EPA/600/A-97/059

      A Technique for Estimation of Dry Deposition  Velocities Based on
                        Similarity  with Latent  Heat Flux

         Jonathan  E. Pleim*, Peter L. Finkelstein*, John F. Clarke*
Atmospheric Sciences Modeling Division, Air Resources Laboratory, National Oceanic and
            Atmospheric Administration, Research Triangle Park, NC 27711

                                       and
                              Thomas G.  Ellestad
National Exposure Research Laboratory, U.S. Environmental Protection Agency, Research
                             Triangle Park, NC 27711

Abstract
Field measurements of chemical dry deposition are needed to assess impacts and trends of
airborne contaminants on the exposure of crops and unmanaged ecosystems as well as for
the development and evaluation of air quality models. However, accurate measurements of
dry deposition velocities require expensive eddy correlation measurements and can only be
practically made for a few chemical species such as ozone and CO2- On the other hand,
operational dry deposition measurements such as used in large area networks involve
relatively inexpensive standard meteorological and chemical measurements but rely on less
accurate deposition velocity models.  This paper describes an intermediate technique which
can give accurate estimates of dry deposition velocity for chemical species which are
dominated by stomatal uptake such as ozone and SO2- This method can give results that
are nearly the quality of eddy correlation measurements at much lower cost. The concept is
that bulk stomatal conductance  can be accurately estimated from measurements of latent
heat flux combined with standard meteorological measurements of humidity, temperature,
and wind speed. The technique is tested for a field experiment where high quality eddy
correlation measurements were made in a soybean field in Kentucky. Over a four month
 on assignment to the National Exposure Research Laboratory, U.S. Environmental Protection Agency

-------
period, which covered the entire growth cycle, this technique showed very good agreement
with eddy correlation measurements for ozone.

Introduction

       Dry deposition is an important removal process of atmospheric trace chemical
species and therefore important in ecosystem research and assessment. However, accurate
measurements of dry deposition are very difficult and expensive to make on a long-term
operational basis. Currently, the most accurate and widely-used technique is eddy
correlation, which relies on fast response instruments and requires constant on-site
supervision. Consequently, such measurements are typically made as part of special
research studies for relatively short periods (weeks to months).
       In lieu of difficult and expensive direct measurements, existing dry deposition
networks and ecosystem exposure studies often combine simple meteorological  and
chemical measurements with dry deposition models to estimate deposition fluxes on a
continuing basis (Clarke and Edgerton 1993).  While such measurements give valuable
information on spatial distributions and long term trends of dry deposition fluxes,
comparison to eddy correlation measurements on an hourly basis show considerable scatter
(e.g. Padro et al, 1991; Meyers et al, in preparation).  Thus, for ecosystem and crop
exposure studies, as well as dry deposition networks, there is a need for more accurate
field measurements of dry deposition of a variety of chemical species.
       The dry deposition models used in existing networks are similar to the dry
deposition components of air quality modeling systems such as the Regional Acid
Deposition Model (RADM) (Chang et al, 1987), and  the Urban Airshed Model (UAM-V)
(Morris et al, 1992). The main difficulty with these models is the estimation of bulk
(canopy level) stomatal conductance, which is the dominant dry deposition pathway for
species such as ozone and SO2 in areas of active vegetation. Generally,  stomatal

-------
conductance is parameterized as functions of environmental factors such as solar radiation,



air humidity and temperature, and soil moisture conditions. Vegetation parameters such as



leaf area, vegetation coverage, surface roughness, and plant specific minimum stomatal



resistance are also considered. Dry deposition models differ in their details, but they have



the common problem of estimating the physiological functions of plants. However, since



plant species differ considerably in their response to environmental conditions, and



important environmental factors such as soil moisture are very difficult to realistically



estimate (or even measure), stomatal conductance estimates are not very accurate.



Therefore, a more direct way of determining bulk stomatal function is needed.



       During the growing season in areas of dense vegetation, water vapor flux is



dominated by the stomatal pathway. Therefore, measurements of latent heat flux can be



used to infer bulk canopy conductance, which can then be used to estimate dry deposition



velocity of some chemical species. This concept of similarity between chemical dry



deposition and evapotranspiration was applied to ozone deposition at a site in eastern



Colorado by Massman (1993). Since that site was sparsely vegetated, he used a two



source model (Shuttleworth and Wallace, 1985) to estimate the partitioning of latent heat



from plants  and soil. At more densely vegetated sites, such as forests or crops in the



eastern part  of North America, the canopy component should dominate. For example,



Baldocchi et al. (1987) tested a similar approach to estimate stomatal resistance using the



Penman-Monteith equation for a soybean canopy. In the current study, field measurements



over a  soybean field in southern Kentucky during the summer of 1995 are used to



demonstrate that ozone dry deposition velocities estimated assuming similarity to



evapotranspiration compare remarkably well to dry deposition velocities measured by eddy



correlation.  If this technique  performs as well at other highly vegetated sites, it could be



used for low-cost, accurate dry deposition measurements or field research on ecosystem



exposure.

-------
       Field deployment of low-cost measurement systems would include standard



meteorological measurements, which are needed to estimate aerodynamic and laminar layer



resistances, as well as measurements of latent heat flux.  Off the shelf systems, such as



energy balance Bowen ratio (EBBR) systems, could be used. Such systems require only



periodic attention (about once per week) and are therefore relatively inexpensive to operate.



However, the results shown here used an eddy correlation (EC) system for the



measurement of latent heat flux.  Therefore, the utility of a low-cost system specifically



designed for dry deposition measurement using latent heat similarity must be inferred from



comparisons between EBBR and EC systems. The current study is intended as  a proof of



concept for the latent heat similarity technique.  To the extent that EBBR systems, or any



other low-cost measurement system, can produce similarly accurate estimates of latent heat



flux as EC systems, the quality of results shown here should be achievable.  EBBR and EC



measurement systems each  have their respective advantages and disadvantages for realistic



measurement of latent heat flux as outlined by Kanemasu et al. (1992).  The First



International Satellite Land Surface Climatology Project (ISLSCP) Field Experiment



(FIFE) made over tall grass prairie in 1987 afforded an excellent opportunity for



comparison of the two types of systems as reported by Smith et al. (1992), Fritschen et al.



(1992), and Nie et al. (1992). Other studies have also shown the comparability of EBBR



and EC systems (Tanner 1988).








Field  Measurements








       A comprehensive set of meteorological and chemical flux measurements was made



in a soybean field in Keysburg, KY, which is near the Tennessee border about 60 km



NNW of Nashville.  Since these data include eddy correlation measurements of ozone flux



and dry deposition velocity, it provides a good testing environment for the development of



techniques to derive chemical dry deposition velocity estimates from latent heat flux

-------
measurements.  A description of the measurements system is presented by Finkelstein et al.



(1995) and Katuletal. (1996).



       The experiment site was set up in a soybean field at 36.65 N 87.03 W, about 2 km



west of Keysburg, KY. Elevation of the site is 585 meters with a gentle NW-SE slope of



about 1.5%. The instrument boom was 4.55 m. above the ground, pointing toward 208°.



Favorable wind directions were from 110 degrees through  270 degrees. Gently rolling



uniform soybean fetch extended out to at least 1500 m through the SE and southern



quadrants.  A corn field was adjacent to the soybeans, 140 m to the west at the closest



point. The boundary ran North-South.  Measurements with winds from  the Southwest will



have some influence from the corn, but are far enough away that the influence is small.



Examination of the data by wind direction showed no directional effects.



       The soybeans (Asgrow 5560) were no-till planted within wheat stubble June 13.



Dry deposition sampling was initiated June 22. An herbicide (Roundup) was applied July



4 which killed most of the weeds and slowed the growth of the beans for several days.



The beans went through a rapid growth period from July 10 through August 5.



Precipitation was adequate, LAI increased from 1 to about 6., midday leaf stomatal



resistance, measured by porometer, was about 40-80 s/m, and the  crop attained its



maximum height of 1.2 m.  Precipitation became very light after late July and by mid



August the beans were under water stress. Leaf stomatal resistance increased to about 800



s/m, and LAI gradually decreased to about 3 by the end of September and to 1 by October



11, when the beans were mostly stalks and pods (most of the leaves had  fallen). The corn



reached full height of 2.5 m in late July and was harvested  on August 25.








Theory







       Dry deposition velocity (Vj) is usually estimated from a series of resistances to




vertical transfer and surface uptake:

-------
           = (ra+rd+rs)"1                                                   (1)
where ra is the aerodynamic resistance, rd is the deposition or laminar layer resistance, and



rs is the surface resistance. Aerodynamic resistance is a function of turbulent transfer in the




atmospheric surface layer and can be estimated in several ways depending on the



instrumentation available.



       The deposition layer resistance accounts for diffusional transfer across a thin



laminar layer adjacent to surfaces.  Because of the no-slip condition, turbulent eddies



cannot penetrate to a surface. Therefore, there exists a thin layer of non-turbulent air where



molecular diffusion is the primary mechanism for transfer.  While this concept is not



relevant for momentum, it is relevant for any quantity which directly interacts with the



surface such as heat, moisture, and chemical deposition.  Therefore, for these quantities,



the addition of a resistance based on molecular diffusion is necessary.



       The estimation of the surface resistance is generally the most critical and most



difficult since it depends on chemical interactions with various surfaces as well as stomatal



uptake by the plants. In highly vegetated areas, the surface resistance for chemical species



such  as ozone and SO2 is dominated by stomatal uptake.  Since evaporation is also




dominated by the stomatal pathway, there exists an opportunity to derive bulk stomatal



resistance from measurements of latent heat flux. Surface water vapor flux (E) can be



estimated using a resistance model similarly to dry deposition flux:
where qa is the ambient specific humidity, qs(Tg) is the saturation specific humidity at the



surface temperature (skin temperature), p is the ambient air density, and rdw is the

-------
deposition layer resistance for water vapor. If latent heat flux is measured along with the
ambient humidity and the surface skin temperature and ra and r
-------
where u is the wind speed at the measurement height, and a is the standard deviation of


the wind direction in radians. The coefficient Ca depends on solar radiation and wind

                                                              7
speed such that Ca = 9 when solar insolation is greater than 10 W/m ; otherwise, Ca = 4


when u > 2 m/s, and Ca = 50-23u when u < 2 m/s.  At a site where fast response wind


measurements are available (i.e. sonic anemometers), ra can be computed using surface


layer similarity theory from eddy correlation measurements of friction velocity and sensible


heat flux. For this study, both methods were tested to evaluate the consequence of using


the simple system without fast response instruments.


       Surface skin temperature is very important to this technique since it defines the


surface saturation humidity, which is a component of the evaporative flux (Equation 2).


Since this measurement  was not made at Keysburg, the surface temperature (Tg) was


derived from eddy correlation measurements of sensible heat flux (H) as follows:
                                                                         (4)
where Ta is the air temperature at 3 m and Cp is the specific heat of air. Note that this


calculation also depends on ra, which in this case is computed from the fast response


measurements since Equation 4 is used only to compensate for the lack of a skin


temperature measurement which will be part of future systems. The deposition layer


resistance r
-------
            -Sc*                                                        (5)
            M.
                                                                                2
where Sc is the Schmidt number defined as the kinematic viscosity of air (y= 0.146 cm /s)


divided by molecular diffusivity (y/D),  For heat, the molecular thermal diffusivity is 0.206

   2                                                      2
cm /s; for water vapor, molecular diffusivity (Dw) is 0.244 cm /s; and, for ozone

                                     2
molecular diffusivity (DQ3) is 0,159 cm /s.


       Once the aerodynamic and deposition layer resistances have been estimated the bulk


stomatal resistance for water vapor can be computed by rearrangement of Equation 2:
                  •£,,
where EH- is evapotranspiration. In this study, the latent heat measurement divided by the


heat of vaporization (LE/Lv) is assumed to approximate Etr- Stomatal resistance for ozone


is estimated by weighting the stomatal resistance for water vapor by the ratios of molecular


diffusivity:
       It should be realized that Equation 6 is only practical when plants are actively

                                                                    2
transpiring.  Therefore, at night (when solar insolation is less than 10 W/m ), rstQ3 is set


to a relatively high constant value (5000 s/m) to represent resistance through closed


stomata.  This is not a serious drawback to the technique, since deposition velocities are


generally small at night and are primarily modulated by the aerodynamic resistance in the


nighttime stable surface layer. In this way, we were able to include all the nighttime data in


the analyses presented below.

-------
       If the stomatal pathway were the only important avenue for dry deposition of
ozone, then rstO3 could be used in place of TS in Equation 1. However, ozone and other
chemical species can deposit to surfaces such as leaves, stems, and soil. For example,
Massman (1993) included other pathways to the ground or exterior leaf surfaces for both
evaporation and ozone deposition in the sparse vegetation of eastern Colorado. Most dry
deposition models represent non-stomatal pathways as resistances in parallel to the bulk
stomatal resistance. However, there is very little agreement as to the magnitude of these
surface resistances. For example,  ground resistance for ozone as used in various models
ranges from 100 - 2000 s/m. Erissman et al. (1994) suggests that 100 s/m is appropriate
for dry soil while 500 s/m should be used for  wet soil since ozone has limited solubility.
Others have suggested the opposite effect of wetness such that resistance decreases for wet
surfaces. Meyers and Baldocchi (1993), on the basis of eddy correlation measurements
within a forest canopy, estimated surface resistance at the forest floor to be about 2000 s/m
regardless of wetness. Massman et al. (1994) found that, at times, surface resistance
decreased when dew was present.
       Resistances to ozone deposition on leaf exteriors or cuticles is also somewhat
controversial.  Cuticle resistance is generally expressed at the leaf level so that the bulk
effect at the canopy level is divided by the Leaf Area Index (LAI). Values of cuticle
resistance range from  1600 - 15000 s/m. In any case, it is a simple matter to provide for
non-stomatal  pathways in the calculation. These may be  made as elaborate as desired with
multiple branches of parallel and serial resistances just as in the various existing dry
deposition models. Given the uncertainty in these  processes, however, the current study
includes a constant resistance as the sum of all non-stomatal pathways (rsurf) in parallel
with the stomatal pathway such that Equation  1 is modified as:
                                                                         (8)
                                                                                 10

-------
       A further complication occurs when surfaces are wet from rain or dew. This can



affect both the moisture fluxes and dry deposition fluxes. The Keysburg field data



suggests an enhancement of ozone dry deposition during periods of surface wetness.



Therefore,  in this study, rsurf is specified at 200 s/m for wet surfaces and 600 s/m for dry




surfaces. Surface wetness may also affect moisture fluxes since there is clearly an



important non-stomatal source of moisture. When surface wetness is caused by dew,



Equation 6 tends to breaks down and does not give a useful-result. This is because the



difference between air humidity and saturation humidity at the surface temperature is very



small and often negative. Clearly, in these saturated conditions, moisture flux is not a good



indication for stomatal  function.  Before sunrise the stomata are closed so that Equation 6 is



not needed. However,  after sunrise the stomata may be open for a time before the dew



evaporates. During such times stomatal resistance should be specified in some other



manner, either as a constant or by some modeling approach.  Fortunately,  this is usually a



rather brief period during which the stable surface layer often presents the limiting



resistance to deposition. For this study, the bulk stomatal resistance for wet conditions is



limited by a minimum value of 100 s/m, while the minimum stomatal resistance for dry



conditions is set to 25 s/m.



       When surface wetness is caused by rain, there is the possibility that a significant



portion of the measured moisture flux is direct evaporation from wet surfaces and therefore



not stomatal.  In these cases, Equation 6 should underestimate stomatal resistance since the



evaporative flux overestimates the evapotranspiration.  However, during periods of



appreciable rainfall, the air quickly saturates and Equation 5 again becomes a poor estimator



of stomatal resistance.  Therefore, to improve estimates during wet periods, a back-up



model to parameterize stomatal resistance could be added.








Comparison  analyses

-------
       The method for estimating ozone dry deposition velocity assuming similarity to


evapotranspiration as outlined above is compared to deposition velocities derived from


eddy correlation measurements. Figure 1 shows a scatter plot of calculated versus


observed 30 minute average dry deposition velocities. The observed velocities are ozone


flux measured by eddy correlation divided by the measured ozone concentration at


approximately 4,5  in above the ground. This plot includes all data points from June 22 -


October 11, 1995 for which the quality of the measurements was considered acceptable,


Screening criteria included wind direction between 110°-270°, for uniform fetch over

                                              2
soybeans, and net energy balance within 100 W/m . The soybeans varied from a canopy


height of 0.15 m with a LAI of 1 at the beginning of the period to 1.15m and LAI of 5.75


at the peak growth stage in mid to late August and back down to 1,1 m with an LAI of 2 by


the end of the period.  Thus, the data include a variety of vegetation and soil moisture


conditions including nighttime and periods of rain.


       Figure 1 indicates that the moisture similarity technique shows considerable skill in


estimating ozone dry deposition velocity. With judicious constraints on the calculation,


such as a dry minimum bulk stomatal resistance of 25 s/m which is reasonable for


soybeans, the technique results in realistic values of deposition velocity between 0 and 1.8


cm/s. Point by point comparison shows excellent agreement considering the short


averaging time (1/2 hour) and the variety of conditions. A linear regression on Figure 1


gives a correlation coefficient of r = 0,82. Figure 2 shows a histogram of observed minus


predicted deposition velocities which shows that 56% of the data are within 0,1 cm/s and


92% of the calculated values are within 0.3 cm/s of the measurements.  The mean bias of


observed minus predicted is 0.0062 cm/s and the standard deviation of the bias is 0.18.


       As discussed above, the use of moisture flux measurements to estimate bulk


stomatal resistance is not very valuable in wet conditions where moisture gradients are very


small. Therefore, stomatal resistance is set to the minimum for most of the wet data points.


Also, at night the stomata are closed and the stomatal resistance is very high.  Thus,  the
                                                                                 12

-------
moisture similarity technique is important only for a fraction of the data points, namely


daylight, dry conditions, which is about half of the total dataset. The fact that the results


compare so well to EC measurements for the full data set suggests that for wet and/or


nighttime conditions the surface resistance is usually not the controlling process.


       Figure 3 shows a scatter plot of daytime dry conditions only. The linear regression


correlation coefficient is r = 0.86.  The criteria for this subset is that the solar insolation be

                   2
greater than 10 W/m  and that the wetness sensor indicate dry conditions. However, since


there was only one wetness sensor, there may be times, such as after a rain event or when


the morning dew is evaporating, when parts of the canopy  may still be wet when the sensor


indicates dry conditions.  Some of the most extreme outlying points in Figure 3 are


probably associated with  mistaken classification as dry by the wetness sensor. For


example the two highest predictions in the "dry" subset were during the day on August 5


which was quite wet from rainfall and all the other data points during this day were


classified as wet. Therefore, accurate determination of wetness of the leaves and ground is


helpful to this technique.


       A time series of measured and computed ozone dry deposition velocity for six days


in early August is presented in Figure 4.  The first four days show remarkable agreement


between the computed and measured values during both day and night. These days  had


mostly clear skies with light to moderate winds from the south  and southeast. This


suggests that under dry, clear sky conditions the technique  is nearly exact in reproducing


measured dry deposition velocities. The latter two days of this period were quite different,


starting on August 5 when the remnants of Hurricane Erin  passed through, dropping over


an inch of rain at the site. Other than the two outliers, which were probably misclassified


as dry, the predictions compared quite well with the observations even though the surfaces


were wet. The very windy conditions lead to high deposition velocities which were  well


simulated by the calculations of the aerodynamic and deposition layer resistances. August 6


was characterized by variable cloudiness, very light winds, and no rain but continuous
                                                                                 13

-------
surface wetness according to the sensor. During the middle of the day when the winds
were light, the calculations significantly underestimate the deposition velocities compared to
the measurements. It is curious that surface wetness was indicated for the whole day when
no rainfall was recorded and solar insolation was substantial. If these data points were
classified as dry, the predictions would be much closer to the measured values as shown in
Figure 5.
       To help assess how well this technique would perform in an inexpensive network,
a test was made using aerodynamic resistance computed from variations in wind direction-
(1 minute averages) according to Equation 3 rather than sonic anemometer measurements.
Using this estimate of ra in Equations 6 and 8 resulted in very similar estimates of dry
deposition velocity such that the linear correlation coefficient was r = 0.79 rather than 0.82.
Thus, simple measurements not requiring fast response instrumentation should be
sufficient for estimation of 1*3.

Discussion

       The technique described in this paper can be considered as a combination of field
measurement and modeling methods for discernment of dry deposition velocities at a
relatively low cost. By using similarity of gaseous dry deposition flux to moisture flux the
bulk stomatal resistance, which is the most important and difficult to model component of
the dry deposition process, is derived from  surrogate measurements. The advantage of this
technique with regard to field networks is that measurements of moisture flux can be made
much more inexpensively than can direct measurements of ozone flux. From a modeling
perspective, an estimate of bulk slomatal resistance derived from moisture flux is more
realistic and responsive than parameterizations based on functions of environmental factors.
Therefore, this technique is not meant replace current models but to augment them by
providing more accurate estimates of the stomatal pathway which is then used in
                                                                                14

-------
 combination with parameterizations of the other important dry deposition pathways.
 Similarly, this method will not replace direct eddy correlation measurements, and in fact
 relies on such studies for development and verification, but will enable relatively cheap
 high quality deployment of dry deposition networks.
       This study has demonstrated the potential of using latent heat flux measurements to
 estimate ozone dry deposition velocities. The data used here were from an intensive field
 study which included eddy correlation measurements of heat, moisture, and ozone flux.
 Clearly, the measurements envisioned for an inexpensive network would be different
 which may affect the accuracy of the results.  In particular, EBBR systems are not
 considered as accurate as eddy correlation for measurement of moisture flux. On the other
 hand, a system designed for this use would include skin temperature measurements, via IR
 radiometry, which should improve the calculations due to better estimates of surface
 humidity. Therefore, the next step is to deploy a system designed for this method,
 including EBBR and skin temperature, as part of a more extensive field experiment which
 includes eddy correlation measurements of ozone and moisture fluxes.  Another challenge
 is to extend this technique to other chemical species. Theoretically,  it should work for any
 species which has a significant stomatal pathway such as SO2- The  main obstacle to
 testing the technique for SO2 is making accurate eddy correlation measurements for
 comparison. Also, this work should be extended to other environments,  such as  other
 types of crops, grasslands, and forests.

 DISCLAIMER
 The information in this document has been funded wholly or in part  by the United States
 Environmental Protection Agency. It has been subjected to Agency review and approved
for publication. Mention of trade names or commercial products does not constitute
 endorsement or recommendation for use.
                                                                                15

-------
REFERENCES



Baldocchi, D.D., Hicks, B.B., Camara, P., 1987. A canopy stomatal resistance model for



       gaseous deposition to vegetated surfaces. Atmospheric Environment, 21: 91-101.



Chang, J.S., Brost, R.A., Isaksen, I.S.A., Madronich, S., Middleton, P., Stockwell,



       W.R. and Walcek, C.J., 1987. A three-dimensional Eulerian acid deposition



       model: Physical concepts and formulation, J. Geophys. Res.,92:  14,681-14,700.



Clarke, J.F., and Edgerton, E.S., 1993. Dry deposition flux calculations for the National



       Dry Deposition Network, EPA-600/R - 93/065, U.S. Environmental Protection



       Agency, Research Triangle Park, NC 27711, 91pp.



Erissman, J.W., Pul, A.van and Wyers, P., 1994. Parameterization of surface resistance



       for the quantification of atmospheric deposition of acidifying pollutants and ozone.



       Atmospheric Environment, 28: 2595-2607.



Finkclstein, P.L., Clarke,  J.F., and Ellestad, T.G., 1995. Measurement of dry deposition



       for deposition velocity model evaluation, Proceedings of the 88th Annual AWMA



       Meeting & Exhibition, San Antonio, Texas, June 18-23, 1995.



Fritschen L. J., Qian  P., Kanemasu E. T., Nie D., Smith E. A., Stewart J. B., Verma S.



       B., Wesely M. L.,  1992. Comparisons of surface flux measurement systems used



       in  FIFE 1989. J. Geophys. Res., 97:  18697-18713.



Kanemasu, E.T., Verma,  S.B., Smith E. A., Fritschen,  L.J., Wesely, M., Field, R.T.,



       Kustas, W.P., Weaver, H., Stewart, J.B., Gurney, R., Panin, G.  and Moncrieff,



       J.B., 1992.  Surface flux measurements in FIFE: an overview. J.  Geophys. Res.,



       97: 18,547-18,555.



Katul, G.G., Finkelstein, P.L., Clarke,  J.F. and Ellestad, T.G., 1996. An investigation of



       the conditional sampling method used to estimate fluxes of active, reactive, and



       passive scalars.  J. Appl. Meteorol., 35: 1835-1845.



Massman, W.J., 1993. Partitioning ozone fluxes to sparse  grass and soil and the inferred



       resistances to dry deposition.  Atmospheric Environment, 27A: 167-174.

-------
Massman, W.J., Pederson, J., Delany, A., Grantz, D., den Hartog, G., Neumann, H.H.,



       Oncley, S.P., Pearson Jr., R., and Shaw, R.H.,  1994.  An evaluation of the



       regional acid deposition model surface module for ozone uptake at three sites in the



       San Joaquin Valley of California. J. Geophys. Res., 99: 8281-8294.



Meyersl, T.P. and Baldocchi, D.D., 1993. Trace gas exchange above the floor of a



       deciduous forest 2. SO2 and O3 Deposition. J. Geophys. Res., 98: 12,631-



       12,638.



Meyers, T.P., Finkelstein, P.L., Clarke, J.F., Ellestad, T.G.,  and Williams, P.P., 1997.



       Description and evaluation of a multi-layer model for inferring dry deposition using



       standard meteorological measurements.  In preparation.



Morris, R.E., Yocke, M.A., Myers, T.C., and Mirabell  V., 1992.  Overview of the



       Variable-Grid Urban Airshed Model (UAM-V).  Presented at the 85th Annual



       Meeting & Exhibition, Kansas City, MO, June 1992.



Nie D., Kanemasu, E.T., Fritschen, L.J., Weaver, H.L., Smith, E.A.,  Verma, S.B.,



       Field, R.T., Kustas, W.P., and Stewart, J.B.,  1992.  An intercomparison of



       surface energy flux measurement systems used during  FIFE 1987.  J. Geophys.



       Res., 97: 18715-18724.



Padro, J., den Hartog, G., and Neumann, H.M., 1991. An investigation of the ADOM dry



       deposition module using summertime O3 measurements above a deciduous forest.



       Atmospheric Environment, 25A: 1689-1704.



Shuttleworth, W.J. and Wallace, J.S., 1985. Evaporation from sparse crops-an energy



       balance combination theory. Q. J. R. Met. Soc., Ill: 839-855.



Smith, E.A., Hsu, A.Y., Crosson, W.L., Field, R.T., Fritschen,  L.J.,  Gurney, R.J.,



       Kanemasu, E.T., Kustas, W.P., Nie, D., Shuttleworth, W.J., Stewart, J.B.,



       Verma, S.B., Weaver, H.L. and Wesely, M.L., 1992.  Area-averaged surface



       fluxes and their time-space variability over the FIFE experiment domain. . J.



       Geophys. Res., 97: 18,599-18,622.
                                                                              17

-------
Tanner, B.D., 1988. Use requirements for Bowen ratio and eddy correlation determination



       of evapotranspiration.  Proceedings of the 1988 Specialty Conference of the



       Irrigation and Drainage Division, American Society of Civil Engineers.



Wescly, M.L. and Hicks, B.B., 1977. Some factors that affect the deposition rates of



       sulfur dioxide and similar gases on vegetation. JAPCA, 27: 1110-1116.

-------
FIGURES



Figure 1. Ozone dry deposition velocity computed from latent heat flux similarity versus



ozone dry deposition velocity derived from eddy correlation (EC) measurements for entire



Keysburg dataset, June 22 - October 11, 1995.



Figure 2. Histogram of observed (EC measurements) minus predicted (similarity



computations) ozone dry deposition velocity for entire dataset.



Figure 3. Same as Figure 1 except for dry daytime conditions only.



Figure 4. Time series of observed (EC measurements) and predicted (similarity



computations) ozone dry deposition velocity for August 1 - 6, 1995.



Figure 5. Same as Figure 4 except that the similarity computations assumed dry conditions



for the daytime portion of August 6.
                                                                                19

-------
             Keysburg, KY,  June 22  - October 11,  1995
o
T3

13
Q.
E
o
O
          -0.5
0
0.5
1.5
                          EC Measured V  (cm/s)
                                       d

-------
      500
O
O
      400  -
      300
      200  -
      100  -
        0
         •1.5
1      -0.5      0       0.5       1

   Observed  -  Predicted (cm/s)
1.5

-------
                            Daytime,  Dry Only
o
Q_
E
o
O
        1.5
        0.5
          0

          -0.5
0
0.5
1.5
                           EC Measured V  (cm/s)

-------
                     EC-Obs
                     Predicted
-0.5
          Aug/1
Aug/2
Aug/3
Aug/4
Aug/5
Aug/6

-------
                      EC-Obs
                —-x— - Predicted
-0.5
          Aug/1
Aug/2
Aug/3
Aug/4
Aug/5
Aug/6

-------
                                                   \ j_J4- WAV »
 1.  REPORT NO.
   EPA/600/A-97/059
 4,  TITLE AND SUBTITLE
                                                                  5.REPORT DATE
 A Technique for  Estimation of Dry Deposition
 Velocities  Based on  Similarity  with Latent  Heat  Flux
                                                                  6.PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)

 Jonathan E,  Pleim,  Peter L.  Finkelstein,  John F,
 Clarke,  and  Thomas G.  Ellestad
             8.PERFORMING ORGANIZATION REPORT NO.
 9. PERFORMING ORGANIZATION NAME AND ADDRESS

 Same as Block 12
                                                                  10.PROGRAM ELEMENT NO.
                                                                  11. CONTRACT/GRANT NO.
 12. SPONSORING AGENCY NAME AND ADDRESS

 U.S. Environmental  Protection Agency
 Office  of Research  and  Development
 National Exposure Research Laboratory
 Research Triangle Park,  NC 27711
             13.TYPE OF REPORT AND PERIOD COVERED

             Journal,  FY-97
             14.  SPONSORING AGENCY CODE

             EPA/600/9
15. SUPPLEMENTARY NOTES
16. ABSTRACT

Field measurements of chemical  dry deposition are needed to assess impacts and trends of  airborne
contaminants on the exposure of crops and unmanaged ecosystems as well  as for the development and
evaluation of air quality models.  However, accurate measurements of dry deposition velocities require
expensive eddy correlation measurements and can only be practically made for a few chemical  species such
as ozone and C02.   On the  other hand, operational dry deposition  measurements such as used in large area
networks involve relatively inexpensive standard meteorological and chemical measurements but rely on
less accurate deposition velocity models.  This paper describes an intermediate technique which can give
accurate estimates of dry deposition velocity for chemical species which are dominated  by stomatal uptake
such as ozone and S02.   This method can  give  results that  are  nearly the quality  of  eddy  correlation
measurements at much lower cost.  The concept is that bulk stomatal conductance can be  accurately
estimated from measurements of  latent heat flux combined with standard  meteorological measurements of
humidity,  temperature, and wind speed.  The technique is tested for a field experiment  where high quality
eddy correlation measurements were made in a soybean field in Kentucky.  Over a four month period, which
covered the entire growth cycle, this technique showed very good agreement with eddy correlation
measurements for ozone.
17.
                                   KEY WORDS AND DOCUMENT ANALYSIS
                    DESCRIPTORS
b.IDENTIFIERS/ OPEN ENDED
TERMS
                                                                                   c.COSATI
18.  DISTRIBUTION STATEMENT
RELEASE  TO PUBLIC
19. SECURITY CLASS (This
Report)

UNCLASSIFIED
                                                                                   21.NO. OF PAGES
                                                    20. SECURITY CLASS  (This
                                                    Page)

                                                    UNCLASSIFIED
                              22. PRICE

-------