x>EPA
             United States
             Environmental Protection
             Agency
             Environmental Sciences Research
             Laboratory
             Research Triangle Park NC 27711
EPA-600/3-78-100
December 1978
             Research and Development
Global Distribution of
Selected Halocarbons,
Hydrocarbons,
SF6, and N2O

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination of traditional grouping was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental  Health Effects Research
      2.  Environmental  Protection Technology
      3.  Ecological Research
      4.  Environmental  Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency  Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on  the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed  for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                               EPA-600/3-78-100
                                               December 1978
         GLOBAL DISTRIBUTION OF SELECTED
     HALOCARBONS, HYDROCARBONS, SF,,  AND NO
                                  o       2
                       by
                Hanwant B.  Singh
                   L. J. Salas
                   H. Shigeishl
                    E. Scribner
          Atmospheric Sciences Laboratory
                  SRI International
           Menlo Park, California  94025
              Grant Number 8038020-02
                Project Officer

                   John Spence
    Atmospheric Chemistry and Physics Division
    Environmental Sciences Research Laboratory
        U.S. Environmental Protection Agency
  Research Triangle Park, North Carolina  27711
  ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
      OFFICE OF RESEARCH AND DEVELOPMENT
     U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA   27711

-------
                               DISCLAIMER
     This report has been reviewed by the Environmental Sciences Research
Laboratory, U. S. Environmental Protection Agency and approved for
publication.  Approval does not signify that the contents necessarily
reflect the views and policies of the U. S. Environmental Protection
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
                                   ii

-------
                               ABSTRACT
     Northern and Southern hemispheric distributions of halogenated species,
hydrocarbons, SFg and N£0 are presented.  The atmospheric growth rates of
selected halocarbons and ^0 are characterized.  The fluorocarbon 11 and
12 global burden and hemispheric distribution is consistent with the view
that no significant sinks in the troposphere exist.  The north-south gra-
dients of fluorocarbon 11, 12, 113, 114, CCl4,and SFg suggest rapid global
mixing with an interhemispheric exchange rate of about one year.  Within
each hemisphere, these species are well mixed.  ^0 shows the least varia-
tions around the globe.  The global distribution of 0130013 is found to be
complex and suggests higher HO levels in the southern hemisphere and around
the equator, when compared to the northern hemispheric HO levels.  The glo-
bal distribution of C^Cl is almost uniform and a significant natural source
has been identified in the ocean.  It is also shown that large anthropogenic
primary or secondary sources of CH3C1 and 0014 exist.  Species such as CHC13,
CH2Cl2> 02^13, and 0201^ show very large north-south gradients.  The atmo-
spheric growth of fluorocarbons 11, 12, 0^0013, and 001^. appear to be con-
sistent with the emissions of these constituents.

     It is found that the world oceans may provide a sink for 001^ which
is about half as effective as the stratospheric sink.  For CH3C1 a signifi-
cant source appears to exist in the ocean.  Fluorocarbon 11 and 12 concen-
tration in Pacific seawater show some evidence of supersaturation.  The
lowest levels measured indicate that the surface ocean waters are in es-
sential equilibrium with the atmospheric concentrations of F12 and Fll.

     The Pacific Ocean (45°N-40°S) is found to be a significant source of
N20 despite the high oxygen content in the South Pacific.  The N20 varia-
bilities in the troposphere are small and an N20 residence time of as much
as 70 years is possible.  No increase of ^0 in the atmosphere was observed
over the last 27 months.  Our data imply that continued use of fertilizer
is unlikely to pose a threat to the stratospheric ozone.

     This report was submitted in fulfillment of Grant Number R8038020-02
by SRI International under the sponsorship of the U.S. Environmental
Protection Agency.  This report covers the period July 21, 1977, to
March 1, 1978.
                                  iii

-------
                               CONTENTS
                                                                  Page

Abstract	   ill

Figures	    vi

Tables	   vii

Acknowledgments	viii

     1.   Introduction 	     1

     2.   Objectives	     3

     3.   Summary of  Results	     5

     4.   Experimental Plan and Sampling and Analysis Protocol  .     9

               Air Sampling and Analysis Experiments 	     9
               Water Sampling and Analysis—Experiment 4 ....    11.
               Sampling Vessel 	    11
               In-Situ Air Sampling	    14
               Water Sampling	    16

     5.   Atmospheric Abundance and Variability of
          Trace Constituents	    19

               Nitrous Oxide 	    22
               Fluorinated Trace Constituents	    27
               Other Halogenated Trace Constituents	    37
               Light Hydrocarbons	    48

     6.   Pacific Seawater Measurements--Distribution and Flux  .    53

               Nitrous Oxide 	    53
               Halocarbons	    62

     7.   Conclusions.	    68

REFERENCES  .  .	    71

-------
                                FIGURES
Number
   1     Map Showing the Sampling Locations  for  the
         Four Experiments	    10
   2     Stainless Steel Canister for  Collecting Grab
         Samples	    12
   3     One-Liter Glass Sampling Vessel  ...  	    13
   4     Air Sampling in Stainless Steel  Vessel	  .    15
   5     Deepwater Sampling in the Pacific Ocean 	    17
   6     Global Distribution of N20	    23
   7     Growth of N20 with Time	    25
   8     Global Distribution of F12	    28
   9     Growth of F12 with Time	    29
  10     Global Distribution of Fll	    31
  11     Growth of Fll with Time	    32
  12     Global Distribution of F113	    33
  13     Global Distribution of F114	    35
  14     Global Distribution of F21	    36
  15     Global Distribution of SF&	    38
  16     Global Distribution of CCl4	    39
  17     Growth of CC^ with Time	    41
  18     Global Distribution of CH3CC13	    42
  19     Growth of CH3CCl3 with Time	    44
  20     Global Distribution of CH3C1	    45
  21     Global Distribution of CH3I	    47
  22     Global Distribution of City	    49
  23     Global Distribution of C2H6	    51
  24     Global Distribution of C2H4	    52
  25     N20 Variation with Depth in the  Pacific Ocean	    56
  26     N£0 Supersaturation in the Pacific  Ocean
         (46°N to 40°S)	    59
                                 vi

-------
                                TABLES
Number                                                              Page
   1     Average Background Concentrations of Important
         Trace Constituents. .  ..	     6
   2     Average Atmospheric Growth Rates of Important
         Species from November  1975 to December 1977	     7
   3     Summary of the Average Concentrations of Measured
         Trace Constituents	    20
   4     Least-Squares Error Coefficients of Third-Order
         Polynomial Used to Define the Global
         Distribution of Trace  Constituents	    21
   5     Average Atmospheric Growth Rates of Important
         Species from November  1975 to December 1977	    26
   6     Ambient N20 Concentrations (ppb) in 1975 and 1977 ....    26
   7     N20 Solubility in Seawater	    53
   8     N20 Concentration in the Pacific Ocean	    55
   9     N20 Concentration in Pacific Surface Water	    58
  10     F12 Concentrations in the Pacific Ocean	    63
  11     Concentrations of Halocarbons in Surface Seawater ....    64
  12     Average Concentration of Halocarbons in the
         Pacific Ocean	    65
                                  vii

-------
                            ACKNOWLEDGMENTS

     This project was supported in part by federal funds from the United
States Environmental Protection Agency under Grant No.  R-8038020.  We
thank the U.S. Coast Guard and the crew of the Burton Island for their
cooperation and help during this study.  Suggestions and comments from
Dr. P. L. Hanst and Mr. J. Spence of EPA are gratefully appreciated.   We
especially thank Prof. E. Robinson of Washington State University for
providing us with air samples from the South Pole.
                                  viii

-------
                               INTRODUCTION

     This report presents the Phase II results obtained during the second
year of a continuing study dealing with the atmospheric fate of halogenated
compounds.  We attempted to characterize the global abundance of important
atmospheric trace constituents likely to affect the stratospheric ozone.
The abundance of these trace constituents was determined in the atmosphere
and in the ocean in both hemispheres.  These data are essential for obtain-
ing important information on the lifetimes of trace constituents, inter-
hemispheric exchange rates, and the strength of atmospheric and oceanic
sinks.  This information is also necessary for calculation of stratospheric
ozone depletion.

-------
                              OBJECTIVES
     The objectives of Phase II were:

     •  To determine the atmospheric distribution and abundance of
        selected halocarbons, hydrocarbons, SFg, and ^0 in the
        northern and southern hemispheres.

     •  To determine the ability of the oceans to act as a source
        or a sink of the important trace constituents.

The objectives of Phase II were designed to complement  the following over-
all objectives of this project:

     •  To determine the distribution, atmospheric loading, sources,
        and sinks of halogenated compounds, selected photochemical
        pollutants, and natural trace constituents with possible
        stratospheric impact.

     •  To use halocarbons to improve understanding of complex
        regional and global atmospheric transport phenomena having
        a bearing on pollution control strategies.

     •  To use halocarbons as reactive tracers to improve under-
        standing of natural tropospheric chemistry.

-------
                          SUMMARY  OF  RESULTS

     Global atmospheric budgets and distributions of a large number of
halogenated species, hydrocarbons, SF,, and N?0 were determined between
63° N and 90°S at different longitudes.  Background air samples were ana-
lyzed both in- situ and by collecting pressurized air samples in specially
constructed vessels.  Pacific seawater samples from 46°N to 40°S were also
analyzed; the samples were obtained from depths of 0 to 300 m at 50-m
intervals.
     Table 1 shows the background concentrations of important trace con-
stituents in the northern hemisphere (N.H.) and the southern hemisphere
(S.H.).  When a significant gradient exists within one hemisphere, the
average concentration is defined as the concentration that, when mixed
uniformly within the hemisphere, represents the total burden of the species
in that hemisphere.
     The north-south average concentration difference for inert fluorocar-
bons is about 5 to 10%, and each of the hemispheres is reasonably well
mixed.  We did not observe any evidence of important longitudinal varia-
tions within each hemisphere.  For N20, CC1,, CH3C1, and CH, the north-
south concentration differences are even smaller and vary between 0 and
47o.  Other species such as CH-CCl- and CLH, show a more complex latitu-
dinal distribution.  The latitudinal distribution of CH-CC1,,, for example,
is best described by the following cubic function:

          (PPt) = 89.710 + 0.818 L + 7.584 X 10"4 L2 - 7.894 X 10"5 L3  (1)
where ppt is parts per trillion, L is the latitude, varying from 0 to +90°
for the N.H. and 0 to -90° for the S.H.  Indeed, CH-CCl- distribution shows
a rapid drop from 20°N to 20°S latitude.  The analysis of CH^CCU distri-
bution supports our earlier findings of low global HO abundance, and higher
HO levels near the equator and in the S.H. than in the N.H.  We suggest
that this gradient is caused by atmospheric carbon monoxide (CO) , which
is an effective sink for HO.
                                   5

-------
                           Table  1

        AVERAGE BACKGROUND CONCENTRATIONS OF IMPORTANT
                      TRACE CONSTITUENTS
Compound
N20
CC12F2 (F12)
CC13F (Fll)
CC1-FCC1F- (F113)
/ z
CCIF-CCIF. (F114)
2. i
CHC12F (F21)
SF6
cci4
CH3CC13
CH3C1
CH3I
CHC13
CH2C12
C2HC13
C2C14
CH4
C2H6
C2H2
Concentration*
(ppt)t
N.H. Average
311 X 103
230
133
19
12
5
0.31
122
113
611
2
14
44
16
40
1430 X 103
1060
<200
S.H. Average
311 X 103
210
119
18
10
4
0.27
119
75
615
2
S3
20
<3
12
1390 X 103
524
<200
Global
Average
311 X 103
220
126
18
11
4
0.29
120
94
613
2
8
32
8
26
1410 X 103
792
<200
 For those species where  significant variations within each
 hemisphere were observed, the average concentration within
 each hemisphere is  the concentration that, when uniformly
 mixed in the  hemisphere, represents the total burden of the
 species  in that hemisphere.

fppt = ID'12 v/v.

-------
     While most trace halospecies seem to be of uniquely anthropogenic
origin, dominant natural sources of CH«C1 and CH«I are indicated.  Con-
trary to the currently held view, our measurements in urban areas suggest
that significant anthropogenic sources of CH«C1 and CC1, must exist.
     The N.H. and S.H. distributions of inert fluorocarbons and SF.
                                                                  6
suggest rapid global mixing, with an interhemi spheric exchange rate of
about one year.  The budgets of . Fll and F12, as measured by us, are com-
patible with an overall residence time of 45 years for Fll and about 60
years for F12.  This is consistent with our earlier assertion that there
are no significant tropospheric sinks (Singh, 1977a).  For other species
such as CHCl-j, CH-CU, C-HC1 , and C^Cl, a significant north-south gradient
is evident, and is indicative of rapid tropospheric removal processes.
     Table 2 shows the average growth rates of N90, CC1 F_, CC1-F, CC1, ,
                                                ^      £ L.     J      H-
and CH»CC1_ at around 40°N.  The average growth rates of halocarbons agree
with projected emissions.  It appears that the background levels of N?0
are unchanged.
     The N~0 distribution in the globe is both uniform and nonvariable.
The atmospheric variability of
of 20 to 60 years.
                                   (a = 0.7%) suggest a residence time
                                Table 2
             AVERAGE ATMOSPHERIC GROWTH RATES OF IMPORTANT
              SPECIES FROM NOVEMBER 1975 TO DECEMBER 1977
Compound
N20
CC12F2 (F12)
CC13F (Fll)
cci4
CH3CC13
Average Growth Rate*
0.9 ppb/yrt
18.5 ppt/yr
12.9 ppt/yr
2.3 ppt/yr
15.5 ppt/yr
0.3 («0) %/yr
10%/yr
12%/yr
2%/yr
17%/yr
              Based on data collected between 35°N - 40°N.
              More precise comparisons indicate an average
              change of -0.170 per year (see Table 6).

-------
     The Pacific Ocean is significantly supersaturated with N»0.   Average
surface saturation of 133% and a maximum saturation of 190% were  measured
between 46°N and 39°S.  The N«0 concentration below the mixed layer varied
from 1 to 7 times the surface concentration, with a maximum near  the equa-
torial region.  The super saturation below the mixed layers was marginal at
midsouthern latitudes.  An oxygen content of 3 to 4 ml/1 is not too high
for significant N^O production.  This observation suggests that either
nitrification processes that are efficient in the presence of relatively
high 0_ content play a more important role in N_0 production than hitherto
believed, or denitrification processes are poorly understood and  can pro-
ceed under conditions significantly different from anoxic.
     A global N»0 oceanic flux of 32.4 million tons per year has  been
calculated.  When one includes fresh-water sources, we estimate an N~0
source of 35 million tons per year.  This corresponds to a turnover rate
of 67 years, which agrees with the upper limit of 60 years estimated from
atmospheric variability data.  Since the stratospheric sink has been esti-
mated to result in a turnover time of 120 years, a unknown sink with a
N90 turnover time of 150 years can be inferred.  Soil may be such a sink.
As long as the ocean is the major source of N«0, any effect on N?0 abun-
dance from increased use of fertilizer is likely to be insignificant.  The
atmospheric abundance of N?0 in December 1977 was within 0.2% of the Sep-
tember 1975 abundance.
     The average Pacific surface water concentrations (ng/1) of
CC13F, CCl^, CH3C1, and CHC13 were found to be 0.28, 0.13, 0.40, 26.8, and
<0.05,  respectively.   Both CCl^F- and CC13F surface water concentration
indicated some supersaturation.  The lowest measured concentrations of
CCl^F- and CC1,.F in seawater were essentially in equilibrium with their
atmospheric concentrations.  In the case of CC1, , world oceans were found
to be a sink about half as effective as the stratospheric sink.  From the
limited data, it would seem that the world oceans could provide a CH»C1
source of about 5 million tons per year.  This source can provide a turn-
over rate for atmospheric CH«C1 of about 2 years.

-------
            EXPERIMENTAL PLAN AND SAMPLING AND ANALYSIS PROTOCOL

     The Phase II study consisted of four experiments.   In each case
samples were collected from the northern and the southern  hemispheres
and analyzed either immediately or later.  The highest  latitude in the
northern hemisphere from which samples were collected and  analyzed was
about 64°N.  In the southern hemisphere samples were collected up to
42°S, and additional samples were obtained from 90°S to obtain maximum
coverage.  Sampling always  began in the northern hemisphere and continued
into the southern hemisphere.  Briefly the four experiments were as fol-
lows:

 AIR SAMPLING AND ANALYSIS
 Experiment 1
          Pressurized air samples were collected between 64°N and 20°S
in specially treated 1-liter stainless steel (SS) and  glass vessels.   The
sampling vessel and the sampling procedure are discussed in detail below.
Typically, four samples--two with glass and two with stainless steel ves-
sels—were collected at each site.  Figure 1 shows the  locations from
which air was collected (Trip 1).  The air samples were always collected
at the cleanest possible locations under the most favorable meteorologi-
cal conditions.  A few urban samples were also collected.   Trip 1 sampling
continued  from 15 September to 26 October 1977.  These samples were ana-
lyzed after 26 October 1977.

Experiment 2
          Air samples were collected on-board the U.S.  Coast Guard ice
breaker, Burton Island, which traveled from Oakland, California to Welling-
ton, New Zealand, covering 36°N to 42°S  latitude between 120°W and 175°E
longitude.  The actual itinerary is shown in Figure 1 (Trip 2).  Air
samples were collected in SS vessels, as in Experiment 1.   This experiment

-------
100      120     140      I60E      160      160*     140      IJO      IOO      BO      60       4O       2OW      0       20E      4O      6O      80      KX>
                                                                                                        (15 Sept - 30 Oct 1977)
                                                                                                             Experiment 1
                                                                                  AZORES

                                                                                   LAS PALM AS
                    (20 Nov - 13 Dec 1977)
                     Experiment 2. 3, and 4
IOO      I2O     140      I60C
                                                                            60      40      20W     0       2OE     40      60       80       IOO
                    FIGURE 1    MAP  SHOWING THE SAMPLING  LOCATIONS  FOR THE FOUR  EXPERIMENTS

-------
was conducted from 20 November to 13 December 1977, and these samples were
later analyzed at our laboratories in Menlo Park.

Experiment 3

          This experiment was conducted during the same oceanographic
cruise and at the same time as Experiment 2.  The major difference in
Experiment 3, however, is that the air samples were analyzed in-situ with
two 3920 Perkin Elmer gas chromatographs, each equipped with two electron-
capture detectors and one flame ionization detector, and a coulometric
gas chromatograph that were installed on-board.
WATER SAMPLING AND ANALYSIS
Experiment 4
          Water samples from the Pacific Ocean were analyzed in-situ.
Much of the water analysis was conducted on Trip 2.  In addition, water
samples from 40°N to 46°N in the Pacific Ocean were also analyzed.  Sea-
water samples were collected from the surface (0 to 2 m) and from depths
of 0 to 300 m at intervals of 50 m.
SAMPLING VESSELS
     Both types of sampling vessels were so designed that they could be
easily pressurized to 40 psi.  Figures 2 and 3 show these sampling vessels.
The stainless steel sampling vessels were electrochemically polished to
have an inert surface and were commercially obtained from D and F instru-
ments.  The glass vessels were constructed at SRI International.  Prior
to use, they were heated to 250°C and flushed with ultra-pure helium that
was cryogenically cleaned.  A hundred volumes of the sampling vessel were
exchanged over 100 minutes (helium flow rate ?»1  liter/minute).   The  sam-
pling vessels were then checked for background contamination, and the pro-
cedure was repeated until the background contamination of the sampling
vessels was reduced to less than 2 to 370 of the expected background con-
centration of a given trace constituent.
     The sampling vessels were also placed in a contaminated room and re-
checked for pressure stability and background contamination.  Typically
                                   11

-------
                                                              L-653D83-5
FIGURE 2   STAINLESS STEEL CANISTERS FOR COLLECTING GRAB SAMPLES
                               12

-------
                                                    L-653D83-6
FIGURE 3   ONE-LITER GLASS SAMPLING VESSEL
                     13

-------
the vessels that maintained the pressure were not affected by room con-
tamination because of the high positive pressure maintained in these ves-
sels.
     Prior to air sampling, the sampling vessels were carried at a positive
pressure of helium («5 psi) to prevent any contamination enroute.  During
sampling each vessel was flushed with 100 to 150 liters of ambient air and
then pressurized to 40 psi with a special stainless steel Metal Bellows
compressor pump (Model MB 158), which could maintain continuous flow rate
of 25 liters/minute.  Figure 4 shows a sampling operation.  A 200-m elec-
trical cord was carried so we could sample away from the source of elec-
tricity, usually a farmhouse or an abandoned building near the ocean.  Two
blank samples were also carried to ensure that no contamination occurred
during sample collection and analysis.

IN-SITU AIR SAMPLING
     Aboard the Burton Island, in-situ air sampling was conducted using
a  stainless steel manifold with two inlet positions.  We hoped that con-
tamination from the ship exhaust could be avoided by switching from one
manifold inlet to the other if meteorological conditions turned unfavor-
able.  Both manifold inlets were at the front of the vessel, about 10 m
above sea level and 20 m apart.  At any given instant only the manifold
inlet that would obtain the cleanest possible air was operational.  The
decision to select the right manifold inlet was made on the basis of wind
data.  On extremely calm days (wind speed <5mph) it was impossible to
avoid exhaust contamination from either of the sampling manifolds, and
air samples were collected in glass syringes from the best possible loca-
tion.  If clean air samples could not be obtained, analysis of air samples
was terminated.  Because the Burton Island is round-bottomed and rolls
excessively, we could not monitor on highly stormy days.
     All air samples were analyzed for halocarbons, light hydrocarbons,
N_0, and SFfi.  The methods of analysis have been published elsewhere
(Singh et al., 1977a, 1977b, and 1977c).  Electron capture gas chroma-
tography was used for all species except the hydrocarbons, for which
                                  14

-------
-  •

                                                   §


                                                          -:r   ;
                                                      -
                                                      ;         5   -x^$*:
                                                                 ;;           *-!
                   FIGURE 4   AIR SAMPLING IN STAINLESS STEEL VESSEL

-------
flame ionization detector (FID) was used.   Cryogenic preconcentration of
air samples was conducted at liquid oxygen temperature prior to  analysis.
Typically, sample size was less than 500 ml.   In some instances,  direct
5-to-10-ml air injections were used; for example,  cryogenic  analysis  was
not necessary for N-O.  Secondary standards were routinely analyzed to
monitor any changes in instrument responses.   In the case of N-O ultra
high precision was sought, so each air injection was accompanied by a
standard air injection.  The hydrocarbon primary standards were  obtained
from Scott-Merrion.

WATER SAMPLING
     The task of water sampling from depths of 0 to 300 m was handled by
the Burton Island crew.  Paired samples were  obtained from 0 to  300 m at
50-m intervals.  Water temperature at these depths was also  measured.
Each water sample provided 0.5 liters of seawater and was usually analyzed
within an hour or two of collection.  Figure  5 shows the water sampling
being conducted.  Water sample analysis was conducted by equilibrating
the water with an equal volume of ultra clean air.  The constituent of
interest was then analyzed in air, and the corresponding equilibrium  con-
centration in water was calculated from available solubility data.  The
trace constituents in air and water were then added to obtain the water
concentration.
                                  16

-------
FIGURE 5  DEEPWATER SAMPLING IN THE PACIFIC  OCEAN

-------
                 ATMOSPHERIC ABUNDANCE AND VARIABILITY  OF
                          TRACE CONSTITUENTS
     The experiments went satisfactorily at all sites,  and  no  serious
problems were encountered.  Stormy weather occasionally made it  impossible
to operate on the oceanographic cruise, but such days were  few.   Some  of
the glass containers were broken enroute to SRI due to  airline handling,
but during Trip 1 samples were collected in quadruplicate,  so  these handi-
caps did not pose any serious problem.  Samples that clearly showed signs
of contamination were discarded.  Two urban samples collected  from Lisbon
were treated separately.  The analysis of data reported here is  by no
means complete.  Additional analyses will appear in future  reports and
publications.
     In the figures showing the atmospheric distribution of trace con-
stituents the following symbols have been used to identify  the sampling
techniques used to obtain the data:
     V  Trip 1, Experiment 1, stainless steel vessels
     A  Trip 1, Experiment 1, glass vessels
     O  Trip 2, Experiment 2, stainless steel vessels
     D  Trip 2, Experiment 3, in-situ air sampling and  analysis
     Table 3 summarizes the data on background concentrations  of the mea-
sured trace constituents in the northern and southern hemispheres.  For
species that showed concentration gradients within each hemisphere, the
average background concentration is defined as the concentration that,
when present uniformly in one hemisphere, represents the total burden  of
a given pollutant in that hemisphere.  To derive this uniform  background
concentration a weighted average of the latitudinal concentration varia-
tion has been taken.  A third-order polynomial was fitted to all of the
data, and the polynomial coefficients for each constituent  are shown in
Table 4.  In the calculations of averages and the polynomial fit, visible
outliers are deleted.
                                   19

-------
                                    Table 3

               SUMMARY OF THE AVERAGE  CONCENTRATIONS OF MEASURED
                              TRACE CONSTITUENTS
Compound
N20
CC12F2 (F12)
CC13F (Fll)
CC10FCC1F0 (F113)
2 Z
CC1F0CC1F. (F114)
2 2.
CHC12F (F21)
SF6
cci4
CH3CC13
CH3C1
CH3I
CHC13
CH2C12
C2HC13
C2C14
CH4
C2H6
C2H2
Concentration
(ppt)*
N.H. Average
311 X 103 (2.3 X 103)
230 (25.5)
133 (13.4)
19 (3.5)
12 (1.9)
5 (2.6)
0.31 (0.04)
122 (4.9)
113*
611 (83.7)
2 (1.0)
14 (7.0)
44 (14.0)
16 (8.0)
40 (12.0)
1430 X 103 (64.7 X 103)
1060*
<200
S.H. Average
311 X 103 (2.8 X 103)
210 (25.1)
119 (11.7)
18 (3.1)
10 (1.3)
4 (1.0)
0.27 (0.01)
119 (4.0)
75*
615 (103.0)
2 (1.2)
*3
20 (4.0)
<3
12 (3.0)
1390 X 103 (51.4 X 103)
524 (14.8)
<200
Global
Average
311 X 103
220
126
18
11
4
0.29
120
94
613
2
8
32
8
26
1410 X 103
792
<200
*        -12
 ppt = 10    v/v.
f
 Parentheses indicate standard deviation.

 For those species where significant variations within each hemisphere were
 observed, the average concentration within each hemisphere is the concen-
 tration that, when uniformly mixed within the hemisphere,  represents the
 JtataJLJiurden of the species in that hemisphere.
                                      20

-------
                                Table 4
   LEAST-SQUARES ERROR COEFFICIENTS OF A THIRD-ORDER POLYNOMIAL USED
        TO DEFINE THE GLOBAL DISTRIBUTION OF TRACE CONSTITUENTS
Compound
N20
CC12F2 (F12)
CC13F (Fll)
CC10FCC1F0 (F113)
/ L
CC1F-CC1F. (F114)
Z f.
CHC12F (F21)
SF6
CC14
CH3CC13
CH3C1
CH3I
CH4
C2H6
Coefficients of Polynomial*
a
3.112 (2)t
2.206 (2)
1.290 (2)
1.787 (1)
1.153 (1)
4.212
0.292
1.198 (2)
8.971 (1)
6.190 (2)
1.711
1.409 (3)
0.769
b
-1.451 (-2)
0.363
0.498
5.249 (-2)
5.591 (-2)
2.952 (-2)
9.221 (-4)
4.412 (-2)
0.818
0.697
-9.422 (-3)
0.828
9.926 (-3)
c
-3.325 (-4)
1.096 (-3)
-3.178 (-3)
2.346 (-4)
-4.059 (-4)
5.583 (-5)
-5.817 (-6)
6.337 (-4)
7.584 (-4)
-3.128 (-3)
6.931 (-4)
4.435 (-5)
6.526 (-5)
d
-8.675 (-8)
-1.574 (-5)
-8.510 (-5)
-6.662 (-6)
-7.693 (-6)
-3.038 (-6)
1.370 (-7)
4.967 (-6)
-7.894 (-5)
-1.204 (-4)
-5.878 (-6)
-8.113 (-5)
5.561 (-8)
   The polynomial is Y = a+bL4CL +dL ,  where Y  is  the  concentration  in
   ppt for all species except N20,  CH^, and  C2Hg,  for  which  it  is  ppb.
   The independent variable L is the latitude in degrees  and varies
   from -90° to +90° (N.H.  = 0 to +90°, S.H. =0 to -90°).
  ^3.112 (2) = 3.112 X 102.
The hemispheres are by no means well mixed when one is dealing with reac-
tive species, and we wish to emphasize that data collected during this
study are ground-level data.  Except in a few instances it is safe to
assume, however, that no significant vertical concentration gradients
exist.  This has been shown by Cronn et al. (1976).  In the following
discussion, we shall consider the atmospheric abundance and distribution
of the measured trace constituents in greater detail.
                                  21

-------
NITROUS OXIDE
     Figure 6 shows the distribution of N_0 between 63°N and 90°S.   It
seems clear that no significant variation exists.   The following data are
most important:
     Average 1LO concentration in the N.H. = 311 ±2.8 ppb
     Average N20 concentration in the S.H. = 311 ±2.3 ppb
     Average global N_0 concentration = 311 ± 2.6 ppb
Thus we calculate a 311 ppb N20 concentration in the globe with a 0.8%
overall standard deviation (a ).  Part of this variability is due to the
precision of our analysis, which was determined to be within about 0.3%
(a ), with standard air samples.  Such high precision can be obtained
  P
only if each air injection is followed by a standard air injection and
the effects of relative humidity (R.H.) are carefully subtracted.  The
data on N20 presented here has been corrected to a constant R.H. of 50%.
The R.H. corrections were made based on 3-hour average R.H. values.  All
samples analyzed within a given 3-hour period were subject to the same
correction.  Samples collected on Trip 1 were not corrected for R.H.,
but a gas chromatographic check indicated that the R.H. was probably not
far from 50% and that these errors could not be more than 0.27o (an   ).
                    2                                            R.H.
The real variance (cra) due to atmospheric variability can be defined as:

                2    2    2    2
               CTa = CTt - CTp - CTR.H.

                  = (0.8)2 - (0.3)2 - (0.2)2 = (0.7)2   .           (2)

     Thus, an atmospheric variation of about 0.7% (a ) can be calculated.
                                                    3.
Using the formula derived by Junge (1974) , a residence time estimate for
N90 can be obtained:
                               14
                                a
T    = — = 20 years   .                     (3)
 est   a                                    v '
However, as Hahn and Junge (1976) have stated, this estimate is probably
low by as much as a factor of 3, primarily because the atmospheric vari-
ability of N~0 are approaching the precision of our analysis and the true
variability is probabily less than 0.7%.  An additional factor also must be
                                  22

-------
   340
    320
a
a
a
O
c\)
   300
   280
   260
S -80     -60
                                   I
                                         i
                          -40    -20      0      20

                                 LATITUDE — degrees
                      FIGURE 6  GLOBAL DISTRIBUTION OF N20
                                                         -a...
60
80
                                     23

-------
considered.  All of our data were collected within 15 m of the  ocean  sur-
face; for this reason a greater N-0 variability can be expected because
the ocean was supersaturated with N»0 (see Section VI-A).   Thus the possi-
bility that the residence time of N-0 in the atmosphere is 20 X 3  = 60
years cannot be ruled out.  Data collected by Weiss (private communication
of R. F. Weiss, Scripps Oceanographic Institute, California) support  an
upper limit of 84 years for the residence time of N00 (a  = 0.5%).
                                                   2.    a
     Figure 7 shows a plot of N20 levels at several locations between 1975
and 1977.  The mean values show that no significant change 0=i0.3% per
year) in the atmospheric abundance of N»0 has occurred during this period
(Table 5).  The standard deviations, however, are quite large,  probably
because extra care was not taken to obtain high precision, since such a
need was not identified.  Typically, calibration standards were injected
infrequently, and this is probably the single most important reason for
higher standard deviations.  In the past, corrections for  R.H.  changes,.
which could account for an error as large as ±0.5%, were also not  made.
     To remedy these deficiencies, air samples collected in 1975 were
compared with air samples collected and analyzed during Trip 2. Two  sam-
ples of clean air collected in September 1975 at a clean coastal site in
California  (»40°N)  were used.  The N-0 in these 1975 samples was compared
with that in ambient air between 24°N and 35°S.  Extreme care was  taken
to obtain maximum precision.  Since the R.H. in the two samples of 1975
air was measured only roughly at about 50%, all ambient measurements  of
N-0 were reduced to the 50% R.H. value.
     Table  6 shows the N-0 levels in the September 1975 air sample and
in ambient  air.  It is clear that the average N_0 concentration of 311.8
ppb in December 1977 is about 0.27» less than the September 1975 value of
312.3 ppb.  If one considers the standard deviations around the mean, a
change of +0.57o to -0.8%, in the atmospheric abundance of N~0 over  a 27-
month period can be calculated.  Since this change is within the atmo-
spheric variability of N?0, a more definitive conclusion cannot be drawn.
We can, however, safely conclude that no significant change (±0.5%)  in
the atmospheric abundance of ELO over the last two years has occurred.
This is contrary to the changes projected by Crutzen (1976), and
                                   24

-------
    340
    320
1
a
   300
   280
   260
       0
    (NOV. 1975)
                                          I     '    T    T      I     '

                                          AVG. GROWTH RATE -  0.9 PPB/YR
10         15
     TIME — months
20
                       FIGURE 7   GROWTH OF NjO WITH TIME
                               T
25
30
                                      25

-------
                     Table 5

  AVERAGE ATMOSPHERIC GROWTH RATES OF IMPORTANT
   SPECIES FROM NOVEMBER 1975 TO DECEMBER 1977
Compound
N20
CC12F2 (F12)
CC13F (Fll)
cci4
CH3CC13
Average Growth Rate*
0.9 ppb/yrt
18.5 ppt/yr
12.9 ppt/yr
2.3 ppt/yr
15.5 ppt/yr
0.3 («0) %/yr
10% /yr
12% /yr
2 %/yr
17%/yr
   Based on data collected between 35°N -  40°N.
  f
   More precise comparisons indicate an average
   change of -0.1% per year (see Table 6).
                     Table 6
AMBIENT N20 CONCENTRATIONS (ppb)  IN 1975 AND 1977
Latitude
24°N
18°N
14°N
4°N
1°N
4°S
15°S
25°S
30°S
35°S
Average
concentration (±a)
September
1975
312
311
313
312
313
313
313
312
313
311
312.3 ± 0.81
November-December
1977
310
312
313
311
312
312
313
310
313
312
311.8 ± 1.14
                       26

-------
McElroy et al. (1976).  It is possible, however, that the relative yield
of N»0 during denitrification is significantly lower than the current
estimates of 7 to 50% (Crutzen, 1976; Hahn and Junge, 1977).   It  is also
possible that the soil is not a source of N^O and that application of
fertilizer is likely to have virtually no impact on the atmospheric abun-
dance of N?0.  This hypothesis is developed further in Section VI.

FLUORINATED TRACE CONSTITUENTS
     Six important fluorinated species--CCl2F2 (F12) , CCUF (Fll) ,
CC12FCC1F2 (F113), CC1F2CC1F2 (F114) , CHC^F (F21) , and SFg--were mea-
sured.  While data on F21 have been provided, we are not sure if F21 is
an experimental artifact or a true atmospheric constituent.  Therefore,
all data on F21 must be considered tentative.

Fluorocarbon 12
          Figure 8 shows the latitudinal variation of F12.  A third-order
polynomial fits the data well.  F12 is nearly uniformly mixed in the
southern and northern hemispheres.  The following data were obtained:
          Average F12 concentration in the N.H. = 230 ±25.5 ppt
          Average F12 concentration in the S.H. = 210 ±25.1 ppt
          Average global F12 concentration = 220 ppt
          Thus there is a north-south gradient, with the S.H. average con-
centrations about 9.5% less than the N.H. values.  The data support our
observation that F12 is well mixed at least within each hemisphere.  Using
a simplified two-box model approach previously developed (Singh, 1977b),
we can easily conclude that the interhemispheric exchange time must be
less than 1.3 year.  While detailed calculations will be presented later,
the data in Figure 8, when coupled with extrapolated emissions data pro-
vided by E. I. DuPont de Nemours and Company (1977), do not disagree
significantly with a 50-year residence time calculated by Singh  (1977a).
          Figure 9 shows the growth rate of atmospheric F12 calculated
over a 27-month period.  This growth rate of 18.5 ppt/year (»10%per year)
is reasonably consistent with the actual growth in the F12 emissions.  The
growth data are summarized in Table  5.
                                   27

-------
IUUU
800
a 60°
a
1
1
(VJ
**• 400


200 <


1 ' 1 ' 1 ' 1
—
—

—
0 a
CBQ Q p Q

mffa
1,1,1,1,
S -80 -60 -40 -20 (
1 1 ' 1 ' 1 ' 1
—
—

—
a
a
itffa "Ho — ^^~ * — "^ '
QD ^D
,.1,1,1,1
) 20 40 60 80
           LATITUDE — degrees
FIGURE 8   GLOBAL DISTRIBUTION OF  F12
                28

-------
    300
    250
    200
c\J
    150
                                                       T
                                            MG. GROWTH  RATE - 18.5 PPT/YR
    100
       0          5
    (NOV. 1975)
10          15          20
     TIME — months
25
30
                       FIGURE 9   GROWTH OF F12 WITH TIME
                                       29

-------
          The urban samples of F12 from Lisbon contained F12 in concentra-
tions of about 800 ppt.  Similar urban concentrations have also been re-
ported in the U.S. (Singh, et al., 1977b, 1977c).

Fluorocarbon 11
          The global distribution of Fll was very similar to that of F12.
The following concentrations were obtained:
          Average Fll concentration in the N.H. = 133 ±13.4 ppt
          Average Fll concentration in the S.H. = 119 ± 11.7 ppt
          Average global Fll concentration = 126 ppt
Figure 10 shows the global variation of Fll; a north-south concentration
difference of about 10.57o can be calculated.  This gradient is consistent
with an interhemispheric exchange rate of less than 1.3 year.   Figure 11
clearly shows that, on the average, Fll has increased at a rate of 12.9
ppt/year (?»1270 per year).  This rate of growth agrees with available emis-
sions data when a residence time of 30 to 40 years is assumed.
          Our measured Fll distribution is inconsistent with that reported
by Lovelock (1973) , but agrees reasonably well with data reported by
Rasmussen et al. (1976).  The calculated burden of Fll based on the north-
south distribution provided by Lovelock (1973) suggests much lower concen-
trations of Fll in the S.H. than we found.  The net effect of using Fll
concentrations lower than ours is that in any analysis of Fll residence
time, unusually low residence times can be calculated.  This is a major
cause of disagreement between the residence time of 10 to 20 years calcu-
lated by Jesson et al. (1977) and that of 30 to 45 years calculated by
Singh (1977a).

Fluorocarbon 113
          The latitudinal distribution of F113 is shown in Figure 12; a
third-order polynomial fit is also shown.  The following data were calcu-
lated:
                                   30

-------
   500
   400  -
-  300
a
a
   200
    100
           I
        S -80    -60
                                      %
                                         Do
                          I
                                                           I-*.
                                                                v
   -40    -20     0      20      40


          LATITUDE — degrees




FIGURE 10   GLOBAL DISTRIBUTION OF F11
60      80
                                    31

-------
    200
    180
    160
_   140
a
a
    120
    100
    80
    60
       O

    (NOV. 1975)
                                                      T
                                           AVG. GROWTH RATE  =  12.9 PPT/YR
10          15

     TIME — months
20
25
30
                       FIGURE 11   GROWTH OF F11 WITH TIME
                                       32

-------
    100
a
a
    80
    60
    20
                    I    r
        S  -80     -60
    -40     -20      0       20

            LATITUDE —  degrees



FIGURE 12  GLOBAL DISTRIBUTION OF F113
60
80   N
                                       33

-------
          Average F113 concentration in the N.H.  = 19 ± 3.5  ppt
          Average F113 concentration in the S.H.  =18 ±3.1  ppt
          Average global F113 concentration = 18  ppt
A north-south concentration difference of about 57° was found.  Accurate
emissions data on F113 are not currently available, but the  residence
time of F113 should be similar to those of Fll and F12.  We  calculate the
1978 global burden of F113 to be 0.6 million tons.

Fluorocarbon 114
          Figure 13 shows the latitudinal variation of F114.  The follow-
ing data were obtained:
          Average F114 concentration in the N.H.  = 12 ± 1.9  ppt
          Average F114 concentration in the S.H.  = 10 ± 1.3  ppt
          Average global F114 concentration = 11  ppt
The global average concentration corresponds to a burden of  0.3  million
tons.  Accurate emissions data for F114, however, are also not available.
Reasonable estimates suggest that the measured F114 burden is  comparable
to cumulative F114 emissions.  A hemispheric concentration difference of
167<> is also consistent with a fast interhemispheric exchange rate.

Fluorocarbon 21
          As stated earlier, we cannot yet confirm that F21  is a true
atmospheric constituent or an experimental artifact.   The global distri-
bution of F21 is shown in Figure 14.  The average concentration is about
4 ppt.  The lack of a large gradient in the N.H.  and  S.H. for a species
with such a low atmospheric residence time suggests that it  may be an
artifact, but we were unable to determine its nature.  Thus  the following
data are tentatively presented:
          Average F21 concentration in the N.H. = 5 ±2.6 ppt
          Average F21 concentration in the S.H. = 4 ± 1.0 ppt
          Average global F21 concentration = 4 ppt.
                                  34

-------
     50
     40  -
E   30 h
a
     20  -
      10
1 ' 1 ' 1 ' 1
a
o
oo

^~ °- wndJmlffl o* 7
' 	 oo Q
1,1,1,1,
S -80 -60 -40 -20 (
1 ' 1 ' 1 ' 1
0 *
a an
gP- ~~£ v Qf v - - , 1^
O O O y "~

1,1,1,1
) 20 40 60 80
                                    LATITUDE — degrees
                       FIGURE 13   GLOBAL DISTRIBUTION OF F114
                                         35

-------
    50
~   30
Q.
a
    20
     10
I    .    I
        S -80     -60
                                    I
                           l
,	I
  -40     -20      0      20

          LATITUDE  — degrees


FIGURE 14  GLOBAL DISTRIBUTION OF F21
                                               40       60      80   N
                                        36

-------
Sulfur Hexafluoride
          We have earlier reported a background concentration of SF, arid
                                                                   o
about 0.28 ppt in the northern hemisphere (Singh et al.,  1977b and 1977c).
Because of the very low concentration of SF,, extensive  data from other
sources have not been available.  Figure 15 shows the latitudinal distri-
bution of SF, and its concentration in the northern and  southern hemi-
            o
spheres.  The following data can be reported:
          Average SF, concentration in the N.H. = 0.31 ± 0.04 ppt
          Average SF  concentration in the S.H. = 0.27 ± 0.01 ppt
          Average global SF, concentration = 0.29 ppt
The hemispheric gradient of SF, of about 13% supports the hypothesis that
                              o
SF, is farily well mixed in the atmosphere.  The atmospheric burden of
  o
SF, can be calculated to be about 7 thousand tons.   Because SF, has a
  O                                                           O
much longer lifetime than either F12 or Fll, it is an excellent tracer
of global mixing processes.  Continued emissions of, and  increases in,
SF, could result in deposition of sulfur at altitudes above 50 km with
  6
unknown effects.
OTHER HALOGENATED TRACE CONSTITUENTS
     Eight other atmospheric trace constituents--CCl, , CH-CCU, CH-C1,
CH-I, CHCU, CH2C12, C2HC13, and C2Cl,--were measured.  Reliable data for
both northern and southern hemisperes were obtained only for the first
four.  Some sample contamination occurred in our sampling vessels for the
others.                                .

Carbon Tetrachloride
          Unlike fluorocarbons, CC1, is rather uniformly distributed
around the globe (Figure 16).  This is not surprising, since emissions
of CC1, in the -last decade have been very small, and the concentration
is fairly uniform.  A very small difference of about 2.57o exists between
the N.H. and S.H. burdens of CC1,.  The following data were obtained:
                                   37

-------
I.U
0.8
B. 0.6
a
£

!
0.2


1 ' 1 ' 1 ' 1
—
—
~~

.

1,1,1,1,
S -80 -60 -40 -20 (
LATITUDE -
1 ' 1 ' 1 ' 1
—
—
A
A 7
?"* v # X "A*"
—
1,1,1,1
) 20 40 60 80
— degrees
                                                                N
FIGURE 15  GLOBAL DISTRIBUTION OF SFg
                  38

-------
    200
    150  -
a
a
o
o
    100  -
     50  -
1 ' 1 ' 1 ' 1 '
1 O A n D
—
1.1.1,1,
S -80 -60 -40 -20 (
1 ' 1 ' 1 ' 1
7
^•*-V-*r~i— "f

1,1,1,1
) 20 40 60 80
                                   LATITUDE  — degrees
                          FIGURE 16  GLOBAL DISTRIBUTION OF CCI,
                                          39

-------
          Average CC1, concentration in the N.H. = 122 ±4.9 ppt
          Average CC1, concentration in the S.H. = 119 ±4.0 ppt
          Average global CC1,  concentration = 120 ppt
The background concentration of CC1, indicates a global burden of 3.2
million tons.  Our calculated burden of CC1, is somewhat higher than the
estimated emissions data compiled by Singh et al. (1976) , Altshuller
(1976), and Galbally (1976).  Secondary sources of CC1, exist, but their
quantitative contribution to the atmospheric budget of CC1,  cannot be now
determined.  No natural sources of CC1, are known.
          The increase in CC1,  in the atmosphere over the last few years
has been small.  Figure 17 shows the average growth rate suggested by
our data.  The growth rate of 2.3 ppt/year («*2%per year) is in good agree-
ment with the growth rate projected by Singh et al. (1976) from emissions
data alone.
          It was also found that CC1, levels in the urban samples collected
in Lisbon were about 180 ppt, about 50% higher than the background level.
This observation is consistent with our earlier finding that CC1, levels
are 20 to 50% higher in urban areas (Singh et al. , 1977c).

Methyl Chloroform
          CHoCClo has been found in the atmosphere for several years, but
only recently has interest in its fate been shown.  This interest resulted
from two observations first reported by Singh (197_7a, 1977b) , who found
that the atmospheric burden of CH,CC1  was consistent with an 8-year resi-
dence time, contrary to all previous estimates that placed its residence
time between 1 and 2 years.  A longer residence time meant that CH,CC1_
could be a potential depleter of stratospheric ozone and that the average
hydroxyl radical (HO) abundance in the troposphere was lower than the
earlier estimates by a factor of 5 to 10.  Much of the analysis, however,
was based on point measurements.  Therefore a continuous data base showing
the latitudinal variation of CH-CCl, was needed.
          Figure 18 shows the global distribution of .CHoCCl- and a poly-
nomial fit to it.  Note that CHnCCl- shows a latitudinal distribution

                                  40

-------
   200
    ISO
    160
o.   140
a
    120
    too
    80
    60
       0

    (NOV. 1975)
                                                      T
                                           AVG. GROWTH RATE  *  2.3 PPT/YR
T
                                                                T	
         10          15

              TIME —  months
20
25
30
                       FIGURE 17   GROWTH OF CCI4 WITH TIME
                                       41

-------
250
200
 I   150
3
8
5   100
 50
        I
                                       Q D
S -80     -60     -40     -20      0      20
                          LATITUDE — degrees

             FIGURE 18   GLOBAL DISTRIBUTION OF
40
                                                                 60
                                                                     80
                                    42

-------
quite different from that of the fluorocarbons.   Up to about 30°N,  the
N.H. seems to be well mixed with CH_CC1 ,  with an average concentration
of about 120 ppt.  A fairly sharp decline  seems to occur between 20°N and
20°S; the levels of O^CCL, seem to level  off to about 70 ppt.  This decline
cannot be attributed to normal mixing processes, since fluorocarbons do
not show such a rapid decline.  A more plausible explanation of such a
phenomenon would be that the HO radical is more abundant around the equa-
tor because of the intense sunlight and higher water vapor concentration
in this region.  Two-dimensional models, however, would be required to
adequately model CH-CCl., distribution.
          The average, uniformly mixed CH»CC1_ concentrations of about
113 ppt in the N.H. and 75 ppt in the S.H. best describe the burden of
CH~CC1~ in each hemisphere.  When used in  a two-box model (Singh, 1977b),
this hemispheric distribution would indicate an HO concentration that is
higher in the southern hemisphere than in  the N.H.  More detailed calcula-
tions will be presented later.
          Figure 19 shows that the average growth rate of CH-CC1_ at about
40°N has been 15.5 ppt/year («17%per year).  Emissions data for CH3CC13
between 1976 and 1978 are not available, but an extrapolation of emissions
data provided by Dow Chemical Company (Singh, 1977b), suggests that an
increase of about 15 to 20%/year can be expected.
          The urban concentration of CHoCCl., in Lisbon was 250 ppt, about
twice the background concentration.

Methyl Chloride
          CH.,C1 has been measured at point sources in the N.H. by a number
of  researchers, but to the best of our knowledge, S.H. data have never
before been reported.  Figure 20 shows the latitudinal variation of CH-C1
that we found.  The following results were obtained:
          Average CH-C1 concentration in the N.H. = 611 ±83.7 ppt
          Average CH-C1 concentration in the S.H. = 615 ± 103.0 ppt
          Average global CH,C1 concentration = 613 ppt.
                                   43

-------
    200
    180
    160
w.

d   I2°
o
K)



    100
     80
    eo  -
       0

     (NOV. 1975)
                                            I      '      I      '      I    ^


                                            AVG.  GROWTH RATE =  I5.5 PPT/YR
10          IS

     TIME — months
20
25
30
                     FIGURE 19   GROWTH OF CH3CCI3 WITH TIME
                                       44

-------
  2000
   1500
a
a
   1000
x
o
   500
                    1   '    1
                             


-------
          The global distribution of CH-C1 is essentially uniform, with
marginally higher concentrations in the S.H.  The 357o larger ocean area
available in the S.H. may be a natural source of CH.,C1.   The higher HO
levels in the S.H. (Singh, 1977b) may also provide an additional sink for
CH-Cl.  The larger reservoir of CH,,C1, coupled with a lifetime we estimate
as 3 years, supports the contention that a significant natural source of
CH-Cl exists.  We show in Section' VI-B that the ocean may be the single
most important source of atmospheric CH«C1.
          The urban samples collected in Lisbon contained CH,Cl levels of
nearly 2200 ppt.  Our earlier studies in Los Angeles (Riverside) showed
an average CH.C1 concentration of 1500 ± 700 ppt (maximum of 3800 ppt),
about 2.5 times the background levels measured (Singh, etal., 1977c).  Thus
a significant urban source of CH,C1 exists.  Since primary emissions of
CH,C1 do not appear to be significant, secondary CH«C1 sources must be
inferred.  The possibility that automobile exhaust may be such a source
should be investigated.

Methyl  Iodide
          While CH.,1 was present most of the time, its quantitation was
difficult because an unknown chemical species or artifact tended to mask
the CHoI peak.  Thus CH-I measurements could be made only when this
artifact was resolved from the true CH.,1 peak.  Therefore the amount of
data was limited.  Figure 21 shows the global distribution of CH-I.  The
following data were obtained:
          Average CH.,1 concentration in the N.H. = 2.1 ± 1.0 ppt
          Average CHLI concentration in the S.H. = 2.3 ± 1.2 ppt
          Average global CH.,1 concentration = 2.2 ppt
CH«I  surface levels in the S.H. are somewhat higher than they are in the
N.H., with a global average of about 2 ppt.  Because of the extremely
short lifetime of CH.,1 (T < 5 days), large vertical variability should
be expected.  Therefore, the surface CH.,1  levels cannot be used to obtain
an estimate of iodine global burden.  The distribution of CH.I also sug-
gests an oceanic  source of CH.,1.

                                   46

-------
     50
                ,        ,
     40
S.   30
Q
to
     20
     10
            1,1
                                               O
        S  -80     -60     -40     -20      0      20      40

                                  LATITUDE — degrees
                       FIGURE 21   GLOBAL  DISTRIBUTION OF CH-jl
60      80   N
                                         47

-------
Chloroform, Methylene Chloride, Trichloroethylene,
and Tetrachloroethylene
          In-situ analysis requiring cryogenic preconcentration of samples
could not be performed on Trip 2 because of turbulent conditions on this
trip.  Therefore, all the data for these species were obtained via ana-
lysis of air samples collected in SS and glass vessels.   A good deal of
contamination of these species was indicated by their very high and often
variable concentrations.  The best data obtained suggested background
concentrations in the N.H. of 14 ± 7, 44 ± 14, 16 ± 8, and 40 ± 12 ppt
for CHC13, CH Cl , C-HCU, and C.Cl,, respectively.  The S.H. background
concentrations were very low:  ^3,20 ± 4, <3, and 12 ± 3 ppt for CHCl,,
CHLCl^, C2HC1_, and C-Cl,, respectively.  The CHC1, levels in the S.H.
were between 2 to 3 ppt.  It must, however, be emphasized that these aver-
ages are based on only 5 to 10 samples for each hemisphere that were
adjudged clean.  They do not adequately take the intrahemispheric vari-
abilities and gradients into account.

LIGHT HYDROCARBONS
     Analysis was conducted for a number of light hydrocarbons.  Due to
the limited number of air samples available and other practical reasons,
quantitation was possible only for methane (CH,), ethane (C«Hfi), and
ethylene (C^) .

Methane
          Figure 22 shows the global distribution of CH,.  It is clear
that CH,, like most other long-lived species, is practically uniformly
distributed.  The following data were obtained:
          Average CH, concentration in the N.H. = 1430 ±64.7 ppb
          Average CH, concentration in the S.H. = 1390 ±51.4 ppb
          Average global CH, concentration = 1410 ppb
Thus the S.H. CH, average concentration is about 370 lower than the N.H.
average concentration.  This, however, is within the variability of about
5% (a) associated with the average.  Even though large anthropogenic sources
                                  48

-------
  2500
  2000
   1500
5
o  1000 -
   500
                                             .?_-..--$--
: ---»•
    i
                                                                        I
        S -80     -60    -40     -20     0      20      40      60     80

                                 LATITUDE — degrees



                       FIGURE 22  GLOBAL DISTRIBUTION OF CH,
                                       49

-------
of CH, exist in the N.H.,  CH,  is largely of natural  origin with a very
high background concentration and a long tropospheric lifetime.   The very
long lifetime of CH,,  when taken with the interhemispheric exchange rate,
allows for a near-uniform distribution of CH,  around the globe.

Ethane
          CLH,, was also ubiquitous in the global atmosphere.   The global
           L D
distribution of C?H, was dissimilar to that of CH,,  probably  because C»H,
has a much shorter tropospheric lifetime than CH,  does.   Figure 23 shows
the nonlinear C-Hg distribution in the N.H.  At least in the  S.H., the
concentration of C-Hfi  is relatively uniform, averaging about  0.5 ppb.   In
the N.H. , however, the concentration approaches 2 ppb at mid-northern
latitudes but drops off to nearly 1 ppb near the equator.  The weighted
average concentration of C~Hfi that represents its burden in the N.H. is
calculated to be 1.1 ppb.   Thus a very large gradient of C^H, between the
N.H: and S.H. exists.   The latitudinal profile of C  H  suggests that there
are significant sources in the northern .hemisphere.

Ethylene and Acetylene
          C~H, distribution is shown in Figure 24.   We emphazine that some
samples were apparently contaminated and unusually variable.   The average
C?H, concentrations in the N.H. and S.H., for example, were 1.4 ± 1.1
ppb and 1.0 ± 0.6 ppb, respectively.  The air samples were probably inad-
vertently contaminated during sampling.  More recently, in-situ analysis
in the N.H. suggests that C_H, background concentrations at ground level
are less than 0.3 ppb.  The problem of sample contamination has also been
observed by others  (Cronn, D.R., WSU, private communication), but has yet
to be adequately resolved.  C»H, concentration near the surface may also be
highly variable and not representative of the tropospheric background.
          Acetylene concentration was found to be below the limit of our
measurement sensitivity in both the N.H. and the S.H. (<200ppt).  Limited
air sample availability did not allow further improvement in  this sensi-
tivity, which would have required sample sizes of 1.5 liters  or more.
                                  50

-------
a
a
a.
$
     10
      8
                                                      v



                                                      7
                                                   1
S  -80     -60     -40
                                  -20      0      20

                                  LATITUDE — degrees
                       FIGURE 23   GLOBAL DISTRIBUTION OF
         1
40      60      80   N
                                         51

-------
a
a.
a
(VJ
a
     15
     10
                                        •*
                                                      9
                                                      7
                                              °f A
S  -80     -60     -40     -20      0       20

                          LATITUDE — degrees



               FIGURE 24  GLOBAL DISTRIBUTION OF
                                                           40
60
80   N
                                         52

-------
            PACIFIC  SEAWATER MEASUREMENTS--DISTRIBUTION AND FLUX

NITROUS OXIDE
     Seawater from the Pacific ocean from 45°N to 39°S was  analyzed  during
Trip 2.  On eight occasions,  paired water samples were collected  from
depths of 0 to 300 m at 50-m intervals.   The water temperature was recorded
at the depth at which a paired water sample was collected.   The data from
the paired samples were averaged.   On two of these occasions only one set
of samples could be analyzed  because bad weather made further operations
impossible.  Surface water samples, however, were always  collected in
duplicate and the results averaged.  Surface water samples  from 16 dif-
ferent locations were analyzed for N»0.
     As soon as the water sample was collected, 25 ml of  the water sample
were equilibrated with an equal volume of nitrogen.  N_0  was then analyzed
in the gas phase, the corresponding water concentration was determined
from solubility data, and the amount of N20 in air and in water were added
to obtain the concentration of N»0 in the original water  sample.  Either
ultra-pure air or nitrogen could be used as the inert gas with no percep-
tible difference in the calculated N?0 concentration.  Table 7 shows the
                                Table 7
                       N20  SOLUBILITY IN SEAWATER
Temperature (°C)
Solubility (S)* of
N?0 in seawater
0

1.04
5

0.89
10

0.75
15

0.63
20

0.53
25,

0.44
30

0.38
    S =
Concentration at the ocean-air interface
    Atmospheric concentration at STP
                                  53

-------
solubility data used in our calculations--a corrected version of data
originally published by Junge et al.  (1971), which derived  them from ex-
periments conducted by Markham and Kobe (1941).   The correct data were
provided by J. Hahn (Max Planck Institute,  Mainz,  West Germany).
     Table 8 shows the measured water temperature and N»0 concentration
at depths of 0, 50, 100, 150, 200, 250, and 300 m, and the  calculated  per-
centage N20 saturation in seawater.  In every instance, a supersaturation
of N_0 was observed.  Figure 25 shows plots of N~0 concentration in sea-
water, N-O saturation, and water temperature as they vary with depth.
Clear evidence of N_0 synthesis below the mixed layer is visible.  The
N?0 production below the mixed layer  showed maximum near the equatorial
region, and only marginal supersaturation was observed at southern mid-
latitudes.  Saturation is most significant near 11°N.  It is least evident
at 39°S, perhaps because the reduced  biological activity results in condi-
tions not greatly favorable to denitrification.   Typically, however, be-
tween 45°N and 40°S in the Pacific, N»0 concentration below the mixed
                                     ^          t'
layer can be 1 to 7 times the surface concentration.  The North Pacific
is low in oxygen and therefore suited for N»0 formation via bacterial
denitrification processes, but in the Western South Pacific the oxygen
content is an order of magnitude higher than it is in the North Pacific
[4 to 5 ml (0?)/1].  We suggest that  N~0 can be synthesized in colder
waters where the oxygen content is considerably higher than 0.3 ml (CL)/1.
All these observations suggest one of two major possibilities.  Either
nitrification processes that are efficient under relatively high 02 abun-
dance play a more significant role in NO production than is now believed,
or denitrification processes are currently poorly understood, and denitri-
fication can proceed under conditions significantly different from anoxic.
The amount of N«0 synthesis in the southern midlatitudes does seem to  be
small when compared to that in the north midlatitude region of the Pacific.
     The surface water was always supersaturated with N^O,  and these waters
largely determine the flux of NO into the atmosphere.  The surface N_0
concentrations and saturation levels  are shown in Table 9.   Figure 26
shows a plot of the N«0 concentration in surface water, N~0 saturation
in surface water, average N_0 saturation in water between 200 and 300  m

                                   54

-------
                                                                                                Table 8




                                                                                     CONCENTRATION  IN THE  PACIFIC OCEAN


Depth
(m)
0
50
100
150
200
250
300
Date 10-29-76
Lat 45-35 'N
Long 125-55 'W
Water
Temp
CC)
13.5
12.2
10.5
9.6
9.0
8.6
8.2
.N,0
<|ig/l)
0.46
0.56
0.73
0.95
1.27
1.37
1.42
I
Sat
115
136
166
213
270
291
296
Date 10-28-76
Lat 41-35'N
Long 125-55 'W
Water
Temp
CO
14.2
13.2
12.0
10.7
10.0
9.5
9.2
N20
(ng/l)
0.46
0.58
0.71
0.96
1.08
1.16
1.18
%
Sat
120
145
169
218
237
252
257
Date 11-25-77
Lat 19°02 'N
Long 137'48'W
Water
Temp
CO
24.3
24.0
24.0
20.6
19.2
17.3
12.4
N,0
(lig/D
0.36
0.78
0.58
0.42
0.40
0.51
0.93
»
Sat
133
289
215
133
121
146
227
Date 11-27-77
Lat 11-52 'S
Long 142-05'W
Water
Temp
CC)
25.4
17.4
12.2
11.8
11.1
10.6
10.2
N20
(W5/D
0.41
0.83
2.48
2.83
2.09
2.10
2.31
I
Sat
152
237
605
690
486
477
519
Date 11-30-77
Lat 01-26'S
Long 150°10'W
Water
Temp
CO
27.6
27.5
27.5
25.8
19.0
13.6
13.4
N,0
(ng/l)
0.48
0.34
0.39
0.40
0.93
2.08
1.35
I
Sat
190
135
154
149
282.
527
342
Date 12-6-77
Lat 25-40'W
Long 167-33 'W
Water
Temp
CO
23.2
21.8
19.4
18.5
17.7
16.3
15.2
N20

0.36
0.40
0.40
0.48
0.42
0.46
0.49
\
Sat
124
133
123
143
120
127
130
Date 12-8-77
Lat 33-40'S
Long 170-50'W
Water
Temp
(°c)
18.5
14.6
13.5
12.7
12.2
11.7
10.7
N,0
Oig/D
0.40
0.44
0.50
0.58
0.59
0.58
0.65
7.
Sat
119
144
125
141
144
141
148
Date 12-11-77
Lat 39-05 'S
Long 179'13'W
Water
Temp
CC)
16.9
16.8
14.7
13.7
13.4
13.1
13.0
N20

0.42
0.47
0.43
0.47
0.47
0.45
0.41
I
Sat
118
132
113
119
118
113
103
Ln

-------
     100
               % SATURATION
     200     300     400     500
600    700  100     200
  % SATURATION
300    400    500    600    700
    50
   100
I  150
Q-
W
Q

   200
   250
  300
                          TEMPERATURE
                               "o
                              LATITUDE 45°35'N
                              LONGITUDE 125°55'w
                           N20 CONCENTRATION
                                  (/Llg/2)
                               N20 SATURATION
                                                                   LATITUDE 41 35 N
                                                                   LONGITUDE  125°55'w
             0.5     1.0     1.5     2.0     2.5     3.0  0     0.5      1.0     1.5     2.0     2.5     3.0
     100    200
    0
   50
   100
I  150
a.
LLJ
a
   200
   250
  300
   /I
 7!
 \  \
  »   \
-A   V
                     % SATURATION
                   300    400     500
            LATITUDE 19 02 N
            LONGITUDE 137°4S'w
               \
             \\l
                                  600    700  100    200
                           % SATURATION
                         300     400     500
                                                                               600     700
                                                                                          \\
                                                                       LONGITUDE 142°0s'w    \\
                    10             20
                   TEMPERATURE — °C
                                                30  0
                                                                  10
                                                                         20
            0.5     1.0     1.5     2.0     2.5     3.0  0     0.5    1.0     1.5    2.0     2.5     3.0
                                                                     N2O
                                                                                       30
                                                                 TEMPERATURE — C
                FIGURE 25   N20  VARIATION WITH DEPTH IN THE PACIFIC OCEAN
                                              56

-------
                 % SATURATION

100    200    300     400     500     600
                                                 700  100     200
                                                                % SATURATION

                                                              300     400    500
   50
   100
   150
Q.
UJ
Q
  200
   250
   300
                     LATITUDE 01°26'S
                     LONGITUDE 150°1()'w
-V.
    !i
                  N2O CONCENTRATION
        N20 SATURATION
                                     TEMPERATURE
                                                           Tl
                                                       I

                                                   \   '
                                                    \   \
                                                           LATITUDE 25°40'S
                                                           LONGITUDE 167°33'w
 0     0.5     1.0     1.5      2.0     2.5     3.0   0
                                                       0.5     1.0     1.5     2.0
                                                                  N2o (jug/2)
                                                                                          600     700
                                                                                           2.5     3.0
                 % SATURATION
100    200    300     400     500     600     700  100    200
   50
   100
   150
   200
   250
   300
0     0.5     1.0
                             LATITUDE 33 40 S
                             LONGITUDE 170°50'w
                                                        I
                      1.5     2.0
                        (/Ug/8)
                                                       IT
                                                        I
                                           2.5     3.0  0      0.5
                                                                      % SATURATION
                                                                    300    400     500
                                                                                          600     700
                                                                         LATITUDE 39°05'S
                                                                         LONGITUDE 174°13'w
                                                                     1.0 •    1.5     2.0     2.5     3.0
                                                                        NO (Alg/2)
                     10             20
                    TEMPERATURE — °C
                                             30   0
                                                                10
                                                                             20
                                                                    TEMPERATURE— C
                                                                                             30
                                       FIGURE 25   (Concluded)
                                                 57

-------
                                                      Table 9
                                        CONCENTRATION IN PACIFIC SURFACE WATER"
Date
10-29-76
10-28-76
11-25-77
11-26-77
11-27-77
11-28-77
11-29-77
11-30-77
12-01-77
12-02-77
12-04-77
12-05-77
12-06-77
12-07-77
12-08-77
12-11-77
Average
Lat
45°35'N
41°35'N
19°02'N
15°43'N
11°52'N
06°35'N
02°40'N
01°26'S
06°07'S
09°07'S
18°24'S
22°38'S
25°40'S
30°14'S
33°40'S
39°05'S

Long
125°55'W
125°35'W
137°48'W
140°05'W
142°05'W
145°14'W
147°43'W
150°10'W
153°13'W
155°35'W
161°42'W
164°26'W
167°33'W
171°18'W
170°50'W
179°13'W

Surface
Water Temp
(C°)
13.5
14.2
24.3
25.5
25.4
28.4
26.5
27.6
28.1
28.8
26.7
24.8
23.2
21.5
18.5
16.9
'
NoO Concentration
(1*8/1)
0.46
0.46
0.36
0.35
0.41
0.31
0.37
0.48
0.31
0.31
0.46
0.34
0.36
0.35
0.40
0.42
0.38
N20 Saturation
115
120
133
130
152
124
141
190
125
124
175
128
124
115
119
118
133
Fluxt x 10'12
(g/cm2 s)
0.173
0.224
0.347
0.323
0.561
0.258
0.438
0.955
0.264
0.261
0.808
0.291
0.261
0.164
0.211
0.204
0.359
Ol
oo
       Water from 0 to 2 meters.
       Flux calculation assumes a stagnant film thickness of 60

-------
oc     o
ui     *•



§ *£ 0"

g  X °~
EC
D
Z  2
   S  CC
   01  3
      <
      DC
      Q_
      5
2  ^  LU
O  OC  K


S  Z  S
<  ID  H
(Cut

?  §  i
S  U  m
   o
1000

 900

 800


 700


 600


 500



 400





 300
         200
         100
                          SURFACE WATER CONCENTRATION  (/^g/C x 103)

                          SURFACE WATER TEMPERATURE (°C x 10)

                          % N2O SATURATION  (200-300 m av.)

                          % N2O SATURATION  IN SURFACE WATER
            60    50    40    30    20   10    0    10    20   30    40    50    60

                           SOUTH                            NORTH
                                       LATITUDE — deg



     FIGURE 26   N20 SUPERSATURATION IN THE PACIFIC OCEAN  (46°N-40°S)
                                       59

-------
in depth, and the surface water temperature as they vary with latitude.
The average N90 between 200 and 300 m shows a decline from mid-northern
to low-northern latitudes, a maximum near the equator, and a rapid decline
in the southern hemisphere.
     From the data in Table 9, it is possible to calculate the flux,  F,
of N-0 into the atmosphere by using a simple film diffusion model:
                          F .      (cw _ cw
where D(T) is the film diffusion coefficient for N?0 in water and is a
function of temperature; Z is the stagnant film depth (Broecker and Peng,
1974) which has been measured to be 60 ± 30 p.m (|j,m = 10~  m) for the
              w                                            w
world ocean; C  is the concentration of N~0 in water; and C   is the N90
concentration in equilibrium with the atmospheric abundance of N90.
     For our calculations we have selected Z = 60 p,m and used the solu-
bility data presented in Table 7.  The variation of D(T) with temperature
was obtained from Broecker and Peng (1974).  Our choice of Z is the one
recommended by Hahn and Junge (1977), but the actual value of Z may be
higher than 60 urn for a biologically active gas such as ^0.  The mean ^0
concentration in the atmosphere was taken to be 311 ppb.
     The flux estimates calculated for each surface site are shown in
                                            -12     2
Table 9.  The average N20 flux is 0.359 X 10    g/cm  s.  The N20 satura-
tion of 133% we found in the Pacific (45 °N to 40°S) can be compared with
the 123% saturation found by Hahn (1974) in the Atlantic between 38°N and
50°N and the 132% saturation found by Rasmussen et al. (1976) in the
eastern tropical Pacific (31°N to 11°S).  Rasmussen's figure, however,
is probably too low, since he used the solubility data published by Junge
et al. (1971), which were found to be in error.  In 1969 Hahn (1974) mea-
sured an average saturation of 180% between 0 and 60 °N, but three of the
seven available data points were unusually high.  Preliminary data from
the Pacific ocean were provided by Craig and Gordon (1963), but the mea-
surement techniques were insensitive.  Our current data remedies this
deficiency significantly.  This is especially so, since the preliminary
data of Craig and Gordon have been extended by some recent investigators
(McElroy et al., 1976) to suggest that oceans are a net sink for N20.
                                  60

-------
     A comparison of our data with those of Hahn (1974) and Rasmussen
et al. (1976) suggests that our measured average N»0 saturation of 133%
in the Pacific is representative of that of global waters between 45°N
and 45°S.  About 74% of the total water surface is between these latitudes.
In the rest of the ocean, the surface supersaturation of N_0 is about half
this average; from the data provided by Hahn (1974) it is reasonable to
assume that this saturation is 115%.
                                           -12     2
     Assuming an average flux of 0.359 X 10    g/cm  s for the world ocean
                                            12
waters, we calculate a net flux of 38.5 X 10   g/year, or 38.5 million
                                        18   2
tons/year for an ocean area of 3.61 X 10   cm .  However, using 133%
saturation for the 74% of the seawater area between 45°N and 45°S
and 115%> for the remaining water surface, the average global N-0 flux is
32.4 million tons/year.  This estimate of ^0 flux from ocean waters is
less than half the best estimate of 70 million tons/year by Hahn and Junge
(1977) and about a quarter of Hahn's (1974) original estimate.  Fresh
                      16   2
water surfaces (2 x 10   cm ) can provide an N_0 source of 1 to 4 million
tons, and combustion sources are likely to be less than 4 million tons
(Hahn and Junge , 1977).  Our best estimate of N-0 flux from oceanic and
fresh water sources is about 35 million tons/year.  The atmospheric burden
                         3
of NO is about 2.34 X 10  million tons/year, which corresponds to an aver-
age N-O level of 311 ppb in air.  Using this atmospheric abundance and a
flux of 35 million tons/year, we calculate an N_0 residence time of 67
years.  This is surprisingly close to the upper limit of a 60-year resi-
dence time estimated from the atmospheric variability of N~0.  The strato-
spheric sink has been estimated to cause a turnover rate of about 120 years.
Since our data indicate that the N20 background is at a steady state, an
unknown sink with a residence time of about 150 years can be inferred.  The
possibility that this sink is the soil is suggested.  Soil as a net sink
for N?0 has also been supported by some recent experiments of Blackmer and
Bremner (1976), who find that although N»0 is synthesized in soil, it is
reconverted to N~ during diffusion upward.
     The saturation of the ocean waters supports the hypothesis that the
ocean is a source of N^O.  McElroy et al. (1976), based on a different
interpretation of Hahn's (1974) and Craig and Gordon's (1963) data,

                                   61

-------
suggested that the soil is the major source of N~0  and  that  the  ocean  is
a net sink.  This allows one to argue that N?0 abundance would increase
with increased use of fertilizer on soil,  but the evidence does  not  sup-
this contention.  As long as the ocean is  a major source of  N~0,  any ef-
fect on the N_0 abundance from increased use of fertilizer  is likely  to
be very marginal.  In fact the possibility that some of man's activities
may actually reduce N?0 (Ellsaesser, 1977) can also not be ruled.out.

B.   Halocarbons
     A preliminary attempt was made to measure halocarbons in seawater,
using the technique used for N_0.  Solubility data for  Fll and F12 were
obtained from Junge (1976), and for other  chlorinated species from Dilling
(1977).  The variations of CC1, and CH-C1  solubility with temperature  were
not available but were assumed to be similar to Fll variation.   It was
possible to obtain quantitative data for F12, Fll,  CC1,, and CH3C1.  The
data on CH-CCl., were unusually high and were discarded  as contaminated.
Table 10 shows the variation of F12 in seawater.  The average surface  con-
centration was 0.28 ng/1.  Typically, the F12 concentration  was  highest
at the surface and declined with depth.  Occasionally,  however,  an in-
crease in concentration was observed with depth.  We suspect that this
reflects contamination of water samples.
     Table 11 shows the surface concentration of F12, Fll, CCl^,  CH,C1,
and CHC1_.  The average concentrations of  individual species at  0 and
300 m are given in Table 12.  The average  surface water concentrations of
of F12, Fll, CC14, CH3C1, and CHCL^ are 0.28, 0.13, 0.40, 26.8,  and  <0.05
ng/1, respectively.  The corresponding average concentrations at 300m  are
shown in Table 12.  It can be seen from Table 11 that the concentration of
CH»C1 shows large variation.  Typically the values  seemed to be  higher near
the equator.  It also appears that CH-C1 is largely synthesized  in the upper
layers of the ocean waters, but the data base is too limited to  draw any
firm conclusions.  CHCl- could not be identified in seawater and we  must
conclude that its concentration in seawater is less than 0.05 ng/1.  Since
we know the surface water concentration of halocarbons, we can determine
                                  62

-------
               Table 10




F12 CONCENTRATIONS IN THE PACIFIC OCEAN
Depth
(m)
0
50
100
150
200
250
300
Date 11-25-77
Lat 19°02'N
Long 137°46'W
ng/1
0.40
0.20
0.32
0.15
0.15
0.22
0.20
Temp
24.3
24.0
24.0
20.6
19.2
17.3
12.4
Date 11-27-77
Lat 11°52'N
Long 142°25'W
ng/1
0.32
0.23
0.23
0.56
0.42
0.24
0.07
Temp
25.4
17.4
12.2
11.8
11.1
10.6
10.2
Date 11-30-77
Lat 01°26'S
Long 150°10'W
ng/1
0.35
0.25
0.28
0.33
0.31
0.24
0.24
Temp
27.6
27.5
27.5
25.8
19.0
13.6
13.4
Date 12-6-77
Lat 25°40'S
Long 167°33'W
ng/1
0.25
0.20
0.20
0.25
0.25
0.45
0.45
Temp
23.2
21.8
19.4
18.5
17.7
16.3
15.2
Date 12-8-77
Lat 33°40'S
Long 173°50'W
ng/1
0.19
0.20
0.22
0.25
0.24
0.27
--
Temp
18.5
14.6
13.5
12.7
12.2
11.7
10.7
Date 12-11-77
Lat 39°05'S
Long 179°13'W
ng/1
0.53
0.18
0.10
0.09
0.09
0.09
0.09
Temp
16.9
16.8
14.7
13.7
13.4
13.1
13.0

-------
                    Table 11

CONCENTRATIONS OF HALOCARBONS  IN SURFACE  SEAWATER
                     (0-2 m)
Date
11-25-77
11-26-77
11-27-77
11-28-77
11-29-77
11-30-77
12-01-77
12-02-77
12-04-77
12-06-77
12-07-77
12-08-77
12-11-77
Average
Lat
19°02'N
15°43'N
11°52'N
06°35'N
02°35'N
01°26'S
06°07'S
09°07'S
18°24'S
25°40'S
30°14'S
33°40'S
39°05'S

Long
137°48'W
140°05'W
142°25'W
145°14'W
145°14'W
150°10'W
153° 13 'W
155°75'W
161°42'W
167°33'W
171°18'W
173°50'W
179°13'W

Water
Temp
(°C)
24.3
25.5
25.4
28.4
26.5
27.6
28.1
28.8
26.7
23.2
21.5
18.5
16.9

Concentration
(ng/1)
F12
0.40
0.32
0.32
0.47
0.07
0.35
0.14
0.07
0.14
0.25
0.40
0.19
0.53
0.28 ± 0.15
Fll
—
--
--
--
--
--
--
0.15
0.24
0.07
0.10
0.10
0.12
0.13 ± 0.06
cci4
—
—
'
--
--
--.
--
0.41
0.45
0.35
0.38
0.40
0.40
0.40 ± 0.03
Cl^Cl
_ _
.
--
--
.
--
--
34.5
85.8
21.2
1.4
12.5
5.3
26.8 ± 31.2
CHC13
—
--
—
--
--
--
--
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05

-------
                                                    Table 12




                            AVERAGE CONCENTRATION OF HALOCARBONS IN THE PACIFIC  OCEAN

Depth
(m)
0
300
F12
Cone
(ng/1)
0.28
0.21
No. of
Data
Points
13
5
Fll
Cone
(ng/1)
0.13
0.06
No. of
Data
Points
6
2
cci4
Cone
(ng/1)
0.40
0.15
No. of
Data
Points
6
2
CH3C1
Cone
(ng/1)
26.8
3.3
No. of
Data
Points
6
2
CHC13
Cone
(ng/1)
<0.05
<0.05
No. of
Data
Points
6
2
Ui

-------
the ability of the ocean to act as a source or a sink for halogenated
species using the film diffusion model used for N90:

                               .   (CW-Q                       (5)

where F is the flux from the ocean to the air.
     Fluorocarbons 12 and 11.  Solubility data for F12 and Fll suggest
that, if the surface water is in equilibrium with the atmospheric burden,
the concentrations of F12 and Fll in water should be about 0.05 and 0.06
ng/1, respectively.  These concentrations are lower than the measured
average concentrations of 0.28 and 0.13 ng/1, so it appears that the
ocean water is supersaturated with F12 and Fll.  This means that either
the solubility data are inaccurate or the water samples were inadvertently
contaminated.  It is also possible that the ocean surface waters have been
contaminated by man-made activities on a global scale.  The lowest concen-
tration of F12 and Fll measured, 0.07 ng/1, is about what one would ex-
pect if the surface water were saturated with F12 and Fll.  If the surface
water is saturated, the ocean would be a relatively ineffective sink for
F12 and Fll but could act as a reservoir that contains less than 0.5% of
the atmospheric abundance of F12 and Fll in a steady-state situation.

     Carbon Tetrachloride.  The average surface water concentration for
CC1, was 0.40 ng/1, and the average ocean water temperature for these
samples was about 23°C.  It is thus possible to calculate the flux of
CC1, into the ocean using Eq. (5) with the following constants:

               D = 10"5 cm2/s
               Z = 90 pan
           SCC14 = 0.85.

A high Z is used because CC1, is rapidly absorbed in fatty tissues and
may be biologically active.  For such species the upper limit of the stag-
nant film thickness calculated from Radon data, (63 ±30 |j,m) is more ap-
propriate.  Using Eq. (5) we can calculate a flux:
                                  66

-------
                      "Fcci  = 2-8 x 10~16 s/01"2 s

If we assume that this flux is typical of all oceans, we can calculate
an exchange rate of 3.2 X 10   g/year.  The atmospheric burden of CC1,
                                               12
from our measurements is calculated as 3.2 x 10  g.   Thus the ocean should
be a sink for CC1, .  The residence time Tnri1  can be calculated as follows:
                                   12
                           •i o v in
                   Tcci  =         in
                    CC14   3.2 X 1010

Thus our measurements indicate that the ocean can provide a sink for CC1,
that is about half as effective as the stratospheric sink.
     Methyl Chloride.  It is obvious from Table 11 that the surface con-
centration of CH_C1 in the Pacific is quite variable, with values somewhat
higher near the equator.  The average surface concentration was 26.8 ng/1.
Using an SCH d of 2.65 (Dilling, :
earlier, we estimate from Eq. (5):
Using an SCH d of 2.65 (Dilling, 1977) and other parameters defined
                        F = 2.6 x 10"14 g/cm2 s                     (8)

Extending this to the world ocean body, we calculate an exchange rate of
        12
3.0 X 10   g/year.  From our measurements, the atmospheric burden of CH-C1
                                12
can be calculated to be 5.5 X 10   g.  The turnover time (TQJ QJ) from the
ocean source is:

                          = 5.5 X 1012 = l g  earg                  (9)
                    CH3C1   3.0 x 1012

Thus the ocean seems to act as a significant source of CH-C1 and provides
a turnover rate of about two years.
                                   67

-------
                               CONCLUSIONS

     We conclude that inert halocarbons, such as fluorocarbons 11, 12,
113, and 114 are reasonably well distributed around the globe, with
southern hemispheric background concentrations that are only marginally
different («10%) from the northern hemispheric background concentrations.
Most of the known sources exist in the northern hemisphere.  This implies
an interhemispheric exchange time of about 1 year.  Our data also suggest
that no significant sinks in the troposphere exist for Fll and F12.  The
atmospheric lifetimes of Fll and F12 are found to be much higher than
those calculated by Jesson et al. (1977).  This is because the southern
hemispheric  concentrations measured by us are significantly higher than
those suggested by Lovelock (1973) and used by Jesson et al.
     The abundance of HO in the southern hemisphere and near the equator
is high compared to HO levels in the northern hemisphere.  Atmospheric
carbon monoxide, a sink for HO, is invoked as the reason for an asymmetric
HO distribution in the two hemispheres.
     We found the atmospheric growth of N-0 to be negligible over a 27-
month period, and our data suggest a 60-to-70-year residence time of N~0
in the atmosphere.  The absence of any increase in N~0 in the atmosphere,
when coupled with a long lifetime, suggests that soil is not a significant
source of N»0 and may even be a net sink.  These observations imply that
continued use of fertilizers is unlikely to perturb the N?0 balance for
several decades, and thus no significant effect on the stratospheric ozone
can be expected.  The possibility that our current agricultural practices
may actually increase the stratospheric ozone can also not be ruled out.
     The oceans are also found to be sinks for atmospheric CC1, that are
about half as important as the stratospheric sink.  A large source of
CH»C1 is found in the Pacific Ocean.  Contrary to present belief, we con-
clude that significant primary or secondary anthropogenic sources of CH-C1
also exist.  The ocean surface water appears to be somewhat supersaturated

-------
with Fll and F12.  The lowest measured values of Fll and F12 would suggest
that ocean surface water is in essential equilibrium with their atmospheric
burden.  These observations suggest that while the ocean may be a small
reservoir for Fll and F12, it is an ineffective sink.  It is also possible
that the anthropogenic contamination of sea surface water has occurred on
a global scale.  The burden of halocarbons in the globe is growing at a
rate that is proportional to the man-made emissions.
                                  70

-------
                              REFERENCES
Altshuller, A. P., "Average Tropospheric Concentration of Carbon Tetra-
     chloride Based on Industrial Production,  Usage,  and  Emissions,"
     Env. Sci. Tech.. 10, 596-598, 1976.

Blackmer, A. M. and J. M. Breraner, "Potential  of Soil as  a Sink for Atmo-
     spheric Nitrous Oxide," Geophys.  Res.  Lett. . 3^,  739-742,  1976.

Broecker, W. S. and T. H. Peng, "Gas Exchange  Rates Between Air and  Sea,"
     Tellus. 26, 39-46, 1974.

Craig, H. and L. I. Gordon, "Nitrous Oxide in  the Ocean and in the Marine
     Atmosphere," Geochimica et Cosmochimica Acta. 27. 949-955, 1963.

Cronn, D. R., R. A. Rasmussen, and E.  Robinson, "Measurement of Tropo-
     spheric Halocarbons by Gas Chromatography-Mass Spectrometry," EPA
     Grant R-0804033, Washington State University, Pullman, Washington,
     1976.

Crutzen, P. J., "Upper Limits of Atmospheric Ozone Reductions Following
     Increased Application of Fixed Nitrogen to the Soil," Geophys.  Res.
     Lett.. 3, 169-172, 1976.

Billing, W. L., "Interphase Transfer Processes II:  Evaporation Ratio of
     Chloro Methanes, Ethanes, Ethylenes, Propanes, and Propylenes  from
     Dilute Aqueous  Solutions.  A Comparison with Theoretical Predictions,"
     Env. Sci. Tech.. 405-409, 1977.

E. I. DuPont de Nemours and Company, "World Production and Release of
     Chlorofluorocarbons 11 and 12 Through 1976, July 15, 1977."

Ellsaesser, H. W., "Has Man Increased Stratospheric Ozone?"  Nature.  270.
     592-593,  1977.

Galbally, I. E.,  "Man-Made Carbon Tetrachloride in the Atmosphere,"  Science.
     193. 573-576, 1976.

Hahn, J., "The North Atlantic Ocean as a Source of N20,"  Tellus, 2j?,  160-168,
     1974.

Hahn, J. and C. Junge, "Atmospheric Nitrous Oxide:  A Critical Review,"
     Z. Naturforsch. 32a. 190-214, 1977.

Jesson, J. P., P. Meakin, and L. C. Glasgow, "The Fluorocarbon-Ozone Theory-
     II:  Tropospheric Lifetimes—An Estimate of the Tropospheric  Lifetime
     of CC13F," Atm. Env.. U, 499-508, 1977.

                                  71

-------
Junge, C., B. Bockholt, K. Schutz, and R.  Beck,  "^0 Measurements  in Air
     and Sea Water Over the Atlantic," Meteor. Forsch.  Ergebrisse, Reihe
     B., No. 6, 1-11, 1971.

Junge, C. E., "Residence Time and Variability of Tropospheric  Trace
     Gases," Tellus. .26, 477-488, 1974.

Junge, C. E., "The Role of the Oceans as a Sink  for Chlorofluormethanes
     and Similar Compounds," Z. Naturforsch.  3.1 a, 482-487,  1976.

Lovelock, J. E., R. J. Maggs, and R.  J. Wade, "Halogenated  Hydrocarbons
     in and Over the Atlantic," Nature, 241,  194-196, 1973.

Markham, A. E. and K. A. Kobe, "Solubility of Carbon Dioxide and Nitrous
     Oxide in Salt Solution," J. Amer. Chem.  Soc. , 63_,  449-454,  1941.

McElroy, M. B., J. W. Elkins, S. C. Wofsy, and Y. L. Ying,  "Sources and
     Sinks for Atmospheric N20," Rev. Geophys. Space Phys.. 14,  143-150,
     1976.

Rasmussen, R. A., D. Pierotti, J. Krasnec, and B. Halter,  "Trip  Report
     on the Cruise of the Alpha Helix Research Vessel," Grant  No.  OCE  75
     04688 A03, Washington State University,  Pullman, Washington,  1976.

Singh, H. B., D. P. Fowler, and T. 0. Peyton, "Atmospheric  Carbontetra-
     chloride:  Another Man-Made Pollutant," Science, 192,  1231-1234, 1976.

Singh, H. B., "Atmospheric Halocarbons:  Evidence in Favor  of  Reduced
     Average Hydroxyl Radical Concentration in the Troposphere," Geophys.
     Res. Lett., 4, 101-104, 1977a.

Singh, H. B., "Preliminary Estimation of Average Tropospheric  HO Concen-
     trations in the Northern and Southern Hemispheres," Geophys.  Res.
     Lett.. 4, 453-456, 1977b.

Singh, H. B., L. Salas, and L. A. Cavanagh, "Distribution,  Sources, and
     Sinks of Atmospheric Halogenated Compounds," J. Air.  Poll.  Contr.
     Assoc., .27, 332-376, 1977a.

Singh, H. B., L. Salas, H. Shigeishi, and A.  Crawford,  "Urban-Nonurban
     Relationships of Halocarbons, SFg, N20,  and Other  Atmospheric Trace
     Constituents," Atm. Env. . _U, 819-828, 1977b.

Singh, H. B., L. Salas, H. Shigeishi, and A.  H.  Smith,  "Atmospheric Fates
     of Halogenated Compounds:  Second Year (Phase I) Summary  Report,"
     EPA Grant R-80380202, SRI Project 4487,  Stanford Research Institute,
     Menlo Park, California, 1977c.
                                  72

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
1. REPORT NO.
 EPA-600/3-78-100
                             2.
                                                           3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
 GLOBAL DISTRIBUTION OF SELECTED HALOCARBONS, HYDRO-
 CARBONS,  SF,,  AND N.O
             o       i
                                                           5. REPORT DATE
                                                            Dprpmber 1978
             6. PERFORMING ORGANIZATION CODE
 Hanwant  B.  Singh, L.J. Salas, H.  Shigeishi, and
 E. Scribner
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 SRI  International
 Menlo Park,  California  94025
             10. PROGRAM ELEMENT NO.
                 1AA603 AI-02  (FY-77)
                                                           11. CONTRACT/GRANT NO.
                                                             8038020-02
12. SPONSORING AGENCY NAME AND ADDRESS
 Environmental Sciences Research  Laboratory-RTF,NC
 Office of  Research and Development
 U.S.  Environmental Protection  Agency
 Research Triangle Park, North  Carolina  27711
             13. TYPE OF REPORT AND PERIOD COVERED
               Interim 7/77 - 3/78	
             14. SPONSORING AGENCY CODE
                EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
       Northern and Southern hemispheric distributions of  halogenated species, hydro-
  carbons, SF, and N_0 ai?e presented.   The atmospheric growth rates of selected halo-
  carbons and N_0 are characterized.   The fluorocarbon 11  and 12 global burden and
  hemispheric distribution is  consistent with the view that no significant sinks  in
  the troposphere exist.  The  north-south gradients of fluorocarbon 11, 12, ,113,  114,
  CC1,, and SF, suggest rapid  global  mixing with an interhemispheric exchange rate
  of  about one year.  Within each hemisphere, these species are well mixed.  N20  shows
  the least variations around  the globe.  The global distribution of CH»CC1. is found
  to  be complex and suggests higher HO levels in the southern hemisphere and around
  the equator, when compared to the northern hemispheric HO levels.  The global
  distribution of CH,C1 is almost uniform and a significant natural source has been
  identified in the ocean.  It is also shown that larg«  anthropogenic primary or
  secondary sources of CH Cl and CC1   exist.  Species such as CHC1-, CH Cl  , C HC1-,
  and C Cl  show very large north-south gradients.  The  atmospheric growth of fluoro-
  carbons 11, 12, CH CC1  , and CC1, appear to be consistent with the emissions of
  these constituents.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/Group
  *  Air pollution
  *  Halohydrocarbons
  *  Chemical analysis
     Troposphere
                             13B
                             07C
                             07D
                             04A
18. DISTRIBUTION STATEMENT

  RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)

  TTWnT.ASSTTi'TF.n  	
21. NO. OF PAGES
	  81
                                              20. SECURITY CLASS (Thispage)
                                               UNCLASSIFIED
                                                                         22. PRICE
EPA Form 2220-1 (9-73)
                                            73

-------