United States
                   Environmental Protection
                   Agency
Atmospheric Research and
Exposure Assessment Laboratory
Research Triangle Park NC 27711
                   Research and Development
EPA/600/S4-90/019 Sept. 1990
&EPA          Project Summary
                    1987  Wet  Deposition  Temporal
                    and  Spatial  Patterns in  North
                    America
                   A. R. Olsen
                    The  purpose  of  this  research
                   project  is to determine the spatial
                   pattern  of wet deposition in North
                   America during  1987 and  to
                   investigate trends in wet deposition
                   in North America from 1979 through
                   1987.   The   project  considers
                   precipitation-weighted  average
                   concentration and deposition for nine
                   ion species:  hydrogen  ion, sulfate,
                   nitrate, ammonium, calcium, chloride,
                   sodium, potassium, and magnesium.
                   To accomplish the research an Acid
                   Deposition System (ADS) data base
                   has been established that  integrates
                   wet deposition monitoring  data from
                   several  regional  and  national moni-
                   toring networks.  Networks included
                   are the  National  Atmospheric
                   Deposition Program/National  Trends
                   Network (NADP/NTN),  the  MAP3S
                   precipitation chemistry  network, the
                   Utility  Acid Precipitation  Study
                   Program (UAPSP)  network,  the
                   Canadian Precipitation Monitoring
                   Network (CAPMoN), and the daily and
                   4-weekly Acidic Precipitation Moni-
                   toring  in Ontario Study  networks
                   (APIOS-D and APIOS-C).  The ADS
                   data base and statistical summaries
                   are available to  others to conduct
                   additional research studies.
                    This   Project Summary  was
                   developed by EPA's Atmospheric
                   Research and Exposure Assessment
                   Laboratory, Research Triangle Park,
                   NC, to announce key findings of the
                   research project  that is fully
                   documented  in a separate report of
the same title (see  Project  Report
ordering information at back).

Data Preparation and
Methodology
  The spatial  and  temporal  pattern
analyses include data  from  the six
networks identified  above.  These
networks operated 69 sites in 1979 and
increased the number of sites to 315 in
1987. During this period, each  network
applied  specific  protocols for selecting
site locations, field operation procedures,
laboratory analysis procedures, and data
quality  procedures.  In  addition,  they
participated in the development of unified
data screening and summary procedures
through  the  Unified Deposition Database
Committee (UDDC) (Olsen et at. 1989).
This project   applied  the  UDDC
procedures  in  selecting  sites and
preparing data summaries for statistical
analysis. The number of sites that meet
the UDDC criteria for annual summaries
increased from 39 in 1979 to 154  in 1987.
In  1987, approximately  50% of the sites
operating met the UDDC criteria. This is
a  reflection of  the  strictness  of the
criteria. The 1987 spatial analyses, 1979-
1987 trend analyses,  and 1982-1987
trend analyses  use specific subsets of
these sites.
  The report presents 30 mosaic  maps of
1987 annual and seasonal wet deposition
spatial  patterns. For  each  map (ion
species  and summary period), the sites
used are selected so that the sites meet
the UDDC criteria for that ion species and
summary period. The  number  of  sites

-------
used  range  from  154 to  240.  For
example, the number of sites for the 1987
annual sulfate maps  is  154.
  A primary consideration in selecting a
site for  temporal pattern analysis is the
continuous  operation of the site  for the
entire period  of  interest. Preferably, the
site would  also meet  the UDDC criteria
for each year during the period. Applying
these  criteria  severely  restricts the
number of sites. A slightly relaxed criteria
was  developed that applied to  multiple
years of data and still  met  the objective
of identifying  sites with good quality data
throughout the period.  The  criteria
resulted in 39 sites being selected for the
1979-1987  trend analyses and 148 sites
for the  1982-1987  trend analyses. The
1979-1987  sites have  six   sites  located
west of  the Mississippi river, two sites in
Canada,  two  sites  south  of  North
Carolina, and the rest  in eastern United
States.  The geographic pattern  of sites
for 1982-1987 trend sites  is  similar but
Ontario  Province has 42 sites.
  Mosaic maps display the 1987 spatial
patterns. The procedure for  developing
the maps  is  to estimate the  spatial
patterns  by using a  kriging  spatial
estimation  procedure that  includes
measurement error. A separate  kriging
analysis was applied to   eastern  and
western North America, since the regions
have distinctively  different  spatial
structure.  The surface is  estimated  for
approximately 3700 hexagons  with  an
area of 2000 square kilometers. The gray
scale map presentation is  based on the
percentiles  from  the  frequency
distribution of  the estimates  for the
eastern  and western United  States.
  The project used two approaches  for
trend analysis:  graphical  displays  of
annual  summaries  and statistical  trend
analysis  of  four-weekly  summaries.
Qualitative  assessment of trend  patterns
relies on time trend  graphical displays of
percentiles of annual summaries  for each
year  in the sites associated  with  either
the 1979-1987  or  the 1982-1987  trend
period.  Quantitative trend assessment
relies on the modified Kendall seasonal
tau  trend  statistical  test and Sen's
median slope  estimation procedure
applied to  four-weekly summary data at
each site.  The  results of  the  trend
analyses are  qualitatively  summarized
across sites.

Spatial Patterns for 1987
  Thirty mosaic maps  summarize and
display  the estimated spatial patterns in
1987 for five ion species  concentration
and  deposition  on  annual, winter and
summer basis. The five ion  species are
hydrogen  ion  (pH),  sulfate,  nitrate,
ammonium,  and calcium.  Simpson  and
Olsen  (1990)  contains the maps  and
extensive descriptions of the patterns.
  Areas of maximal sulfate  and  nitrate
deposition vary slightly from year to year,
but have generally remained within the
area bounded  by  Indiana on the  west,
southern  Ontario  on  the   north,
Pennsylvania and New  York  on  the east,
and  northern  Kentucky on the  south
(Figures  1  and 2).  In  1987, 10% of all
sites used for these spatial analyses had
deposition values exceeding  29  kg/ha for
sulfate and  21  kg/ha  for nitrate.  Both
maxima occur in  western Pennsylvania
and southern Ontario. In 1987, 5% of the
eastern United  States   is  estimated to
have  annual sulfate deposition greater
than 30 kg/ha;  25% with annual  sulfate
deposition  greater  than  22  kg/ha. For
annual nitrate deposition, 5% is greater
than 20 kg/ha and 25%  is greater than 13
kg/ha in the eastern United States. In the
western  United  States,  annual sulfate
deposition is estimated to be  less than 19
kg/ha and annual nitrate deposition less
than 12 kg/ha.
  Summertime  wet deposition of  sulfate
is more  than  double   wintertime  wet
sulfate deposition  in most areas  of the
eastern  United  States.  Summertime
deposition  of nitrates is also generally
higher at most  of the eastern United
States locations.  In  the west,   seasonal
patterns  are generally  similar,  although
higher uncertainty exists due to  terrain
influences on  precipitation  patterns and
the low density  of wet deposition sites
limit the  ability  to  develop  patterns  for
this  type  of  evaluation.   Sulfate
concentrations  in summer also  are
generally higher than in winter.  The area
of summertime maximal sulfate con-
centration  extends  west  along  the
southern  Great  Lakes  region as  far as
northeastern    Illinois.    Nitrate
concentrations for the  two seasons are
comparable, with the notable  exception of
the eastern Great  Lakes  region,  where
wintertime  concentrations can be  nearly
twice those of summertime.
  Similar to earlier years, the  area with
highest precipitation acidity (lowest pH) in
1987  is   centered  on  western
Pennsylvania,  eastern  Ohio,   northern
West  Virginia,  and southwestern New
York, where 5% of the sites used in the
analysis  have average  annual pH  values
less than 4.2  (Figure  3). The area  of
lowest pH  generally coincides  with the
areas  of  highest  sulfate,  nitrate,  and
hydrogen ion deposition (Figure  4).
  Maximal wet deposition of ammonium
and  calcium  occurs  in  two  regions
(Figures  5  and 6).  One  region  centers
over southern Ontario;  the  other over
Nebraska and Iowa in  the  Great Plains
region.
Trends from 1979-1987
  Figure 7  shows  that  annual  sulfate
deposition  gradually decreased  from
1979-1987 for the 39 sites in the analysis.
Although not presented, a similar gradual
decrease is present for sulfate deposition
for  the  144 sites during  1982-1987
period. The median sulfate deposition for
the   39  sites  decreased  from
approximately   26 kg/ha  in  1979 to
approximately 21 kg/ha in 1987. All other
ion  species  also decreased  from 1979-
1987, although  hydrogen,  nitrate,  and
ammonium  decreases  are  relatively
smaller.  Figure  7  shows  calcium
deposition as an example cation species.
Although calcium decreases, the pattern
differs from  that for  sulfate:  the  sites
appear more variable within  a year and
inter-annual  variation appears greater.
The median calcium deposition for the 39
sites  decreased from approximately  1.6
kg/ha in 1979 to 1.0 kg/ha in 1987.
  An  overview of  the  quantitative
statistical trend  results is in Table 1. The
most  striking feature  of the  comparison
across ion species  is that trend estimates
are consistent,  except  for  nitrate  and
ammonium concentration, with  over 50%
of the estimates indicating a decreasing
trend  during  the 1979-1987 and the 1982-
1987  periods.  The  percent  changes per
year  for  sulfate  and   calcium
concentration are greater over 1979-1987
than over 1982-1987; -2.2% and -6.1%
over  1979-1987 for sulfate and calcium,
respectively. In contrast,  the percent
changes per year for deposition are more
similar over 1979-1987  and 1982-1987.
For both periods, the percent change per
year  for sulfate and  calcium  deposition
exceeded  the percent  change  in
precipitation.
  From  1979 to  1987,  36 of  39 wet
deposition  monitoring  sites located
mainly in the eastern  United  States used
for  trend  analysis  show an  estimated
decrease in the concentration of sulfate in
precipitation,  although  most  of the
decreases are not  statistically significant.
Nevertheless, the  median  decrease  in
sulfate concentration for these 39 stations
during this period  is approximately  18%
(using  1987  as  base  year).  Sulfate
deposition shows an estimated decrease
for 34  of  39  monitoring  sites and  a

-------
       max - 18.7
      95%
      90%
      75%



      50%



      25%

      70%
6.9



3.7



2.4

1.8
Based on:
NADP/NTN, UAPSP, MAP3S,
CAPMoN, APIOS-C, APIOS-D
                                  Sulfate Deposition
                                  kg ha'1
       mm = 1.1
                                                                                               min =6.4
 Figure 1. Annual 1987 spatial distribution of sulfate deposition.
median decrease  for  all 39  sites  of
approximately 27%  from 1979 to 1987.

Conclusions
  The  research  project statistical
analyses  of  wet deposition  monitoring
network data enables several conclusions
to be made concerning the 1987 spatial
pattern of  major  wet deposition  ion
species and the trend observed from
1979-1987.
  Maximum deposition  of individual wet
deposition ion species are located  in the
northeastern  United States and  south-
eastern Canada. These areas  have  not
changed  significantly  in  eastern  North
America in the  past decade. The  actual
deposition values and  areal coverage of
those locations  have changed  from year
to year,  depending   on  variations  in
emissions and  meteorological conditions.
Spatial patterns of wet deposition are not
as well characterized in the western as in
the eastern region of North America. This
results from the density of the monitoring
networks  being  less in  the west than in
the east, from  the presence  of  large
terrain  features in  the  west,  and from
large separation of emission sources in
the west.
                             From 1979-1987,  more than half of the
                           39 trend sites have  estimated decreasing
                           concentration and deposition trends for
                           all ion  species,  except  ammonium
                           concentration.  For sulfate,  calcium,
                           chlorine,  sodium, potassium,  and
                           magnesium, more than 75% of  the  sites
                           have  estimated decreases;  and  for
                           calcium, sodium, and  magnesium  over
                           90%   of  the  sites  have estimated
                           decreases. In most cases the estimated
                           decreasing trends  are  not statistically
                           significant.  Nevertheless,  the  median
                           decrease  for these 39  sites in sulfate
                           concentration is approximately 18% and
                           in sulfate deposition is approximately
                           27%.
                             From 1982-1987,  less than 10% of the
                           144 trend  sites  have  a  statistically
                           significant  trend in  ion  species
                           concentration and deposition. However, a
                           median decrease of 0% to 4%  per  year
                           in concentration and of 0% to 7%  per
                           year in deposition  is estimated for the
                           sites. The median decrease for these 144
                           sites  in  sulfate  concentration  is
                           approximately 1% per year and in sulfate
                           deposition 3% per year.
                             The  available  monitoring data  and
                           statistical analysis procedures  limits
                                                           quantitative  generalizations of  the
                                                           estimated  magnitudes of decreasing
                                                           trends in concentration  and deposition of
                                                           ion species. The monitoring sites are not
                                                           regularly  spaced  throughout  eastern
                                                           North  America and consequently  some
                                                           regions have  more representation than
                                                           others. The statistical analysis procedure
                                                           addresses trends at individual  sites  and
                                                           discussion of trends across all sites does
                                                           not  consider  regional  trends  in  a
                                                           quantitative way.
                                                           References
                                                           Olsen, A. R., D. S. Bigelow, W. H. Chan,
                                                           T. L. Clark, M. A. Lusis, P. K. Misra, R. J.
                                                           Vet, and E. C. Voldner.  1989.  "Unified
                                                           Wet  Deposition  Data Summaries  for
                                                           North  America:   Data   Summary
                                                           Procedures and  Results   for  1984."
                                                           Atmospheric Environment, in  press.

                                                           Simpson, J. C., and  A. R. Olsen. 1990.
                                                           7987 Wet  Deposition  Temporal and
                                                           Spatial Patterns  in North America.
                                                           EPA/600/4-90/019, U.S.  Environmental
                                                           Protection  Agency,  Research Triangle
                                                           Park, North Carolina.

-------
  max = 11.9
                  Based on:
                  NADP/NTN, UAPSP, MAP3S,
                  CAPMoN, APIOS-C, APIOS-D
                                                                                                   mm  = 6.1
Figure 2. Annual 1987 spatial distribution of nitrate deposition
   max =4.6
  95%
  90%

  75%
  50%


  25%

  70%
4.8
4.9

5.0
5.2


5.3

5.5
Based on:
NADP/NTN, UAPSP, MAP3S,
CAPMoN, APIOS-C, APIOS-0
   min  = 6.1
       4.6


       4.7

       4.9

mm =5.2
Figure 3.  Annual 1987 spatial distribution of precipitation-weighted pH.

-------
                                                        Annual 1987
                                                        Hydrogen Deposition
                                                          ha'1
50%


         0.03

         0.02

  min =0.01
          Based on:
          NADP/NTN, UAPSP, MAP3S,
          CAPMoN, APIOS-C, APIOS-D
                                                                                                 min - 0.04
Figure 4. Annual 1987 spatial distribution of hydrogen ion deposit/on.
                                                       Annual 1987
                                                       Ammonium Deposition
                                                       kg na''
75%


50%


25%

70%
•0.6


•0.5

•0.4
                 Based on:
                 NADP'NTN, UAPSP, MAP3S,
                 CAPMoN, APIOS-C, APIOS-D
   min  =0.3
                                                                                                 min =0.8
Figure 5. Annual 1987 spatial distribution of ammonium deposition.

-------
   max = 2.5
  95%
  90%
  75%


  50%


  25%

  10%
 1.4


 0.7


•0.5

•0.3
Based on:
NADP/NTN, UAPSP, MAP3S,
CAPMoN, APIOS-C, APIOS-D
   mm =02
                                     Annual 1987
                                     Calcium Deposition
                                     kg ha'1
                                                                                                  min =0.3
Figure 6. Annual 1987 spatial distribution of calcium deposition.

-------
60-
50-
o> 40-
o
1 30-
Q.
0)
Q
03
m 20-
V)
10-
0-
e
e
I

MMMM

I
1
0
O
«
<


DM*

(
<
) <
> e


wwt
1
[ J
>
>
>
>
0
o
n T


1
[ 1
I
O
e
o +
fl ^

•mumm __
T
1 1
o
o
0 0
0
o
A
f
1 1

11
1
o
0
0 0

Io
iL


8
              7979   7980   7987   7982   7983   7984   7985   7986   7987
5.0-
TB 4.0-
-C
t
1 3.0-
CO
O
Q.
Q
B 20-
3 *-u
O
O
7.0-
0.0-
<
o J
o
T _
1





i '
9
e
9
i »
0 0
-, f t S
T j f T

1
1 1 -.

i*M*. 	
1
> . T * 1
8 e
o

0
T 1 °

T
. 1 I-J-, 1

° T T
0,0
              7979   7980   7987   7982   7983   7984   7985   7986   7987


 Figure 7. Temporal pattern of (a) annual sulfate ion wet deposition and (b) annual calcium ion wet deposition. Boxp/ofs are the 10th, 25th, 50th,
          75th and 90th percentiles of sites in the 1979-87 trend subset
Table 1. Median and Percent Change of Trend Estimates of Ion Concentration and Deposition for 7979-7987 and 7982-7987 Trend Sites
                                        Concentration (peq/L yr)
Deposition (meq/m2/yr)
1979-1987
Parameter
Precipitation
H +
SO42'
N03
NH4 +
Ca2 +
Est.
-
-0.05
-0.97
-0.07
0.72
-0.35
%
-
-0.4
-2.2
-0.7
0.7
-6.7
7982-7987
Est

-004
-0 28
042
005
-0.14
%

-0.2
-7.0
1.8
0.3
-2.6
7979-7987
Est.
-0.04
-0.09
-0.88
-0.20
-0.07
-0.29
%
-0.9
-0.6
-3.0
-7.2
-0.6
-5.5
7982-7987
Est.
-1.85
-0.50
-1.08
-0.10
-0.25
-0.32
%
-2.4
-2.9
-3.5
-0.8
-2.0
-6.3

-------
A.R. Olsen  is with Pacific Northwest Laboratory, Richland, WA 99352.
David M. Holland is the EPA Project Officer (see below)
The complete report, entitled "1987 Wet Deposition Temporal and Spatial Patterns
    in North America," (Order No. PB90 251 836/AS; Cost: $31.00, subject to
    change) will be available only from:
       National Technical Information Service
       5285 Port Royal Road
       Springfield, VA22161
       Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
       Atmospheric Research and Exposure Assessment Laboratory
       U.S. Environmental Protection Agency
       Research Triangle Park, NC 27711
United States                   Center for Environmental Research
Environmental Protection         Information
Agency                        Cincinnati OH 45268
 Official Business
 Penalty for Private Use $300

 EPA/600/S4-90/019

-------