred to as odd nitrogen.

     Other mechanisms may also deplete the ozone layer.  For

example, metals on fine aerosol particles, traces of which may

                                  -  10  -

or may not be catalytically active are a potential ozone depletor.

Research is actively going on in this area.     Other possible

ozone depletors include among others, most of the nonfluorinated


     Models of chemical-kinetic reactions and transport for

chemicals which directly or indirectly interact with ozone have

been developed which consist of a set of mathematical equations.

Most current models involve about 100 chemical interactions,

many interrelated with each other (see for example table II and

reference 2).  Due to limitations of computers, most of the

models are only one dimensional (altitude).  One development

from modeling is the realization that the major chemical,

catalytic cycles cannot be studied separately.  The models do

include the parameters of strength of sources, rate coefficients

of chemical and photochemical reactions,  and mixing coefficients

as a function of altitude.  They also include absorption

coefficients for solar radiation and quantum yields  (fraction

of light that leads to decomposition) with respect to wave

length.  Even with all this sophistication the picture is not

complete.  What is needed is a 3-D model that includes chemical

kinetic reactions and more precise and complete measurements of

the parameters of the model.

     Gases such as N20, F-ll, F-12 and others can absorb thermal

radiation in the infrared region of the electromagnetic spectrum

and this may lead to a greenhouse effect such as that for C02.

                                 - 11 -

                          Table II


                        (Reference 67)
0 + 02 + M = 03 + M

03 + NO = N02 + 02

N20 + 0(1D) = N2 +

N + 0  + NO + 0
OD) + H20 = 20H

0, + OH = H09 + 09
 •J          £    £

03 + H02 = OH + 202

H + 02 + M = H02 + M
       = 20
0 + N02 = NO + 02

N20 + 0(1D) = 2ND
N + NO = N2 + 0

+0(1D) + CH4 = OH + 2H02 + CO
0 + OH = 02 + H

0 + H02 = OH + 02

03 + H = OH + 02

H0  + OH = H0 +
OH + N02 + M = HN03 + M

H0  + OH = H  + H0
OH + HNO., = H20 + N03
             M =
NO + H02 = N02 + OH

OH + OH = H20 + 0

N02 + °3 = N03 + °2

OH + OH + M = H202 + M

+0 + CH4 = OH + 2H02 + CO

O^D) + M = 0 + M

Cl + OC10 = 2C10
Cl + N02 + M = C1N02 + M
NO + 0 + M = N02 + M

H2 + 0(1D) = OH + H
N + 03 = NO + 02
         = OH
CO + OH = H + CO,
Cl +

        = CIO +
 C1 + CH4 = HC1 + 2H02 + CO

Cl + C1N02 = 2C1 + N02
CIO + 0 = Cl + 0.
NO + CIO = N02 + Cl

                             - 12 -
CIO + 03 = Cl + 02 + 02
CIO + N02 + M = C10N02 + M
CIO + CIO = 2C1 + 02
OH + HC1 = H2 + Cl
0 + OC10 = CIO + 02
N + OC10 = NO + CIO
Cl + OH = HC1 + 0
Cl + HN03 = HC1 + NO.,
Cl +
          0(D) = 2C1
          = HC1 + H0
 CH3C1 + OH = Cl + H20 + H02
N02 + 0 + M = N03 + M
          + M =

         = 2N0
                                   CIO + 03 = OC10 + 02
                                   CIO + CIO = Cl + OC10
                                   HC1 + 0(1D)  = Cl + OH
                                   0 + HC1 = OH + Cl
                                   NO + OC10 = N02 + CIO
                                   H + OC10 = OH + CIO
                                   Cl + H02 = HC1 + 02
                                   *CFC13 + 0(1D)  = 2C1
                                   Cl + H_ = HC1 + H
C10N02 + 0 = CIO + NO-
NO + N03 = 2N02
                                             = NO
M =
N2°5 + H2° = 2HN03
o2 = o + o
HN03 = OH + N02
H02 = OH + 0
HC1 = H + Cl
CIO = Cl + DTD)
OC10 = CIO + 0
 CFC13 = 3C1
                                   °3 = ° + °2
                                   N02 = NO + 0
                                   NO = N + 0
                                   H0  = 20H
                                   C10N02 = CIO + N02
                                   CIO = Cl + 0
                                   C1N02 = Cl +
                                   *CF2C12 = 2C1
                                   *CC14 = 4C1

                               - 13 -

 CH3C1 = 2H02 + CO + Cl            N205 = 2N02 + 0

N03 = NO + 02                      N03 = N02 + 0

H00 = H + OH

 Products such as F, H,/ CO, COF- etc. are not followed when

 produced from halocarBons.

 Methyl radical production assumed to yield 2H02 + CO

                              - 14 -
Although the understanding of the possible effects on climate

are not fully understood, an estimate can be made of the magnitude.

Using a one-dimensional model      it has been predicted that

if the atmospheric concentration of C02 were to increase by

25% the global temperature would increase by 0.79°K.  If the

concentration of N?0 were to double after a period of several

years the global temperature would increase by 0.68°K.

                              -  15  -
II.  Troposheric NO
     Atmospheric or air chemistry for NO  involves two major
regions, the troposphere and the stratosphere, where the
kinds of chemical reactions that occur are different.  In the
troposphere, gases are well mixed and homogeneous.  This region
is shielded by stratospheric gases, mainly ozone and water
vapor which absorb much of the ultraviolet light from the sun.
As gases move away from the Earth the altitude increases and the
temperature generally drops until a reversal occurs, primarily
due to the absorption of ultraviolet photons.  The position of
the reversal is called the tropopause and it is the dividing
line between the troposphere and the stratosphere.  In the
stratosphere the mixture of gases is nonuniform and in general
residence times are long if the possibility of chemical reaction
is excluded.
     A major chemical characteristic of NO  is their oxidative
nature as compared to SO  which are reductive.  This leads to
the primary deliterious effect of the NO  cycle in the troposphere
which is often called smog or photochemical oxidations.  Photo-
chemical smog involves the input of nitric oxide, sunlight and
hydrocarbons.  The kinetic chemical reactions that occur in the
troposphere involve such intermediate as N02 and PAN  (see figure
1).  These reactions have been studied since the 1950's when
this problem manifest itself on a significant level in Los

 R = CH3, C2H5, ETC.
                   /VBCOO,,^+R + C02

                       R02 -.-^
                                          DRY DEPOS.T.ON
         N2 + °2
                            Figure 1     Simplified schematic  description of  NO  chemical

                            kinetics in the  troposphere.  Although not inclusive of all
                            possible detail,  the  main pathways can be seen.

                                - 17 -

Angeles.     It has been demonstrated that the chemical compounds

involved vary in concentration throughout the day depending on

the amount of sunlight and input gases.  The intermediates are

the compounds responsible for eye irritation, adverse respiratory

effects, plant damage and the browning of the atmosphere known

as smog.v ' Some of the color is due to adsorption 'of sunlight

by the intermediate products and some is due to polymerizing

of the oxidation products forming aerosols.

     The two major nitrogen oxide constituents in the atmosphere

are NO  (nitric oxide) and N02 (nitrogen dioxide).  The major

oxide, NO is produced by the high temperature combustion process

in the presence of N~ in the air.  Some N02 is produced by

lower temperature 300°K oxidation.  The amount produced is

dependent more on the concentration of NO than 02 in the

     2NO + 0        2N02
     As seen in figure 1 nitric oxide reacts commonly with the

oxidizing agents such as the transient hydroperoxyl  (H02),

alkylperoxyl (R02) and acylperoxyl  (RC002) free radicals.  In

these reactions R represents methyl  (CH3), ethyl  (C2H5) and

higher alkyl groups.

                                  - 18  -

     Nitrogen dioxide plays a very central role in tropospheric

chemistry being involved directly or indirectly in virtually

every kinetic pathway.  It reacts with water and nitric oxide

to form nitric and nitrous acids in the reactions;
     2 N0  + H0         HONO., + HONO
     N02 + NO + H20           2 HONO
     And as shown in figure 1, N0« finds a sink in the reactions

with HO, R02 and RC002.  If R = CH3 then the latter reaction

involves PAN (CH.,C002N02) or peroxyacylnitrate.  This compound

is the mysterious compound X of the 1950's involved in photo-

chemical smog.

     A source of ozone in photochemical smog involving NO  follows

from the photodissociation of N0_;

     N02 + hv (<4300 A)	!>0(3P) + NO

followed by;
     0(3P)  + 0  + M 	y 0  + M

                                 - 19 -

     The ozone is either quickly ozidized by NO or photolyzed
to produce HO;
     The oxidation of N02 by O., leads to the production of
N205 which forms on aerosol or reaches the ground by dry
deposition.  As shown in figure 1 the acids fromed by N02 also
form aerosols and become part of the clouds eventually reaching
the ground by rain out.

     It is generally concluded by researchers of kinetic chemical
reactions in the troposphere that the major reactions and pathways
are now "reasonably well" understood.     However some reactions
have been recommended for future studies.

     One majo^r area requiring further study is the reaction of
longer chain R02 with NO and N02 as a function of temperature
and pressure.  The chemistry of PAN, particularly for higher
analogues such as PPN need to be investigated.  Particularly
information is required concerning thermal decomposition.
     The reactions of NO., that require further investigation

                                 - 20 -
     N03 + hv	> NO + 02
as well as the reaction
     N2°5 + H2°
     In summary, NO  tropospheric chemistry involves some

mechanisms which require further study.  Besides those mentioned

above research is needed in areas such as aerosol production

(heterogenous chemistry) and acid rainfall.  However those

processes that are related to chemicals involved in ozone

depletion appear to be well enough known that the major

attention should be focused on what happens in the stratosphere

and how the chemicals that get there are produced.

                                 - 21 -
     As shown in Table III the NO  catalytic cycle is a major

cause of depletion of stratospheric ozone.  One of the gases

involved in stratospheric NO  chemistry that is of concern

with regard to ozone depletion is nitrous oxide (N20).
                          TABLE III
Approximate mechanisms for the Depletion of Stratospheric
Ozone * '   (average for the region from 25 to 40 km in altitude) .
                                   Percent of 03 destruction rate

Transport to the Troposphere

Hydroxyl and Hydroperoxyl


NO  Catalytic Cycle

CIO  Catalytic Cycle




     As with many of the atmospheric reactions which interrelate

different chemical cycles, the CIO  and NO  cycles are so
                                  H       A,

related.  For example, chlorofluoromethane decomposes by photolysis

                                - 22 -
in the stratosphere releasing chlorine which consumes ozone

by the reaction;
     Cl + 03	> CIO + 02
But the abundance of CIO may be influenced by
     CIO + NO	^ Cl + N02
and by
     CIO + N02 + M	>C1N03 + M
     CIO + HO 	5"  H02 + Cl
     Nitrous oxide is produced naturally by microbial action

on any nitrifiable substance and from anthropogenic sources

among which are fertilizers and burning of fossil fuels.  It

has been used as an anesthetic and for that use has been

called laughing gas.

     A recent Study     suggests that nitrous oxide  (laughing

gas) may have some direct toxicological effects on anesthesiol-

ogists and dentists.  This conclusion is based on largely











     Figure 2     Relationship between nitrification, denitrification
     and fixation as they relate to valence and compound(reference 42).

                                -  24  -

circumstantial evidence.   However there is some evidence that

the gas increases the fetal death rate and may cause leukopenic


     Animals and plants cannot use nitrogen directly in the

form of N« to produce nitrogen compounds such as amino acids

and proteins.  The nitrogen must be first made availabe in

a fixed form.  This is a process that dissociates molecular

nitrogen and uses the resulting atomic nitrogen to form

compounds such as ammonium (NHt via biological fixation

or nitrate (N0~) or nitrite (NO-)  via a biological fixation.

On land the biological fixation is performed mainly by

organisms symbiotically associated with leguminous plants.

For continuous high levels of productivity over large land

areas fertilizer must be used, but high yields over limited

areas on a continuous basis or on extensive areas inter-

mittently can be obtained by recycling biologically fixed

N2.  In aqueous media fixation is primarily accomplished

by blue-green algae.

     Nitrous oxide can in principle be produced by either

nitrification or denitrification.   Nitrification is the

microbial oxidation of ammonia to nitrate and nitrite.

Biological nitrification is the major source of N20 in the

biosphere, and is mediated primarily by obligate aerobic


                               -  25 -

     Denitrification is carried out by bacteria that can use

nitrate as a terminal electron acceptor when oxygen has been

depleted via the sequence;
nitrate      nitrite     nitric    nitrous   molecular

                         oxide     oxide     nitrogen

(see also figure 2)

The above sequence is somewhat oversimplified.  For example,

it appears that between NO™ and N~0 is an intermediate some-

times referred to as X.     It is possible that X is hyponitrous

acid but the present evidence is not positive and no identi-

fication can be made at this time.  It has been estimated that

about 25-30% of the time does fixed nitrogen go into denitri-

fication in soils^    and about 6% in oceans.

     The ability of soil microorganisms to reduce N-0 to N2

is inhibited by the NO^ content of the soil.      All nitrogen

fertilizers provide nitrate to the soil, either directly, because

they may be wholly or in part in the nitrate form, or because

they are in the ammonium form or are changed to the ammonium

form, which can then be converted to nitrate.  All crop manage-

ment inputs and techniques which increase plant growth and return

                                 - 26 -

more plant biomass to the soil and increase the amount of nitrate

produced in the soil.

     Without the denitrification process the oxygen and

nitrogen of the atmosphere would long ago have been depleted.

Thus one would expect that the rate of denitrification is

on the average equal to the rate of fixation.

     As long as there is sufficient free oxygen the aerobic

microorganisms use it for respiration.   However when the

oxygen level falls below 5% in soils the microbial popu-

lation may use both free oxygen and combined oxygen in the

form of nitrate thus forming N-O.  Since N20 is a gas, some

of it in any case will diffuse away before the denitrification

becomes complete.  Under the situation of extreme lack of

oxygen, N^O will be an intermediate with N~ the final product.

It is thus important to measure the N20/N2 ratio in the

denitrification process for different ecosystems.

     Factors that affect denitrification are;  ^ '  '

I.   Free oxygen content must be very small  (e.g. in soils,

water, sediments).  The redox potential should be 330 mv.    '

2.   The temperature range must be between 5 C and 60 C, although

individual species have narrower acceptable temperature ranges. '   '

                                - 27 -

The production rate usually follows the well known Arrhenius
equation with rate changing expedientially with temperature.
3.   When the pH is lower than 6.5 generation of N20 compared to
N2 is favored     although the total denitrification rate is
4.   There must be an available carbon source.
     B.   Atmospheric Concentration and Residence Time

     In the troposphere up to 12 km the concentration of N~0 is
fairly constant.  However there is some disagreement about the
absolute value of the concentration.  Hahn and Junge      put the
value at 260 ppb while Singh et. al.  v  ' find it to vary less
than 4% between northern and southern hemisphere at an average
concentration of 311± 2.6 ppb.  A recent urban study puts the
value at 329.5i 3.35 ppb.      The value has been shown to
increase 1.5% over the last decade by Weiss & Craig and 0.3%/yr
for the last two years by Singh et al.      However the latter
experimenter observes that they could not detect a change if it
were less than 0.5%/yr.  Some researchers using the above
information claim that there is no N~0 problem.     From balloon
measurements     the change of N~0 concentration is less than

                                - 28 -

     An early study of N~0 concentration      showed a concentration

increase of roughly 20 ppb/yr during the period 1967-1969.  An

extrapolation of this increase would predict the 1978 atmospheric

concentration to be roughly 460 ppb.  Since the concentration is

nowhere near this high and since the data has not been reproduce-

able it is highly suspect.

     According to Tony Brodrick     and Paul Crutzen     there

is no convincing hard evidence that the concentration of N20 is

changing in the atmosphere.  Crutzen feels that the study by

Weiss is "coming close" but he feels that additional data are


     In the troposphere the concentration thus is approximately

310 ppb and decreases to about 50 ppb at 35 km.  This leads to

                                     15   *
a tropospheric burden of M = 1.6 X 10  gm.   A similar calcu-

lation for the stratosphere leads to an additional burden of

0.16 X 10  gm.  Using an average removal or residence time of

T = 80 yr. then the input flux to the troposphere is;

     n -  M  _  1.6 X 10 5 gm      on v in12     * vr n /
     Q - — -    80 yrs.	      = 20 X 10   gm of N20/yr.
  This value' can be arrived by using the information that

1 m  of air = 1.2 kg at ground level and approximately 0.4 kg

at 12 km.  Then let the average mass density for the troposphere

mixing ratio (by volume) of N20 X average density of the tropo-

                                               Q             -3
sphere X volume of the troposphere = (310 x 10  )   (0.8 kg/m )

(TIT)  Jj6412)3 - (6400) 3J m3 = 1.6 X 1015 gm. of N0.

                                 -  29 -
     Uncertainties still exist concerning the value of the
tropospheric residence time of N20.  A method for estimating
residence time (Test) has been described by Junge, Let;
        = real standard deviation  (for concentrations)
        = overall standard deviation
        = variability due to precision
Quantitatively Singh et al     finds these values to be
         =  (0.8)2 -  (0.3)2 =  (0.7)2
Then using the general Junge criterion

     Test = L4   = 20 years

It seems surprising that any meaningful value emerges from this
analysis since 
                                - 30 -
     It has been observed     that Junge ' s model represents a

valid -lower limit for lifetime.  Other modelers put the lifetime
at, 150 yrs^60^ 130 years ^~"'' and 100 years VJ~''. These are

lifetimes with respect to stratospheric photochemical destruction.

Thus they are upper limits and other sinks could shorten the

photochemical lifetimes.  Note that if the variability is less

than 1% in the troposphere this is consistent with a lifetime

larger than 100 yrs.

     Some authors use different names for this quantity.  For

instance Singh et al.    ' determines the turnover rate to be

67 years.  The term removal rate is sometimes used.v     If a

gas is consumed the description used is lifetime but if the

gas is merely passing through to the stratosphere perhaps a

better description is residence time.

     Crutzen and Ehhalt^  ' note that in 1967-68 there was a sharp

increase in N~0 levels indicating that the turnover rate must

be "short".  Using Junge's relationship described above they get

Test = 10 yrs with an uncertainty of a factor of two.

     The estimates of the residence time of N~0 range from

5 to 160 years.      It appears that a more reasonable range

of possiblities is 100-15-0 years.

                                -  31  -

     The equation used above, M = QT  shows that far a given burden

M, Q and T are not independent.     Using the known concentration

of N-O as shown above led to 1.6 x 10   gm.  It would appear that

atmosphere concentration is know within an accuracy of 10%.  Thus

M is the best known quantity in the equation.  The above range

of residence time (100-150 yrs) has a larger uncertainty.  This

range of lifetimes implies an injection rate of between 16 and

11 Mt N20/yr.

     C.   Sources of N~0
1.   Land

     a.   Review

     The denitrification losses in soils depends on a number of

factors among which are the availability of organic carbon sub-

strate, the moisture content of the soil  (02 content or redox

potential), the presence or absence of a cover crop and the type of

fertilizer applied.  The proportion of nitrogen liberated as

N20 varies from 0 to 100%.  Another example of wildly varying

fluxes is shown in the fact that the N20/N2 ratio can vary between

1/0.06 and 1/6. (9)

     There are three main methods used to measure fluxes of

nitrous oxide from the soil, viz;

                                -  32  -

     1.   use soil profile concentration measurements and

     classical molecular diffusion theory  (e.g. reference


     2.   the box method measuring how concentration changes

     with respect to time (27,68,69)

     3.   the micrometeorological method which applies turbu-

     lent diffusion theory to air concentration profiles  (24).

     Example of fluxes of N~0 that have been measured from
soil using the profile method generally are of the order of


magnitude of 10~12g cm"2 s"1 of N2O (3.4 x 10~12   cm"2 s"1 for
                                                       —12    —2  —1
soil covered with grass and trees to less than 0.1x10   gems

for desert soil).  Cropland soils show fluxes of approximately

        -12     -2  -1
0.3 x 10   gm cm   s  .  Compost piles show concentrations as

high as 20 ppm.  '
     As an example of the third method one worker uses the

aerodynamic profile method which calculates fluxes using the

equation; <19'24>
     F = K2 V2   (C   - C    /
               1 n[Z-d)

                                  - 33 -

     F = flux density  (gm/cm /sec)

     K = von Karman's constant  (0.41)

U*,V*  = friction velocity  (cm/sec)

     C = N^O concentation  (gin/cm  ) at height, Z(cm)

     u = windspeed  (cm/sec) at height Z  (cm)

     d = crop height  (cm)

The method does allow correction  for thermal instability.

     Using the above method one experimenter has determined

that N»0 can emanate or be absorbed by  soil.  His calculations

for two separate days showed a flux towards the soil of ^230 X

  -12      -2 -1                                           -12
10    gm cm  s   on a warm dry day call  a flux of ^160 X 10

     -2 -1
gm cm  s   on a colder damper day.

     These values are surprising  in two  ways.  First they are

considerably larger than most other measurements  (a recent

study shows that the concentrations change by as much as an

order of magnitude in one day     ) and  secondly they show that

in one case the soil is a sink.   The method used is the closest

to a direct measurement but it does depend on a theoretical

model.  One possible explanation  of the  large fluxes calculated

in this way is that in general the fluxes are both in and out

of the soil and that those experimentally measured by the first

                                - 34 -

two methods is the difference between the emitted and  absorbed

fluxes.  A recent estimate indicates that the N20 supplied  to

the atmosphere by soils probably does not exceed about 7 M  tons


     b. Analysis
     To understand the magnitude of a flux from  soil of magnitude

          2   1
100 gin cm   s   consider the situation where 2/3 of the global

land mass (1/3 the earths surface area) is emitting at that

       _,3 _ gm

100 X 10    2
          cm  sec
                 [3.15 X 107 secfl2\ ( l)            ~\ fl010cm21=  3000  Mt

                       I        JisJ Uj 411(6400 ) ^ kmT   .  .  2     N00 per
                 t     1 yr.   ^.l    '            J.j_ 1  km  I    2  r
This can be compared to the total global soil nitrogen  loses

due to agricultural activites of 6000-13000 MT. v   '  This  is

larger than the total flux from all sources estimated as 210

Mt/yr earlier.  This is clearly much too large  a  flux and  suggests

that the problem is more complex than has previously been

expected.  It also demonstrates the danger of extrapolating

from a single site to a global level.

     Evidence for the complexity of the relationship between

N20 and soil is shown in a study of Iowa surface  soils  showed

that their capacity for uptake of N20 under conditions  favor-

                                - 35 -

able for denitrification of nitrate was much greater than
their capacity for the release of the gas. ^  '   The experiments
involved incubating the soils in anerobic conditions and showed
that the soils had the capacity to take up all the N20 produced
by denitrification.  At least 90% of the gas released was in
the form of N2.  This information casts doubt on techniques
used to measure the flux of N20 which rely on soil concentration
profiles and indicates that the use of chambers placed over soil
surface is a superior technique.

     An upper limit on the flux of N20 possible from any system
is given by the rate of fixation.  In some cases the rate of
fixation is known more precisely than the rate of N20 emission.
From some "very patchy" evidence   ' it is estimated that nitrogen
                                                    -12     -2
fixation occurs in forests to the extent of 3-5 X 10   gm cm
 -1                                                  9
s  .  For an area of global forests of about 4.1 X 10 ha
         8  2             17  2
(lha = 10 cm )  or 4.1 X 10  cm  this gives a global average of
approximately 50 Mt/yr nitrogen fixed in forests.  A similar
                                         -12      -2 -1
analysis for unused land yields ^-0.6 X 10    gm cm  s   for
4.9 x 10 ha and a global total for nitrogen fixed of 10 Mt/yr.
     The emission and absorption of N20 by soils is the most
diverse and complex part of the situation and would thus deserve
the most attention and investigation in the future  (note also
in the section and sinks that soils could be a very large sink for
N20) .

                               - 36 -

     In summary,  although some experimental measurements have

been made to determine the N20 levels for soil systems they

are too few and specialized to make accurate generalizations

from.  Studies need to be made involving several parameters

among which should be; type of ground cover, kind and amount

of cultivation, moisture and humidity, pH, meteorological

conditions, sunlight, fertilizers or nutrient and soil com-

position.  The measurement of N-O from land areas will likely

be the most difficult complex and involved of any of the

natural flux determinations.

2.   Freshwater and Sediments

     a.   Review

     As recently as 1977 no known measurements had been made

for fresh waters.     However recently it has been shown that

fresh surface water is supersaturated with N20 bearing about

20% more than the air immediately above it.   '   Significantly

larger concentrations supersaturated by factors up to five

have been found by Cicerone et al which are due to anthropogenic

activities such as sewage treatment. v  '  Also municipal tap

water samples from U.S. cities show saturation factors between

1.8 and 89.2 (a factor of one indicates equilibrium and a

factor of 89.2 means that the dissolved N20 was 89.2 times

more concentrated than in equilibrium with the atmosphere).

                                 -  37  -
It was suggested that chlorination of river water and wastes

might be a key step in the production of N20.
     Within a few cm. of the surface of sediments exist the

ideal denitrification condition.  A possible mechanism for

nitrate transport involves the upward seepage of groundwater.

Estimates for the amount of nitrogen denitrified by this process

range from 23 to 81% of that retained in the lake.   '

     In lakes it was found that only waters deeper than 100 ft.

produced significant amounts of N20.      It is suggested that

this occurs since photoplankton are absent and the supply of

organic matter is scarce.  In these circumstances the nitrifying

microorganisms can compete more successfully for the limited

amount of available ammonium.  Also there are often more nutrients

below the discontinuity layer (50-60 M) than in surface water.

     b.   Analysis
     The average flux from some lakes is approximately 2 x 10

gm cm"  s~ ).     It might be expected that the average flux

from fresh water sources would be of this order of magnitude but

                              16  2
slightly higher.  Using 2 x 10  cm  as an estimate of the area

                        -12     -2  -1
of freshwater and 5 x 10   gm cm   s   as a rough estimate of

the average flux of N20 from freshwater sources than the annual

flux would be:

                                 - 38 -

           -12    gm
     5 X 10
              cm  sec
\2  X lO^cmf UJLJLJLOII f-lMt_ 1

L          JL    1 yr.   J[10-" gmj
     = 3.15 Mt of N20/yr.
Some heavily polluted rivers such as the Rhine show fluxes as

large as 70 X 10~12gm cm"2s~1.(28)
     It has been suggested that N20 is produced in water and

soil systems by the oxidation (nitrification) of ammonia in

addition to that produced by denitrification.  If this is so

monitoring N~0 could be used as a way to study nitrification.

     Thus the contribution from freshwater sources appears to

be small however it would appear that a more careful and closer

examination should be made of the contribution of sewage treat-

ment plants.

3. Oceans

     Nitrification appears to take place in the soil, water,

the upper layers of sediment and in the ocean where it is

                                           (1 o \
most efficient below 200 M.  Hahn and Jungev  ' suggested that

this occurs since photoplankton are absent and the supply

of organic matter is scarce.  In these circumstances the

nitrifying microorganisms can compete more successfully for

                                 - 39 -

the limited amount of available ammonium.  Also there are

often more nutrients below the discontinuity layer  (50-60M)

than in surface water.

     The range of fluxes measured from oceans vary widely.

These fluxes are not measured directly but involve using

concentration data with a model.  The most common model is

called a stagnant film diffusion model     or a two layer

model.      The model assumes that both phases  (air and sea)

are well mixed and that there is no turbulence or wind.  The

flux is then proportional to the difference between the con-

centration of N20 in the water and the water concentration

when in equilibrium with the atmospheric abundance.  This is

derived from the one dimensional relationship for molecular

            -   n
            - - D
where F is the flux, c the concentration, Z the depth and D

the coefficient of molecular diffusion of gas in the layer  (air-

sea layer).  The rate of evaporation/condensation determines

the exchange parameters.  The proportionately factor in the

above relationship can be derived for ideal conditions.

     Measurements of 42 ocean samples from the North Atlantic

by Hahn and Junge    , some taken at depths of 3000 m, show

                                - 40 -

a concentration of N20 in the water of 0.4 ppm v.  Some of the

samples were slightly below equlibrium concentration and some

were saturated by as much as 280%.  The flux using the above

simplified model is 70 Mt N20/yr for all the oceans if the average

concentration is 0.4 ppm v.  The individual flow ranged from  (0.82

to 1.66) x 10~12g cm~2s~1 near the equator to  (0.05 to 0.66)

    — 12     —2 —1
X 10    g cm  s   at 60 degrees north latitude.

     The global number was determined by substitution in the


     F = - K  (C - Ceq)

where C = 0.4 ppm v was the concentration found on the water

and Ceq = 0.25 ppm v was the assumed equilibrium value (Ceq is

actually closer to 0.31 ppmv).

     Singh et al     have measured water concentrations in

the Pacific ocean with an average surface saturation of 133%

 (a super saturation of 33%).  They found that some of the

concentration below the mixed layer to vary between 1 and 7

times that of the surface concentration.  These values lead

                                              —12     —? —1
via the model to an oceanic flux of 0.359 x 10   gm cm  s

or 32.4 Mt N20/yr.  If fresh water sources are included the

total goes up to 35 Mt N,,0/yr.

                                - 41 -
     A survey was made of several oceans of the world sampling

N-0 in the surface water and the air by Weiss et al.      Each

day analysis was made of the atmosphere and the ocean for a total

of approximately 8000 analysis in all.  They were analyzed by gas

chromotagraphy with a precision of 0.3 to 0.4%.
     The trip covered the ocean from California through the

Panama Canal to Spain, then through the Mediterranean Sea to the

Arabian Sea, and Mauritius.  The trip then went to Antartica and


     The overall average super saturation of ocean waters with

respect to the atmosphere was approximately 4% with a few

occasional spikes as much as two fold.  These values are much

lower than those of Hahn and Junge.

     The procedure used to determine the flux from the ocean

was to determine the partial pressure of N-O and then estimate

how much was released.  A check is provided in the levels

of radon gas released from the radium in sea water.  The

atmosphere has very little natural radium). This mechanism will

show the rate of release of a gas from the ocean.

     In general the saturation levels were very near the

equilibrium level except in regions where there is upwelling

                                - 42 -

bringing N20 up from the lower depths or frontal boundaries

formed by sharp temperature changes.

     Contrary to the findings of Hahn and Junge     and Singh

et al^14^, the levels found by Weiss     in the North Atlantic

were remarkably uniform and averaged no more than a few percent

from saturation.  The western basin of the Mediterranean showed

some supersaturation and the Red Sea was slightly undersaturated.

But as samples were taken further south in the Red Sea values

they showed significant spikes.  These high values correlated well

with oceanic mixing boundaries.  The highest value of the

expedition of two fold supersaturation was recorded at one point

in this area.

     The Northern Arabian Sea showed isolated high values of

320-330 ppb due to upwelling.  In the Antartic a polar front

structure reflected N-O distribution.  This is a region where

warm water is moving one way on one side and cold water another

way on the other side.

     Using the above concentrations, the model, and the

radon flux information a global flux estimate can be made.  This

estimate assumes that the tract followed by Weiss was representative

of all the ocean.  The estimate of approximately 4 Mt N20/yr and

is considerably lower than the 70 Mt N20/yr estimated by Hahn &

Junge.       Due to the major differences in measurements in the

                                - 43 -

ocean and their importance  (see discussion in Global Balance

N~0 section) the magnitude oceanic fluxes needs to be measured

by an independent researcher.  In terms of biological activity

the surface of the oceans over the world is very heterogerous.

Hahn, Weiss & Singh have simplified different parts of the large

and highly variable ocean.

4.  Fertilizers

     a.  Review

     The dramatic increases in crop yields since WW II have

been largely due to an increased use of nitrogen fertilizers

and plant nutrients.  This has been mainly due to better

genetic stock.  Higher yielding hybrids require more nutrient

and care; therefore, the agricultural chemicals become very

important in allowing the plants to reach their genetic potential.

     The mechanism by which N20 is produced usually involves

an anerobic bacterial activity denitrification.  It is estimated

that about 1/3 of fertilizer used  (57 Mt was fixed industrially

in 1974-reference 23) in agriculture is returned to the atmosphere

by denitrification in a relatively short time  (decades or so).

The yield of N20 from fertilizers is between 5 and 20%.  Thus the

fertilizer used may end up as nitrous oxide although it may take

several years for this to happen.  The time estimates of this

phenomena show a large range of uncertainty.

                               - 44 -

     Commercial fertilizers are made up of nitrogen in the

form of nitrate, ammonium, and/or organic nitrogen compounds.

They are used to fertilize arable soil and in some cases on

pasture land.  They add to the substrate to enhance denitri-

fication and nitrogen uptake.

     Different chemical forms of a fertilizer added to the soil

can lead to N20 production.^  '   If it is added in the form of

a nitrate the mechanism of denitrification may follow and if

it is added in the form of an ammonium it will follow a cycle

of bacteria nitrification leading from NH3 to NH2OH and the

intermediate X.(13'31)

     If fertilizers are added to the soil in the form of

urea they are hydrolyzed by soil urease to form (NH4)2 C0_.

It was observed that this ammonium is nitrified under aerobic

                                                            / on \
conditions to form nitrate.  The amount of the resulting N20   '

released can be greatly reduced by the use nitropyrin.   However,

this has not been demonstrated under field conditions.

     High N20 fluxes are found immediately after a rain or

in anerobic conditions.  Using NaNO  produced fluxes of 50 x

10   g cm   s   for recently cropped areas after a rain,   '

and applying KN03 has yielded fluxes of up to 480 x 10~12g

cm   s  .   This latter flux was 10 days after application.

                                  - 45 -

The N20 flux increased until the 18th day when it reached
         -12    -2  -1
1350-X 10   g cm   s     Thus N20 can be generated by the
anerobic denitrification of nitrate fertilizer.  However the
values were determined by the soil gradient technique and thus
are questionable.
     b.   Analysis
     An estimate can be made of N~0 flux from fertilizers by
assuming that each person on earth requires 1/2 acre for food
                  9                             2
production  (4 x 10  people and 1 acre = 0.004 km ).  Using a
          —11      —2 —1
flux of 10    gm cm  s   produces 25 Mt N90/yr from fertilizers,
     The contribution of fertilizer usage to atmospheric N«0
appears to be the largest anthropogenic source and should be
expected to increase significantly in the future.

5.   Other Anthropogenic Sources

     a.   Review

     The relationship between natural and anthropogenic N-O
is demonstrated by comparing the natural rate of fixation with
that due to industry.  The natural rate of nitrogen fixation
is about 200 Mt/yr.

     Industrial fixation for agricultural purposes was 3.5 MT/
yr. in 1950 and it is predicted to be between 100 and 200 Mt/

                                - 46 -

yr. (the rate of application of nitrogen fertilizer in 1977 was
about 40 Mt.) (8'23)  by the end of the century.      It is
estimated as 79 Mt/yr.*16^  For all other industrial processes
the rate was 15 Mt/yr in 1950, 40 Mt/yr in 1974 and could grow
to 100 Mt/yr by 2000.  These values should be compared to the
natural rate of about 160 Mt/yr.  Thus it can be seen that
the rate of N?0 released from anthropogenic sources could be
growing at a rapid rate.

     As pointed out earlier the concentrations of N^O can be
                                        (27 28)
larger near wastewater treatment plants.   '      This occurs
in part because the easiest and most economical way to treat
wastes involves nitrification followed by denitrification.  It
is the latter step that produces nitrous oxide.

     Samples of air had been taken, sealed and kept from as
far back as 1961 by Barman Craig of the Scripps Institute of
Oceanography.      They have been analyzed for N2O and the con-
centrations show a slow statistically significant increase with
time.   They go from 291 ppb in 1961 to 300 ppb in 1978.  This
increase is very similar to that of CO- for the same time period.
For the last decade the N20 atmospheric concentration has increased
1.5%     while that of C02 has increased 2%.   It suggests from
these measurements that the N20 change in the atmosphere may be

                               - 47 -

     Samples from the exhaust stacks fossil fuel power plants

have been taken and analyzed for the N-O/CO- ratio.  The CO-

concentration value is well known for combustion processes

being in the range 12-14%.  The exhaust stack concentrations

of N20 for two large fossil fuel plants were 25.8 and 25.0 ppm

by volume.  The equilibrium values are of the order of 0.1 ppm

showing concentrations of least two orders of magnitude above

equilibrium.  It was suggested that the major source of N2
                               - 48 -

the value at 2.2 Mt N20/yr.  It is also predicted  that  this  flux

is increasing at a rate of 3.5% per year.

     It has been suggested that the use of catalytic  converters


could lead to the production of N20 by the reactions;
          6NO + 4NH3

     b.   Analysis

     The N20 in automobile exhausts was found to be  less  than

ambient levels unless catalytic converters were used and  then

the concentration was found to be in the range of  1-10  ppm by

volume.   '     Using an average value of 6 ppm for  the

concentration of N20 from autombiles their contribution to the

global flux can be estimated.  Assume that there are 100

million automobiles with an average engine displacement of 2

liters running at 2000 rpm for 2 hours each day each with a

catalytic converter.  Then the total contribution  from

automobiles would be

2000 revolutions         honr-Qi (2000 cm3) ...     -6, (1.2 kg) ,nn8x
                \ *• •«»• j u _> Li\j LIJ. o)    ...—     ^ ^ 2C JL U   )      •• "• ( J_ (J J

                               (revolution)         (m )
     = 0.13 Mt N20/yr

                               - 49 -

     Thus the contribution from automobiles would be minimal

even with the catalytic converters.

     Thus there are several anthropogenic sources other than

fertilizer usage that need further study.  Possible significant

contributions can be made by fossil fuel combustion and sewage

treatment processes.
D.   Sinks of N00
     1.   Stratospheric Sinks
     Once N20 survives long enough to reach the stratosphere

it has two possible sinks  '
     N20 + O^D) 	9- 2NO            K = 6.3 x 10"11

                 	** N  + 0         K = 5.1 x 10"11
Stratospheric N20 can be destroyed by photolysis
The resulting nitric oxide is an ozone depleter
     NO + 03	> N02 + 02       K = 2.1 x 10 12  exp  (-1470/T)

                                -  50  -

                                            2         -1
     The meaning of a rate constant  (unit cm  molecule  ) is

found in the relationship


     T = k~c

where T is the lifetime, k the rate constant and c the concen-


     The amount of N20 consumed by the reaction
                    K = 6.3 x 10'11
can be determined using the following information
Altitude       0(D )           Atmospheric N20     Atmospheric

 (km)          from           Mixing Ratio        denisty-from

               reference      ppb(v) from         reference  (45)

               (45)           reference 13        (particles/cm )

20               1               180                1.7 x 1018

25               6               100                7.7 x 1017

                               - 51 -
Then using the relationship
     -d[N20]  = K[N20]


»] [N20]
(particles/cm )
3060 x 108
770 x 108
252 x 108
63 x 108
8.1 x 108
                                                    3.6 x 1017
                                                    1.7 x 1017
                                                    8.1 x 1016
                                                    4.3 x 1016
                                                    2.3 x 1016
                                                    1.3 x 1016
                                                    7.2 x 1015
                                             molecules/cm /sec
The volume of a spherical shell at these altitudes with a
                            9   3
width of 5 km is =  2.6 x 10  km .
Thus we have at 20 km for example:

                               - 52 -
19.3 molecules [2.6 x 109 km3]  lO^cm3! 44 gm	^Ys.15 x 107sec1

     cm3 sec                    1 km I 6 x 10   molecules    1 yr
   (19.3) [6.0 x 109]  gm
= 11 x 1010  22  - >

      then add the contribution for all altitudes

 (19.3 ^ 29.1 + 42.8 +16.8)  (6.0xl09)  gm =

    1,0 Mt N20yr/ for consumption to form NO

also   1.0 Mt/yr to the reaction producing N~ and 02

Thus most of stratospheric N-0 must be photolyzed in order

to have an overall sink of the order of 20-40 Mt N2O/yr. in

the stratosphere.

     Most of the estimates of the strength of the sink for

N20 with 0(1D)  fall around 1.0 x 108 molecules cm~2s~1.  (25)
A more recent value using Crutzen's 2D model is 5.1 x 10

            -2  -1
molecules cm   s  .  These lead to annual values of 1.2

Mt N20/yr and 7 Mt N20/yr.

                                - 53 -

             (33)                        9             2
     Crutzen*    estimates that 3.4 x 10  molecules/cm /sec of

N_0 cross the tropopause at 15 km and are either photolyzed or

             1                                          8
react with 0( D).  This latter reaction returns 2.5 x 10


molecules/cm /sec of NO, N02 and HN03 back to the troposphere.

                               9             2
The net result is that 3.1 x 10  molecules/cm /sec are consumed

in the stratosphere.  This corresponds to consumption of approxi-

mately 33 Mt of N20/year in the stratosphere.
     Recently reactions in the stratosphere involving H0_ have

been investigated in some detail and new rate   '  ' constants

and temperature dependence has been determined.  This new

information indicates that H0~ consumes NO faster than had been
                               -  54 -
2.   Other Possible Sinks
     It has been demonstrated that soils may act as sinks

as well as sources for N20 through the uptake of atmospheric

N-0 by soil microorganisms.  '     This suggested idea has

also been suggested as a mechanism involving oceans.

     Several possible tropospheric reactions have been

investigated.  The rate constant for N20 + OH is less than

4 x 10~   cm /sec with a corresponding removal time of more

than 10  years.     The experimental turnover time determined

by direct measurement is 5-15 years.  A recent report shows

the rate constant to be (3.8  1.2) x 10    cm  molecule"

sec~ .
     It is suggested that the reactions
           H02	>  N2 + 02 + OH
               	>•  2NO + OH

will be slower than that with OH by factor of 100.

     Using the data of Blackmer and Bremner   ' an estimate

can be made of the possible magnitude of the soil as a sink.

They showed that approximately 1 yg of N20 can be absorbed

by a gm of soil in 10 hrs.  This rate was measured for 30 gm

1 yg N20
gm 10 hr
'0.6 gmj
.cm2 J
"24 hr"
. 1 day
[1 Mt "\
1018 pgj
                               - 55 -

density of roughly 0.6 gm/cm .  Then assuming that 1/6 of

the earth is soil and that the radius of the earth is 6400

km the annual absorption of N20 by solid would be:
                                                       rin5   2-
                                      j-i          99 -i-U cm

                                  -x-4 n  (6400)  km2T^   I
^  112 Mt N20/yr.

Assuming that soil conditions will be appropriate about 1/3 of

the year yields on annual absorption of ^ 40 Mt N20/yr.

This estimate is derived from an enclosed bench top experiment

and there is only weak evidence that measurements in the field

demonstrate such an effect.   (Cicerone showed that on 8 of 60

days his soil was a sink^    and similar results have been

found by Rolston^    & Piebie.      It would seem that before

the microorganisms involved in denitrification can consume

the nitrates and start consuming the N-0 it is blown away.

However in the sense that everything comes from somewhere and

goes somewhere the N20 transported by the wind could be

absorbed at another site.  Recent diurnal variations in N20

atmospheric concentrations found by Bruce, Eggleton and

Penkett     suggest that there is an important ground surface

sink for N-O.  This uptake appears to be due to the reduction

of N20 to N2. l  '

                                -  56  -
     It has been shown that nitrous oxide will undergo both

a thermal and a photochemical decomposition at 296°K when

it is absorbed on various dry sands.      The rate of

decomposition was dependent on pressure and moisture.  The

destructive efficiency under optimum conditions was the order

of 2 x 10   molecules/incident photon.   Assuming that 1/3 of

the global land mass is desert or near-desert and 1.7% of solar

intensity is in the range 280-320 nm then approximately 70 Mt/yr

of nitrous oxide could be consumed at the above rate.  Thus

decomposition by desert soil is a potential sink for N0.
     Thus although some possible sinks for N»0 in the atmosphere

other than photolysis and reaction 0( D)  have been suggested,

none has been definitely shown to be a viable mechanism.
IV. Global Mass/Materials Balance of N.,0

     In analyzing a complex system such as the way that

nitrogen and its compounds move through the environment the

mass or materials balance approach can be a useful organizational

tool. This model examines all possible sources and sinks allowing

the comparison of different fluxes of N20.  It also provides

insight and understanding into seemingly unrelated components,

flows between reservoirs, the relationships between anthro-

pogenic and natural sources and allows for a better design

of monitoring on data gathering activities.

                                 - 57 -

     However the data regarding fluxes of N20 are somewhat

limited and have a large uncertainty.  For example the amount

of N20 evolved from land was estimated as being between 16 and

69 Mt N/yr by one research group based on only two experiments.

And recently it has been shown that the soil could be a sink

of the order of 60 Mt N20/yr.  This wide range of values makes

it very difficult to predict the effect on the stratosphere

due to N.,0.  To get a rough of the order of magnitude of global

flux units assume that the average global production of N20 is

      -12      -2    -1
1 x 10    gm cm   sec  .  Then for the earth:

1 x io12 -2f -   4 n [*6400)2 km2!3'15 x 107 secjf lOJ^ITj

         cm  sec                    u    1 yr.     -"•  1 kmj. 1
                                                             10   gm
     = 160 Mt N20/yr is emitted into the atmosphere
The global net production of N20 is estimated by others to

be between 125 and 475 million metric tons  (Mt) of N20 per year

with 210 Mt as the most likely value.      A recent review

listed the estimates for sources and sinks of N,,0.  These and

other values are shown in Table IV.

                                 - 58 -
                          TABLE IV
                         Amount (10  metric tons of N20/yr)
                         Likely value
land (soil)
fixed nitrogen
including fertilzer
man made  (direct)
fresh water  (13)
                         70   (5)'


                        144  (79)
Stratosphere; Reaction
with UV and 0 ^D)        20
Troposphere; Reaction
with 0( D), OH, negative
ions & HO-
a) recent measurement by Weiss,
c) reference
                                    b) educated evaluation,
     As can be seen from Table IV a large unknown sink could
exist in the atmosphere for N20.  At this time there are some
guesses as to what that sink might be e.g., the soil).  At least

                                 - 59 -
one researcher has suggested that the above marine estimates

are much too large and thus there is no missing sink.
     Some controversy centers on how much comes from  (or goes

to) the oceans.  Hahn,(13) Yoshinari(37) and Singh(14) et al

have estimated fluxes from oceans that require a tropospheric

sink.  Singh concludes that since the oceans are so large,

the contribution from anthropogenic sources must be small.

Weiss     with his lower estimate of oceanic flux would say

all the sources & sinks are known and balance  (his flux

value combined with Hahn & Junge's other estimates place the

lower bound of sources much closer to estimate value  for

sinks).  Some support for this latter view comes from a review

by McElroy et al.(28)

     The Marine estimate shown in Table IV comes from an average

of the few measurements made over oceans.  It would appear

that more measurements are needed over oceans.  The most

important areas are those which are nutrient rich such as the

upwelling zone along the West African coast, the Indian Ocean

and the Arabian sea, and coastal areas near some rivers

connecting to highly populated areas.

     One source puts the estimated input of N~0 from  fertilizer

at 3 to 11 Mt N/yr as the result of a postulated fixation rate

                                 - 60 -

of 200 Mt N/yr.  For comparison with Table IV this would be

5 to 16 Mt N20/yr.(8)

     The land estimate range of values in Table IV could also

be narrowed by better numbers for the fluxes over crops and

                                                     ( JC\

forests.  Using the data given by Blackmer & Bremmer      the

estimate of a soil being a sink could be as large as 40 Mt

N~0/yr.  Thus the soil could turn out to be a significant sink of


     The fertilizer estimate is derived from the amount used and

estimate of how much is released.  Two projects are underway

                           (19 38)
to start such measurements.   '    However, it appears that

the precision of 0 ppb can only detect fluxes down to about

10~   gm cm"  sec   which is two orders of magnitude above

the flux estimates from soils and water.  A recent study

                                                 -13      -2
claimed that a detection limit for flux of 2 x 10    gm cm

 _ -I

s   can be acheived using samples taken over even 30 minute

durations.     The discrepancy concerning the possible limits

of detection appears to be caused by different sample handling

techniques, and the theoretical flux models used.
     In summary the estimates of fluxes associated with sources

and sinks of N20 are based on very few actual measurement.  They

are for the most part rough first order estimates and in some

respects are confusing.  With some authors suggesting that

                                 -  61  -

the oceans and soil are potential sinks and that the flux may

be into and out of these areas indicates that there are many

open questions and much more to be learned about how the

N20 cycle works.

     A recent estimate indicates that the N20 supplied to the

atomosphere by soils probably does not exceed about 7 M tons

N/yr. (63)

     A not unreasonable approach to determining the  best overall

guess from the current inadequate information would be as


     1.  Use the estimate of N20 flux that is determined

         by residence; viz, 11-16 Mt N^O/yr.

     2.  Assume that all sinks are in the stratosphere and

         we use the upper limit; viz, 16 Mt N20/yr.

     3.  Then assume that freshwater and ocean sources

          (including sediments) yield 7 Mt N20/yr. and other

         sources contribute another 1 Mt N20/yr.

     4.  Then on balance land/crop/fertilizer systems must

         contribute 8 Mt N20/yr.  This could be all due to

         sources or a combination of sources and sinks.

                                 - 62 -
II.   Effect of NO on Stratospheric Ozone
     Predictions have been made of the increase in N20 burden

in the atmosphere due to fertilizer use.  Assuming a 6%

annual increase in the use of fertilizers, the increase

by the year 2000 will be between 7 and 31% of the 1975

atmospheric burden of N?0.   '  It is possible that at some

future date (approx. 100 yrs) the N20 production rate

will be 3 times what it is now due to anthropogenic sources.
     Models have been used to predict the change in stratospheric

ozone due to N20.  ' '  '     Using the first order approxi-

mation that the rate of nitrogen fixation is the same as that

for N20 production, the % decrease can be calculated as a

function of increase in nitrogen fixation.

     It has been observed     that the interaction of N~0

in the atmosphere and its resulting effect on 0_ is non-

linear.  Figure 3 shows the relationship between the

multiple of present N-O release and change in total ozone.

However the analysis shown below does give an indication of

the variables involved.  Since this analysis is only a first

order approximation its usefulness is restricted.

     Stratospheric NO  is produced primarily by nitrous oxide

and the relationship between NO  and 0_ has been shown to be^
                               J$*      .3




                               I            I

                               2            3

                        MULTIPLE OF PRESENT N2°
      Figure 3     Effect on stratospheric ozone of increases  in the
      present rate of release of N2O (reference 67).

                                  -  64  -
               A[0-J               A[NO 1

               	£_ =     x       	
                [03]                 [N0x]
     However since the primary source of stratospheric NO  is

nitrous oxide, then;
               A[0 ]                A[N~0]

               	— =     X	
                 [03]                 [N20]

where [0_] and  [N20] are the amounts of ozone and nitrous oxide.

     Recent information   '    would suggest that X is quite small

and positive  (perhaps 1/100).  Previous values were in the range

-1/5 to -1/10.  Thus it is suggested that N20 actually leads to

an increase in ozone concentration.  Such assessments should be

treated with some caution.  There are still some unknowns that could

change the value of X.  Some key measurements that remain involve

methane chemistry, water vapor effects, and tropospheric con-

centrations of HO.

The main reasons for this recent change in effect are:

     1.    CIO has more of an effect than previously thought.

     2.    More NO is consumed by H02 than previously thought.

     3.    The ozone is found in 3-D models to be less sensitive

          to N20 than previously thought.

                                 - 65 -

     The important variables used in the analysis of N-0 mass

balance are:

     ANF = The increase in global nitrogen fixation due to

            human activities (Mt/yr) (from combustion,

            manufacture, and use of fertilizers and

            cultivation of legumes - this assumes that the

            major changes are due to human activities)

       ot  = Fraction of denitrified nitrogen that is N20

       3  = Fraction of fixed nitrogen that is denitrified


            within a few decades

       T  = Atmospheric residence time of N^O
     Of the estimated 200 to 350 Mt N/yr of nitrogen fixation

it is estimated that 79 Mt N/yr results from the synthesis

of nitrogen fertilizer and the cultivation of legumes and another

21 Mt N/yr is fixed as NO  by combustion, another source puts

this at 26 Mt N/yr.      These fixation processes have been

increased rapidly in the last three decades so that between

1/4 & 1/2 of the total global rate of nitrogen fixation is due

to anthropogenic activities.  As part of the nitrogen cycle

this leads to release of gases as shown;

                                 - 66 -
     N0=  bacteria and
          organic matter
                                           , NO
           [N2] +  [N20] + i  [NO]
In soils o(  varies widely as temperature, pH, moisture and  the

availability of oxygen change.  Typical ranges for these

parameters are

     land      0.025 < 
                                -  67 -
from the large global reservoir.  Their conclusion is based

on the fact that the global reservoir of fixed nitrogen is

estimated to be 2 x 10  Mt so that adding even as much as

200 Mt by the year 2000 will be only a very small contribution.

However McElroy et al      point out that in general

agricultural life times are much shorter.
     The rate at which nitrogen is retained in living plants is

directly affected by their life cycle.  For cereal crops and

leaves of trees the lifetime is one year while the average

time nitrogen is retained in living plants is of the order of

15 years.      It is generally felt that in the natural nitrogen

cycle the rate of denitrification equals the rate of retention.

However the increase of anthropogenic N^O may upset the equilibrium.
     The production, ApSI00, of nitrous oxide is:   (where P  _
                        ^*                                 IN A U

is the global rate of production)
     APN 0 =*( 3 A N F
     Where 1SL, is the global rate of nitrogen fixation.  The

change in the N20 inventory must equal the difference between

the production rate and the destruction rate.
     d[N20] _                  [N20]

       dt   "    PN20 -       ~^~

                                 - 68 -

and when a steady state is acheived

     [N20] = T  PN 0

     It is thought that at present the situation is close

to steady state.

     Consider now the situation where the global rate of

production of nitrous oxide is increased from P  Q to P  Q +

AP  n and kept here for a time long  compared to the atmos-

pheric residence time.  A new steady state inventory will be

achieved.  If the change is small the new inventory will be
     The relationship between the production of N20 and fix-

ation of nitrogen is quite complex.  However a first order

approximation may be made using the expression mentioned above,
     A PN() =o( 0 A NF
Where N^ is the natural  (background) global rate of fixation

of nitrogen.  Substituting this expression and P   n =  [N_0] T
                                                IN n \J      £

in the above yields.
     A[0 ]     A[N20]      A P  n    AN

     _ 1 = x  __ _ = x   __i_- x _
      [03]        [N20]        PQ     [N20]

                                - 69 -

     The natural global inventory  [N20] is known to be about

1500 Mt N

     Assume that the rate of anthropogenic fixation of nitrogen

is 100 Mt N/yr  (this is the increase N^ over the natural rate

of fixation) then
     If we assume that all the fixed nitrogen is denitrified

in a few decades then 3=1.  Using T = 150 yrs. ando^ = 0.1

and assuming that X = 1/100 then
     A[°3]  = 0.
or the ozone concentration would increase by 1%
     Some support that the increased N20 production will deplete

stratospheric 0., is found in the phenomena that followed a large

solar proton event that occurred in the stratosphere in 1972.v  '

The proton event produced NO in the stratosphere that led to a

20% decrease in stratospheric ozone at that time.

                               - 70 -

VI.  Conclusion

     A broad and general conclusion regarding the ozone depleting

capability of anthropogenically produced nitrous oxide emerges

from the preceding analysis.  There is a problem which may

be very important on the time scale of 20-50 years.  The

problems could be a serious one but it is not an immediate crisis.


     The two main areas that need further study are

     1.   a more accurate determination of estimated global

          flux from soil/fertilizer/crop systems

     2.   a determination of how much  (if any) the N-0 atmospheric

          concentration is changing.

     The role of N~0 in nitrogen cycle has been the subject of

much study in recent years.  Although N~0 rates have been

deduced from these studies, much more specific information is

needed to determine with some reasonable accuracy the role

of each source in the overall cycle and thus its ultimate

effect or stratospheric ozone.  Perhaps the most important

information that is needed is to know under what conditions

N«0 is released from soil and also how that is related to

                               -  71  -

the amount of and type of fertilizer and how it is used.  These

studies should be conducted in natural field conditions.  Also

the variation of the N-O/N^ ratio for different crops, soils

and growing conditions needs to be determined.  The magnitude

of N20 release must be measured in the field and under different

conditions of moisture, weather, soil pH, type of ground cover

and type of soil.

     A long term (perhaps 10 yrs) study should be made of

atmospheric nitrous oxide to determine if it is changing.

In particular any possible change in the quantity of nitrous

oxide reaching the stratosphere needs to be determined.

     With regard to possible future ocean studies it was

observed that some more coverage is needed.  Particularly

the following areas should be sampled include:  the upwelling

areas of western Africa and the eastern coasts of most continents

and the eastern tropic areas of the Pacific Ocean.

     Also more studies should be made of fresh waters.  It was

noted that the Panama Canal showed high values for N20 and this

suggests a study of tropical inland waters  (a study of the

Amazon River is planned in the near future).

     In view of the studies of exhaust stack concentrations of

N»0 from fossil fuel power plants it would  seem advisable to

                                -  72 -

monitor those in a more careful and complete way.  Care should

be taken concerning where the sample is collected from and what

possible reactions could produce or consume nitrous oxide.  A

similar study should be made of exhaust gases from automobiles

with catalytic converters.

     Since the contributors to the knowledge and understanding

of global nitrous oxide fluxes are from diverse disciplines,

a symposium or workshop would be important and valuable, perhaps

every 2 years.  The major workers appear to be from the areas of

atmospheric chemistry, atmospheric physics, oceanography, micro-

biology and soil sciences.  It is essential that there be direct

communication between workers in these diverse areas.

     Other potentially sizable sources of N~0 such as sewage

treatment plants and feedlots need to be monitored.

     Should cultivation and fertilizer practices be changed?

It appears that anthropogenic sources of N20 can have a sizeable

effect on stratospheric ozone.  The most important of these

contributions involve planting, cultivating and fertilizing

crops.  The implications of the problem on world food production

is immense and beyond the present analysis.  However it can

be seen that political and economic decisions in the not to

distant future will require a better knowledge of the impact

                                -  73 -

of nitrous oxide of stratospheric ozone than is now known.   Some

continued and new studies, investigations and monitoring are

clearly indicated.

                                 - 74 -

        "                      ^~~

Tropospheric Burden                1.7 x 10  gm

Stratospheric Burden               0.17 x 10  gm

Residence Time                     100 - 150 years

Guesstimate of Global Flux         210 x 10  gm N-O/yr

Soil Flux                          0.1 - 3.4 x 10~12gm cm~2s~1

                                                -12     -2  -1
Land Flux                          up to 40 x 10   gm cm   s

                                           -12     -2  -1
Water Flux                         'v 2 x 10   gm cm   s

                                                  —12     —1  — 1
Oceans                             .82 - 1.66 x 10   gm cm   s

Fertilizer     N20                 5-20%

                                         •I /-   O

Area of:  freshwater               2 x 10   cm

                                           17   2
          forests                  4.1 x 10   cm

                                           17   2
          unused land              4.9 x 10   cm

From Crutzen

                             atmospheric scale height   _ „ ,-

     - deposition velocity =  .     .   .      .,     .  .   ~ "l3
         c               J   atmospheric residence time   10 yr.

                           = 2.7 x 10"3 cm sec"1
     - flux of N20 into stratosphere = 9 Mt N/yr.

                                               9             —2  —1
                                     = 1.2 x 10  molecules cm   s
Atmospheric concentration of 0.3 ppm v is equivalent to a

concentration in the soil of 0.013 yg N-O-N/g^  '

       12                                     6
     10   gm = 1 Mt  (million metric tons) = 10  tonnes

     1 ha = 104 m2

                                 - 75 -

1.   Response to the Ozone Protection Sections of the Clean
     Air Act Ammendments of 1977:  An Interim Report,
     National Academy of Sciences, 1977.

2a.   W.C. Wong, Y.L. Yung, A.A. Laeis, T. Mu, and J.E. Hansen,
     Greenhouse Effects Due to Man-Made Perturbations of Trace
     Gases, Science, 194  (1976) 685-690.

2.   P.J. Crutzen, I.S. Isakesen and J.R. McAfee, The Impact of
     the Chlorocarbon Industry on the Ozone Layer, J. Geophys.
     Res. £3 (1978) pgs. 345-363.

3.   P.L. Hanst, Noxious Trace Gases in the Air, Chemistry 5.1  (1978)
     pgs. 8-15.

4.   Nitrogen Oxides, National Academy of Sciences, Washington, DC,

5.   Chemical Kinetic Data Needs for Modeling the Lower Troposphere,
     Workshop sponsored by EPA and NBS at the Sheraton-Reston
     International Conference Center, Reston, VA., pgs. 15-17,
     May 1978.

6.   Nitrates:  An Environmental Assessment, National Academy of
     Sciences  (Scientific and Technical Assessments of Environ-
     mental Pollutants), In Press.

7.   H.S. Johnston, J. Geophys. Res., Analysis of the Independent
     Variables in the Perturbation of Stratospheric Ozone by
     Nitrogen Fertilizer.  82^  (1977) 1767-1772.

8.   P.J. Crutzen and D.H. Ehhalt, Effects of Nitrogen Fertilizers
     and Combustion on the stratospheric Ozone Layer,
     Ambio, ([  (1977), pgs. 112-7.

9.   S.C. Liu, F.J. Cicerone, T.M. Donahue, and W.L. Chameides,
     Sources and Sinks of Atmospheric N_0 and the Possible
     Ozone Reduction Due to Industrial Fixed Nitrogen Fertilizers
     Tellus, 2£ (1977) pgs. 251-263.

10.   D.D. Focht, Soil Sci., 118  (1974) 173-179.  The effect of
     temperature, pH and geration on the production of nitrous
     oxide and gaseous nitrogen-a zero-order kinetic model.

11.   G. Stanford, S. Dzienia and R.A. Vanderpul, Effect of Temper-
     ature on Denitrification Rate in Soils Proc. Soil Sci. Soc.
     Am. 3_9 (1975) 867-870.

12.   S.C. Liu, F.J. Cicerone, T.M. Donahue, and W.L. Chameides,
     Limitation of Fertilizer Induced Ozone Reduction By
     the Long Lifetime of the Reservoir of Fixed Nitrogen.
     Geophys. Res. Letts, 3  (1976) pgs. 157-160.

                                - 76 -
13.  J. Hahn and C. Junge, Atmospheric Nitrous Oxide:  A Critical
     Review, Z. Naturforsch. 329  (1977) pgs. 190-214.

14.  H.B. Singh, L.J. Salas, H. Shigeishi and E. Scribner, Global
     Distribution of Selected Halocarbons, Hydrocarbons,
     SF, and N~0, Stanford Research Institute, unpublished
     report, May 1978.

15.  R.J. Cicerone, J.D. Shetter, D.H. Stedman, T.J. Kelly and
     S.C. Liu, Atmospheric ISUO:  Measurements to Determine It's
     Sources, Sinks and Variations, J. Geophysical Res., JJ3_
     (1978) 3042-3050.

16.  R.F. Weiss and H. Craig, Production of Atmospheric Nitrous Oxide
     by Combustion, Geophys. Res. Lett., 3_  (1976) pgs. 751-3.

17.  Dr. Phil Hanst, EPA/RTP, private communication.

18.  C.E. Junge, Tellus, Potential of Soil as a Sink for Atmospheric
     Nitrous Oxide, 2_6_  (1977) 477-488.

19.  F.R. Lemon, USDA-SEA-FR, Broadfield Hall, Cornell University,
     private communication.

20.  H. Fluhler, M.S. Ardakan:  T.E. Szuszkiewicz and L.H. Stolzy,
     Field Measured Nitrous Oxide Concentrations, Redox
     Potentials, Oxygen Diffusion Rates and Oxygen Partial
     Pressures in Relation to Denitrification, Soil Sci., 122
     (1976) 107-114.

21.  D.E. Ralston, M. Fried and D.A. Goldhomer, Denitrification
     Measured Directly from Nitrogen and Nitrous Oxide Gas
     Fluxes, J. Soil Sci. Am. 40  (1976) 259-266.

22.  R. C. Stefanson, Soil Science, Denitrification from Nitrogen
     Fertilizer Placed at Various Depths in Soil-Plant Systems,
     Soil Sci., 121  (1976) 353-363.

23.  Effect of Increased Nitrogen Fixation on Stratospheric Ozone,
     Council for Agricultural Science and Technology, Report
     #53, 1976 (CAST).

24.  E. Lemon, Nitrogen in the Environment, Vol. 1, Academic Press,
     1978, Nitrous Oxide  (N~0) Exchange at the Land Surface
     pgs. 493-521.

25.  Don Wuebbles, Lawrence Livermore Laboratory Report, UCRL-
     50042-77, Sept. 30, 1977.

26.  A.M. Blackmer and J.M. Bremner, Potential of Soil as a Sink
     for Atmospheric Nitrous Oxide, Geophys. Res. Lett., 3
     (1976) 739-742.

                               -  77  -
27.   R.J. Cicerone, J.D. Shetter and S.C. Liu, Nitrous Oxide in
     Michigan Waters and in U.S. Municipal Waters, Geophys.
     Res. Letts., _5 (1978), pgs. 173-6.

28.   M.B. McElroy, S.C. Wofsy and Y.L. Yung, The Nitrogen Cycle;
     Perturbations Due to Man and Their Impact on Atmospheric
     N90 and 07, Phil. Trans. Roy. Soc., 277B (1977) pgs.
     159-181. J

29.   P.S. Liss and P.G. Slater, Flux of Gases Across the Air-Sea
     Interface, Nature, 247  (1974) 181-184.

30.   R. F. Weiss, private communication.

31.   J.M. Bremner and A.M. Blackmer, Nitrous Oxide Emission from
     Soils During Nitrification of Fertilizer Nitrogen, Science,
     199  (1978), pgs.   295-6.

32.   D.D. Focht and A.C. Chang, Nitrification and Denitrification
     Processes Related to Wastewater Treatment.   Adv. Appl.
     Microbiology, Vol 19, Ed. by D. Perlman, Academic Press,
     New York, 1975.

33.   P.J. Crutzen, Ambio, Estimates of Variations in Total Ozone
     Due to Natural Causes and Human Activities, 3_  (1974) 201-211.

34.   C.J. Howard and K.M. Evenson, The Kinetics of the Reaction
     of H02 with NO, Geophys. Res. Lett., 4_  (1977) 437-440.

35.   Chlorofluoromethones and the Stratosphere,  R.D. Hudson, Editor,
     NASA Reference Publication 1010, August 1977.

36.   W.L. Chameides, D.H. Stedman, R.R. Dickerson, D.W. Rush and
     R.J. Cicerone, NO  Production in Lightning, J. of
     Atmospheric Sciences, 34  (1977) 143-149.

37.   T. Yoshinari, Marine Chemistry 4_  (1976) pgs. 189-202.

38.   L. Porter, USDA-SEA-FR, Federal Bldg.,  301 S. Howes St.,
     Fort Collins, Colo.  80522, private communication.

39.   K.C. Hall and J.R. Burford, Detection of Nitrous Oxide at
     ambient atmospheric concentrations, Annu. Rep. Agric.
     Res. Counc. Letcombe. Lab. S589.L4  (1976) pgs. 55-6.

40.   P.J. Crutzen, Upper Limits on Atmospheric Ozone Reductions
     Following Increased Applications of Fixed Nitrogen in
     the Soil, Geophys. Res. Lett., 3_  (1976) pgs. 169-172.

41.   William White, Vice-President, Ferilizer Institute, Washington,
     DC, private communication.

                                - 78 -
42.  C.C. Delwiche in Denitrification Seminar held by the Fertilizer
     Institute, 1015 18th St., N.W., Washington, DC, 20036,
     October, 1977.

43.  K.A. Brice, A.F.J. Eggleton and S.A. Penkett, Nature, An
     Important Ground Surface Sink for Atmospheric Nitrous
     Oxide, 268 (1977)  127-129.

44.  A.M. Blackmer and J.M. Bremner in Denitrification Seminar
     held by the Fertilizer Institute, 1015 18th St., N.W.,
     Washington, DC  20036, October, 1977-

45.  J. Heicklen,  Atmospheric Chemistry, Academic Press, 1976.

46.  Private Communication, Anthony Broadrick, AEQ-10, FAA, 800
     Independence Ave., Washington, DC  20591.

47-  Private Communication, Paul Crutzen, NCAR, Boulder, Colo.,

48.  K. Schutz, C. Junge, R. Beck and B. Albrecht, Studies of
     Atmospheric N20, J. Geophys. Res., 75 (1970) 2230-2246.

49.  C.N. Liong and M.A. Tabataba:, Effect of Trace Elements on
     Nitrification of Soils, J. Environ. Qual. , 1_ (1978)

50.  Private Communication, Dr. John Duxbury, Cornell University

51.  R. E. Rebbert and R. Ausloos, Decomposition of N?0 Over
     Particular Matter to be published in the Geophysical Research

52.  Private Communication, Julius Chang, Lawrence Livermore

53.  Private Communication, Harold Johnston,  Chemistry Dept.,
     University of California at Berkeley.

54.  D.F. Heath, A.J. Krueger and P.J. Crutzer, Solar Proton
     Event:  Influence on Stratospheric Ozone, Science, 197 (1977)

55.  Private Communication, D.F. Rolston, University of California
     at Davis.

56.  Private Communication, Jim Piebie; Michigan University.

57.  Private Communication, Mike McElroy, Harvard University.

58.  Private Communication, Paul Golden, NOAA.

                                   - 79 -
    59.   Private Communication, Art Hornsby, EPA, Korr Laboratory,
         Adon Oklahoma.

    60,   Private Communication, Jerry Mahlman, Princeton University.

    61.   Private communication, J.H. Bremner, Iowa State University.

    62.   A.M.  Blackmer and J.M. Bremner, Inhibitory Effect of Nitrate
         on Reduction of N90 to N9 by Soil Micro organisms, Soil Biol.
         Biochem, 1£ (1978J 187-191.

    63.   J.M.  Bremner,  Effects of Soil Processes on the Atmospheric
         Concentration of Nitrous Oxide, Nitrogen in the Environment,
         Vol.  1, Academic Press, 1978, pgs. 477-491.

    64.   I.G.  Poppoff,  R.C. Whitter, R.P. Tures, and L.A. Copone,
         An Assessment of the Effects of Supersonic Aircraft Operations
         on the Stratospheric Ozone Content, NASA Reference Publication
         1026, August 1978.

    65.   Private Communication, Roland Hauck, TVA - National Fertilizer
         Development Center, Muscle Shoals, Ala.

    66.   J.T.  Jastak and W. Greenfield, Trace Contamination of
         Anesthetic Gases:  A Brief Review, JADA, 95  (1977) 758-762.

    67-   Private Communication, Don Wuebbles, Lawrence Livermore

    68.   D.J.  McKenney, D.L. Wade & W.I. Findlay, Rates of N20
         Evolution from N-Fertilized Soil, Geophys. Res. Letts.,
         5_ (1978) 777-780.

    69.   J.R.  Freney, O.T. Denmead and J.R. Simpson, Soil As a Source
         or Sink For Atmospheric Nitrous Oxide, Nature 273  (1978)
MJ.S. GOVERNMENT PRINTING OFFICE: 1979 620-007/3750 1-3

                              TECHNICAL REPORT DATA
                        (Please read Instructions on the reverse before completing]
AjPreliminary Analysis of Nitrous  Oxide (N20)
including a Materials Balance
            5. REPORT DATE
               January 1979
C.  Richard Cothern
                                                   8. PERFORMING ORGANIZATION REPORT NO
                                                   10. PROGRAM ELEMENT NO.
                                                   11. CONTRACT/GRANT NO.
Environmental Protection Agency
Office of Toxic  Substances
401 M St.,  S.W.
Washington, D.C.   20460
                In House
 This preliminary analysis of  the  sources, sinks  and effect of nitrous
 oxide(N2O) in  the atmosphere  is a snapshot in time  of a rapidly  moving
 object.  Much  has been learned about this gas in recent years but much
 remains to be  determined.  The most reasonable residence time for N2O in
 the atmosphere is in the range 100-150 years.  Since the change  in atmos-
 pheric concentration of N20 is less than 1% per  year, this implies that
 the release rate is in the range  10-15 Mt N2O/year.   The combination of
 freshwater.- ocean and sediment sources appears to be about 7 Mt  N2O/year
 which means that the most that land/crop/fertilizer systems can  contribut
 is 8 Mt N2O/year.   The unknown in this balance is the concentration of
 land/crop/fertilizer systems.  Measurement of such  fluxes are needed. It
 appears that the only sinks for N20 are in the stratosphere.  The overall
 effect of N20  on stratospheric ozone is predicted to be small and could
 lead to an increase in stratospheric ozone.  The contribution of man
 made nitrous oxide appears to be  small at the present time but is
 expected to grow.   The overall problem is an important one on the time
 scale 20-50 years and is not  an immediate crisis.
                           KEY WORDS AND DOCUMENT ANALYSIS
                                       b.IDENTIFIERS/OPEN ENDED TERMS
                                                              c.  COSATI Field/Group

19. SECURITY CLASS (ThisReport)'
                                       20. SECURITY CLASS (Thispage)
                       22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE