& EPA United States Environmental Protection Agency Office of Air Quality Planning and Standards Research Triangle Park. NC 27711 EPA-454/R-94-018 July 1994 Air COMPARISON OF ISC2 DRY DEPOSITION ESTIMATES BASED ON CURRENT AND PROPOSED DEPOSITION ALGORITHMS ------- COMPARISON OF ISC2 DRY DEPOSITION ESTIMATES BASED ON CURRENT AND PROPOSED DEPOSITION ALGORITHMS Office of Air Quality Planning and Standards U.S. Environmental Protection Agency Research Triangle Park, NC 27711 ------- DISCLAIMER This report has been reviewed by the U.S. Environmental Protection Agency (EPA) and has been approved for publication as an EPA document. Any mention of trade names or commercial products does not constitute endorsement or recommendation for use. 11 ------- PREFACE The ability to accurately estimate deposition of particulate matter is of special concern in assessing environmental impacts from a variety of sources including Superfund sites, municipal waste incinerators, and surface coal mines. The limitations of the Industrial Source Complex (ISC2) model (version dated 92273) for use in such assessments are discussed elsewhere and are noted in the following. The deposition algorithm currently employed in ISC2 was developed for applications to large particles dominated by gravitational settling (i.e., particles greater than about 20 fw. diameter) . The current algorithm was not intended for use with small particles and, in fact, includes an assumption that particles less than 5.7 fj.m in diameter are totally reflected at the surface and, thus, experience no deposition. This is in conflict with recent observations. In light of these limitations, a new deposition algorithm which will handle the full range of significant particle sizes (0.1 to 100 /zm) is being considered for use in ISC2. A description of the new algorithm including comparisons with alternative methods for estimating deposition is provided in an April 1994 EPA report "Development and Testing of A Dry Deposition Algorithm (Revised)", EPA-454/R-94-015. The new deposition algorithm has been tested within the framework of the ISC2 model and comparisons of deposition estimates using the old (current) and new deposition algorithms have been made for a range of source types and particulate emission scenarios. Similar comparisons have been made of particulate concentration estimates as affected by the old and new deposition algorithms. This report documents these analyses. The Environmental Protection Agency must conduct a formal public review before the Agency can recommend the new deposition algorithm for use in regulatory analyses. Such a review will enable EPA to assess the potential consequences of replacing the old deposition algorithm in ISC2 with the new algorithm. This report is being released in accordance with this review process. The report is one of several reports on the ISC2 models that must be considered before any formal changes can be adopted. 111 ------- TABLE OF CONTENTS Page 1.0 Introduction 1 2.0 Methodology 2 2.1 Modeling Scenarios 2 2.1.1 Particle Size 2 2.1.2 Meteorological Data 3 2.1.3 Release Height 4 2.1.4 Receptor Location 4 2.2 Diagnostic Comparisons 5 2.3 Operational Comparisons 5 3.0 Results 6 3.1 Diagnostic Comparisons 6 3.2 Operational Comparisons 8 3.2.1 Deposition 8 3.2.2 Concentration 10 4.0 Summary and Conclusions 12 4.1 Deposition 12 4.2 Concentration 12 5.0 References 13 Appendix A A-1 Appendix B B-l IV ------- LIST OF TABLES Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Particle diameter, gravitational settling velocity, and reflection coefficient used in modeling deposition. Source parameters employed in modeling . . . Highest 1-hour ISCST deposition flux estimates (g/m2) based on the old and new algorithms for a 35-m release. Meteorological Conditions used in sensitivity , analyses. Comparison of ISCST deposition flux estimates , based on the old and new algorithms for selected averaging times, release heights, and particle diameters as indicated. Comparison of ISC2 annual deposition flux . . estimates based on the old and new algorithms for release heights and particle diameters as indicated. Comparison of ISCST concentration estimates as affected by the old and new deposition algorithms for selected averaging times, release heights, and particle diameters as indicated. Comparison of ISC2 annual concentration . . . estimates as affected by the old and new deposition algorithms for release heights and particle diameters as indicated. 4 7 14 15 16 17 v ------- LIST OF FIGURES Figure 1. Comparison of highest 1-hour ISCST deposition estimates based on the old and new deposition algorithms for a 35-m release for 50 /zm diameter particles. Figure 2. Comparison of highest 1-hour ISCST deposition estimates based on the old and new deposition algorithms for a 35-m release for particle diameters as indicated. Figure 3. Comparison of ISCST deposition estimates based on the old and new deposition algorithms for selected extreme meteorological conditions and particle diameters as indicated. Figure 4. Fractional difference of highest 1-hour ISCST deposition estimates based on the old and new algorithms for particle diameters and release heights as indicated. Figure 5. Fractional difference of highest 3-hour ISCST deposition estimates based on the old and new algorithms for particle diameters and release heights as indicated. Figure 6. Fractional difference of highest 24-hour ISCST deposition estimates based on the old and new algorithms for particle diameters and release heights as indicated. Figure 7. Fractional difference of highest annual ISCST deposition estimates based on the old and new algorithms for particle diameters and release heights as indicated. Figure 8. Fractional difference of highest 1-hour ISCST concentration estimates considering effects of the old and new deposition algorithms for particle diameters and release heights as indicated. 19 . 20 21 22 . 23 . 24 . 25 VI ------- LIST OF FIGURES (continued) Figure 9. Fractional difference of highest 3-hour ISCST . 26 concentration estimates considering effects of the old and new deposition algorithms for particle diameters and release heights as indicated. Figure 10 Fractional difference of highest 24-hour ISCST . 27 concentration estimates considering effects of the old and new deposition algorithms for particle diameters and release heights as indicated. Figure 11 Fractional difference of highest annual ISCST . 28 concentration estimates considering effects of the old and new deposition algorithms for particle diameters and release heights as indicated. VI1 ------- Comparison of ISC2 Dry Deposition Estimates Based on Current and Proposed Deposition Algorithms 1. INTRODUCTION The Industrial Source Complex dispersion model [ISC2 (USEPA, 1992)] is a widely recommended refined model for use in assessments of air quality impacts from particulate sources and incorporates many features essential to the modeling of such sources. These include area and volume source algorithms, building wake algorithms, and a deposition algorithm based on Dumbauld et al. (1976) and Overcamp (1976). The latter was included in recently completed evaluations (USEPA, 1994) comparing the performance of several dry deposition algorithms and several plume depletion algorithms. As a result of these evaluations, EPA is considering replacing the current deposition algorithm in ISC2 (hereafter referred to as the "old" deposition algorithm) with the deposition algorithm now employed in the Acid Deposition and Oxidant Model [ADOM (Pleim et al., 1984)]. This algorithm will be referred to hereafter as the "new" deposition algorithm. In addition, EPA also proposes to incorporate a plume depletion algorithm in ISC2. The proposed plume depletion algorithm, a modified version of the source depletion technique (Horst, 1977), was selected following the evaluations in USEPA, 1994. This report documents analyses to assess the potential consequences (i.e., differences in estimates of deposition and design concentrations) of replacing the old deposition algorithm in ISC2 with the new algorithm. The old deposition algorithm in ISC2 simulates deposition as the movement of particles toward the surface by the combined processes of atmospheric turbulence and gravitational settling. The method was developed for applications to large particles dominated by gravitational settling; these are typically particles greater than 20 /zm diameter. At the surface, a portion of the plume determined by a user-specified reflection coefficient1 is reflected from the surface; the remainder is deposited, resulting in a reduction in concentration within the plume. It should be noted that for small particles, the ISC2 reflection coefficient was derived based on extrapolation of the gravitational settling-dominated data for larger particles (Dumbauld et al., 1976). The extrapolation led to the assumption that particles less than 5.7 /im in diameter are totally reflected 1 The empirical reflection coefficients recommended for use with ISC2 are based on experiments involving aircraft releases of two spray carriers (Duphar and No. 2 fuel oil) in which most of the particles (more than 99 percent) exceeded 20 /xm diameter. Particles of this large size are nearly completely controlled by gravitational settling. ------- at the surface and thus experience no deposition. This is contrary to recent observations which indicate significant deposition velocities for these small particles. The new deposition algorithm is intended to simulate processes important over the entire range of particle sizes. For example, the proposed algorithm includes effects due to inertial impaction and Brownian motion which control the deposition of small particles. These effects, which are dependent on meteorological and surface conditions, are not easily parameterized within the framework of a reflection coefficient algorithm such as used in the current ISC2. The new deposition algorithm has been tested within the framework of the ISC2 models [ISCST2 (Short-Term) and ISCLT2 (Long-Term)] and comparisons of deposition estimates using the old and new deposition algorithms have been made for a range of source types and particulate emission scenarios. Similar comparisons have been made of particulate concentration estimates as affected by the old and new deposition algorithms. Documentation of these comparisons is provided in the following. 2. METHODOLOGY 2.1 Modeling Scenarios 2.1.1 Particle Size In the development and testing of dry deposition algorithms it was noted that the important physical processes affecting dry deposition could be segregated by particle size (USEPA, 1994). These processes include gravitational settling, which is dominant for large particles (greater than 20 pan diameter), inertial impaction, which is dominant in the size range from 1.0 to 20 /im, and Brownian motion, which is important for small particles (less than about 0.1 jzm) . In the comparisons to be presented, seven particle size categories were employed to represent the important range of particle sizes. These were: 0.1, 1.0, 10, 20, 50, 80, and 100 /urn diameter. The characteristic settling velocities (assuming spherical particles with a density of 1 g/cm3) and reflection coefficients assigned to these particles for use in modeling based on the old deposition algorithm are given in Table 1. For use in modeling, each particle size category was treated as a separate source and assigned a mass fraction of 1.0. ------- Table 1 Particle diameter, gravitational settling velocity, and reflection coefficient used in modeling deposition Particle Diameter (p.m) 0.10 1.00 10.0 20.0 50.0 80.0 100. Settling Velocity (m/s) 0.00000 0.00003 0.003 0.012 0.074 0.190 0.300 Reflection Coefficient 1.00 1.00 0.87 0.76 0.55 0.27 0.00 2.1.2 Meteorological Data 2.1.2.1 ISCST2 The meteorological data employed in the ISCST2 evaluations consisted of one year (1984) of hourly surface data and concurrent twice-daily mixing heights for the National Weather Service (NWS) station at Oklahoma City, Oklahoma. These data were processed using the EPA recommended Meteorological Processor for Regulatory Models [MPRM (USEPA, 1990)]. In addition to the standard set of meteorological variables required by Gaussian dispersion models (i.e., wind direction, wind speed, stability, temperature, and mixing height), the proposed new deposition algorithm also requires estimates of the Monin-Obukhov length and the surface friction velocity. These additional variables were calculated using software adapted from the meteorological preprocessor for the Hybrid Plume Dispersion Model (Hanna and Chang, 1991). To facilitate subsequent analyses, all hourly wind directions were set to a fixed value of 270 degrees (flow vector of 90 degrees). 2.1.2.2 ISCLT2 The same meteorological data (Oklahoma City hourly surface data for 1984) were also processed for use in ISCLT2. This processing was accomplished using the PCSTAR software which generates a joint frequency distribution of wind direction and wind speed by stability class. The Monin-Obukhov length and friction velocity required in the deposition algorithm are generated internally within ISCLT2. The algorithm for computing ------- the friction velocity is based on Wang and Chen (1980) and is the same as was used for ISCST. The algorithm for computing the Monin-Obukhov length is based on Colder (1972). 2.1.3 Release Height Both deposition and concentration are dependent on release height. The modeling performed for these evaluations, included a surface release, and three elevated releases, 35-m, 100-m, and 200-m. The source parameters associated with these four releases are given in Table 2. Table 2 Source parameters employed in modeling Stack Height (m) 0.1 35 100 200 Stack Diameter (m) 1.0 2.4 4.6 5.6 Exit Velocity (m/s) 0.1 11.7 18.8 26.5 Exit Temp. (K) 293 432 416 425 Buoyancy Flux (m4/s3) 53 288 633 Emission Rate (g/s) 1000 1000 1000 1000 2.1.4 Receptor Location All receptors were assumed to be located^ in flat terrain and were assigned an elevation equal to the base elevation of the source. Receptor locations for use in ISCST and ISCLT are described in the following. 2.1.4.1 ISCST2 Receptors for use in ISCST2, were placed at intervals along the X axis beginning at 0.1 km from the source and extending to 50 km. A total of 35 receptors were employed in the modeling. This arrangement of receptors in combination with a fixed 90 degree flow vector facilitates the identification of the highest 1-hour results which, in this analysis, are independent of wind direction. The results obtained for longer averaging times (i.e., 3-hour, 24-hour, and annual) which are dependent on wind direction, will be conservative (higher) compared to results where the wind direction is allowed to vary. ------- 2.1.4.2 ISCLT2 Screening analyses employing coarse polar grids were conducted to determine the distance to the maximum deposition using ISCLT2. Additional screening analyses were conducted using a refined polar grid employing 26 rings extending to 10 km. The screening showed that the highest deposition consistently occurred at receptors located along the 360 degree radial. Consequently, all subsequent analyses using ISCLT2 were conducted with receptors located on the 360 degree radial. These receptors were located at 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.2, 1.4, 1.6, 1.8, 2.0, 2.5, 3.0, 3.5, 4.0, 5.0, 6, 7, 8, 9, and 10 km. 2.2 Diagnostic Comparisons In conducting the ISC2 analyses, it became apparent that a limited sensitivity study was needed to illustrate the sensitivity of the new and old algorithms to particle size, surface characteristics, meteorological conditions, release height, and source/receptor distance. Consequently, sensitivity analyses were constructed using 1-hour deposition and concentration estimates from ISCST2. These sensitivity analyses complement and extend the analyses in USEPA (1994). 2.3 Operational Comparisons In a regulatory air quality analysis, ISC2 is used to estimate expected extreme (highest and second highest) concentrations for averaging times of interest, commonly, 1-hour, 3-hour, 24-hour and annual. Hence, a primary purpose of this analysis is to assess the operational consequences of the new algorithm by examining its effect on extreme value deposition and concentration estimates. This is accomplished by exercising ISCST2 and ISCLT2 in an operational mode and comparing the deposition and concentration estimates based on the old and new deposition algorithms. The performance measure used in the comparisons is patterned on the 'Fractional Bias' (Cox and Tikvart, 1990) and is defined here as the 'Fractional Difference' (FD) : FD = 2 (NEW - OLD) / (NEW + OLD) ------- 3. RESULTS 3.1 Diagnostic Comparisons The results of the diagnostic comparison are presented first since they are essential to the understanding of the results of the operational comparisons. An initial pilot analysis was conducted for the 35-m release height to determine the maximum deposition as a function of particle size for all receptors and all meteorological conditions (i.e., unpaired in space and time), and to identify the meteorological conditions associated with these maxima. Results (old and new deposition estimates as a function of downwind distance) for the 50 /*m diameter particle are presented in Figure 1. The maximum deposition using the old algorithm occurs at 0.5 km under C stability with a wind speed of 15.4 m/s. By comparison, the maximum deposition using the new algorithm occurs at 3 km under F stability with a wind speed of 1.5 m/s. The same conditions (F stability and 1.5 m/s) were associated with a secondary maximum for the old algorithm which also occurs at 3 km downwind. The higher maximum with C stability for the old algorithm is a result of a combination of effects: first, with the higher wind speed (15.4 m/s versus 1.5 m/s) the plume rise from the 35-m stack is lower (19 meters versus 124 meters) and second, the plume disperses more rapidly in slightly unstable conditions (C stability) as compared to strongly stable conditions (F stability). Similar graphs for other particle sizes evaluated (10, 20, 50, 80, and 100 /im) are given in Appendix A. The old ISC2 deposition algorithm employs a 'tilted-plume' approach for characterizing the gravitational descent of the plume as it disperses downwind. The degree of tilt is dependent on the ambient wind speed and the gravitational settling velocity. The latter, as shown in Table 1, is a function of particle size, density, etc. The old algorithm characterizes deposition at the surface through the use of an empirical reflection/absorption coefficient which is also a function of particle size. The effective deposition velocity, defined by the ratio of deposition flux and concentration, is dependent on plume tilt and the interaction of reflection terms with the underlying surface. The end result is that the effective deposition velocity in the old algorithm decreases with downwind distance. The new deposition algorithm also employs a tilted plume approach for characterizing the gravitational descent of the plume. Deposition at the surface is computed as the product of a deposition velocity and the ground-level air concentration. However, in contrast to the old algorithm, the deposition velocity in the new algorithm is a function of surface conditions ------- and meteorology, but is independent of distance from the source. These differences preclude summarization of the results of these comparisons in simple statements. Table 3 and Figure 2 summarize the results of the above analyses for all particle sizes (maximum deposition estimates unpaired in space and time) for 1-hour averaging. For the smaller (0.1 and 1.0 ^m - the old algorithm always results in no deposition for these particles) and larger (50, 80, and 100 /zm) particle sizes, the new algorithm results in higher deposition estimates than the old algorithm. However, for intermediate particle sizes (10 and 20 /zm) the new algorithm results in lower deposition estimates. Table 3 Highest 1-hour average ISCST deposition flux estimates (g/m2) based on the old and new algorithms for a 35-m release. Particle Diameter (Mm) 0.1 1.0 10 20 50 80 100 Maximum Deposition (g/n?) OLD - - 1.04 (*) 1.94 (*) 3.99 {*) 20.8 (**) 69.9 (**) NEW 0.01 0.00 0.62 0.71 4.86 32.7 75.9 (*) (*) (*) (*) (**) (**) (**) Distance to Max (m) OLD - - 500 500 500 1400 900 NEW 600 600 600 600 3000 1200 800 Key to meteorological conditions noted (*) C stability at 15.4 m/s (**) F stability at 1.54 m/s above Additional modeling using the meteorological conditions identified with the modeled extremes (see Table 4) was conducted to assess the sensitivity of the algorithms to meteorology. In the course of these analyses, two additional meteorological cases were included: the first, C stability at 3 m/s, was included to assess the sensitivity of the algorithms to wind speed; the second, E stability at 1.54 m/s, was included to aid in understanding the concentration results for a 200-m release of 100 jum particles. These latter two sensitivity analyses are discussed in Appendix B. ------- Table 4 Meteorological conditions used in sensitivity Analyses P-G Stability Class A C C E F Wind Speed (m/s) 2.57 15.4 3.00 1.54 1.54 Temperature CIO 307 295 295 297 306 Mixing Height (m) 1536 808 808 1368 2756 Friction Velocity (m/s) 0.29 1.35 0.30 0.07 0.07 M-0 Length (m) -8 -839 -60 10 10 Figure 3 shows the results of the analyses of the Table 3 extremes. It should be noted that this figure is similar to Figure 2 except, in this case, the results are paired in time (i.e., the old and new estimates were based on the same meteorological conditions). As shown in Figure 3, the relative magnitude of the deposition estimates based on the new and old algorithms is dependent on the meteorology and, as indicated by the solid lines, the ratio (slope) appears to be a constant for particle sizes above about 20 /zm in diameter for a given meteorological condition. In general, deposition estimates based on the old algorithm exceed the estimates based on the new algorithm for the unstable conditions, while the reverse is the case for stable conditions. The two cases for stability class C (3 m/s and 15 m/s) show the effect of wind speed on the deposition estimates. At the lower wind speed the estimates from the two algorithms are similar. At the higher wind speed, the deposition estimates based on the old algorithm exceed those based on the new algorithm. 3.2 Operational Comparisons 3.2.1 Deposition 3.2.1.1 ISCST2 Operationally, one is often interested in maximum deposition estimates, unpaired in space and time. These would be cumulative estimates; i.e., summed over some period of time (e.g., annual) and over all particle sizes. The latter would be a weighted sum based on the mass fraction of each particle size in the source term. Deposition estimates were computed for four time periods (1- hour, 3-hour, 24-hour, and annual) for each of four release scenarios and each of seven particle size categories. The results are given in Table 5. Note that for each old/new pair of estimates, the corresponding Fractional Difference (FD) comparison statistic is also given. For example, the estimates 8 ------- for a 35-m release of 50 /xm diameter particles over a l-hour period are 3.99 g/m2 using the old algorithm and 4.86 g/m2 using the new algorithm. The FD comparison statistic in this case is computed as: FD = 2(4.86 - 3.99)/(4.86 + 3.99) = 0.20 The resulting value, 0.20 in this example, indicates that the estimate based on the new algorithm increased (in this case by 22 percent) relative to the estimate based on the old algorithm. For use in interpreting the FD statistic, it should be noted that a value of -0.67 (+0.67) indicates that estimates based on the new algorithm decreased (increased) by a factor of 2 relative to estimates based on the old algorithm. The FD values from Table 5 are presented graphically in Figures 4 (1-hour), 5 (3-hour), 6 (24-hour) and 7 (annual average). It should be noted, since the old algorithm assumes that particles less than 5.7 /urn in diameter do not deposit, that the FD statistic for the first two particle size categories (0.1 and 1.0 /zm) will always be 2. For surface releases, the new algorithm results in higher deposition estimates for all except the 10 and 100 /xm particle size categories for the 1-hour period, for all except the 100 jum particle size for the 3-hour period, and for all particle size categories for the 24-hour and annual periods. In general, the magnitude of the increase in the deposition estimate for a surface release increases as the time period of the deposition increases. Annual deposition estimates from a surface release (Figure 7), for example, are increased by more than a factor of 2 using the new algorithm as compared to the old algorithm. For elevated releases, the new algorithm results in higher estimates for the smaller (0.1 and 1.0 ion) and larger (80 and 100 /zm) particle sizes. The increase in the deposition estimates for 80 and 100 /zm particles is generally less than a factor of two. For intermediate particle sizes (10 and 20 /zm) , deposition estimates using the new algorithm are decreased relative to the corresponding estimates based on the old algorithm. The decrease generally exceeds a factor of 2 for 100-m and 200-m releases. Results vary for elevated releases of 50 /zm particles; for example, deposition estimates for a 35-m release increase (generally by less than a factor of 2), whereas estimates for 100-m and 200-m releases may increase (annual) or decrease (1- hour and 3-hour) depending on the averaging period. ------- 3.2.1.2 ISCLT2 Annual deposition estimates were also computed using ISCLT2. These results are presented in Table 6. For a surface release, annual deposition estimates based on the new algorithm exceed estimates based on the old algorithm for all particle sizes. The FD statistic for the surface release ranges from 1.5 to 2.0. For elevated releases, deposition estimates based on the new algorithm may be higher or lower than estimates based on the old algorithm, depending on particle size and release height. 3.2.2 Concentration Ambient concentrations are reduced by the deposition of particulate matter and, as such, model estimates of ambient concentrations depend on the algorithms employed to model deposition processes. In the case of the old algorithm, the fractional reduction in concentration is equal to one half of the fraction deposited (i.e., one minus the reflection coefficient). Thus, concentrations of 100 /zm particles, which experience no reflection, are reduced by factor of 2. It should be noted that the treatment of the gravational settling in the old deposition algorithm results in a 'bouncing plume'. As such, concentration estimates using the old deposition algorithm are invalid beyond the point of plume touch down. The latter is a function of particle size (settling velocity), release height, and wind speed. Because of uncertainties resulting from the bouncing plume, most past applications of ISC2 for estimating concentrations of particulate matter have normally been made with the deposition algorithm turned off. In the case of the new algorithm, the reduction in concentration is estimated using a modified source depletion technique based on Horst (1977). With this technique, the source term (emission rate) is adjusted (decreased) to simulate the increased pollutant removal with distance. In addition to these changes, the new algorithm incorporates necessary changes to eliminate the bouncing plume. 3.2.2.1 ISCST2 As was done for deposition, corresponding concentration results for l-hour, 3-hour, 24-hour, and annual averaging are presented in Table 7. The FD values from Table 7 are presented graphically in Figures 8 (1-hour), 9 (3-hour), 10 (24-hour) and 11 (annual average). 10 ------- For particles up to 20 /zm in diameter, the differences in concentration estimates are not significant (less than 10 percent). For surface releases of larger particles, the concentration estimates using the new algorithm are consistently lower than the corresponding estimates using the old algorithm. The decrease in this case ranges between a factor of 1.5 and 2 (50 to 100 percent decrease). For elevated releases of larger particles, the results are mixed. For 50 /zm particles, the concentration estimates using the new algorithm are lower than the corresponding estimates using the old algorithm. However, for elevated releases of the largest particle size category (100 /zm) , the concentration estimates using the new algorithm are increased relative to the corresponding estimates using the old algorithm. The increase approaches a factor of 2 for 100-m and 200-m release heights (for short-term averaging periods). For annual averaging, the increase approaches a factor of 1.5. This apparent contradiction with the deposition results (deposition estimates based on the new algorithm also exceed the estimates based on the old algorithm for elevated releases of large particles) is a direct result of a combination of inherent differences in the two algorithms. These differences, which are manifest in the modeling of elevated releases of large particles, are discussed in Appendix B. The above results are for ISCST2 run with the deposition (old and new) and plume depletion (new only) options activated (turned-on). The concentration estimates from the 'old' and 'new1 versions of ISCST2 are identical when deposition is not turned-on. 3.2.2.2 ISCLT2 Annual concentration estimates were also computed using ISCLT2. These results are presented in Table 8. For a surface release, annual concentration estimates using the new algorithm are less than the corresponding estimates using the old algorithm for all particle sizes. The FD statistic for the surface release ranges from zero (no change) to -0.2 (a decrease of 20 percent). For elevated releases, with the exception of 100 izm size particles, the differences are generally not significant. For 100 urn particles, concentration estimates using the new algorithm are higher by 30 to 40 percent. 11 ------- 4. Summary and Conclusions The proposed algorithm was expected to provide deposition estimations similar to the old algorithm for particles 20 jzm in diameter and larger, and to provide more realistic (higher) deposition estimates for particles less than 20 /*m in diameter. 4.1 Deposition For surface releases, this comparison shows that the new algorithm results in higher deposition estimates for all particle size categories for the 24-hour and annual periods. The annual deposition estimates are increased by more than a factor of 2 using the new method. For elevated releases, the new algorithm results in higher estimates for the smaller (0.1 and 1.0 /zm) and larger (80 and 100 pirn) particle sizes. The increase in the deposition estimates for 80 and 100 /zm particles is generally less than a factor of two. For intermediate particle sizes (10 and 20 /on) , deposition estimates using the new algorithm are decreased relative to the corresponding estimates based on the old algorithm. The decrease generally exceeds a factor of 2 for 100-m and 200-m releases. The ratio of the deposition estimates based on the new and old algorithms is dependent on the meteorology and appears to be a constant for particle sizes above about 20 /zm for a given meteorological condition. In general, deposition estimates based on the old algorithm exceed the estimates based on the new algorithm for the unstable conditions, while the reverse is the case for stable conditions (e.g., see Figure 3). 4.2 Concentration For a surface release of large particles (50, 80, and 100/zm) , estimated concentrations using the new algorithm are decreased relative to the corresponding estimates using the new algorithm. The decrease ranges between a factor of 1.5 and 2 (50 to 100 percent decrease). For small and intermediate particles (0.1, 1, 10, and 20 /zm) , the differences are not significant (less then 10 percent). For elevated releases of larger particles the results are mixed. However, in all cases, the differences for these larger particles do not exceed a factor of 2. As was the case for surface releases, the differences for small and intermediate size particles are not significant (less than 10 percent). 12 ------- 5. REFERENCES Cox, W.M. and J.A. Tikvart, 1990: A Statistical Procedure for Determining the Best Performing Air Quality Simulation Model. Atmos. Environ., 24, 2387-2395. Dumbauld, R.K., J.E. Rafferty and H.E. Cramer, 1976: Dispersion-deposition from aerial spray releases. Proceedings of the Third Symposium on Atmospheric Turbulence, Diffusion, and Air Quality, October 19-22, Raleigh, NC. Colder, D., 1972: Relations among stability parameters in the surface layer. Boundary Layer Met. 3, 46-58. Hanna, S.R. and J.C. Chang, 1991: Modification of the Hybrid Plume Dispersion Model (HPDM) for urban conditions and its evaluation using the Indianapolis data set. Vol. I. User's guide for HPDM-Urban. Sigma Research Corp., Concord, MA. Overcamp, T.J., 1976: A general Gaussian diffusion-deposition model for elevated point sources. J. Appl. Meteor., 15, 1167-1171. Horst, T.W., 1977: A surface depletion model for deposition from a Gaussian plume. Atmos. Environ., 11, 41-46. Pleim, J., A. Venkatram and R. Yamartino, 1984: ADOM/TADAP model development program. Volume 4. The dry deposition module. Ontario Ministry of the Environment, Rexdale, Ontario, Canada U.S. EPA, 1990: Meteorological Processor for Regulatory Models (MPRM 1.2) User's Guide. EPA-600/3- 88 - 043 (Revised). U.S. Environmental Protection Agency, Research Triangle Park, North Carolina, 27711. U.S. EPA, 1992: User's Guide for the Industrial Source (ISC2) Dispersion Model. Volume 1 - User Instructions. EPA-450/4- 92-008a. U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711. 13 ------- U.S. EPA, 1994: Development and Testing of Dry Deposition Algorithms. EPA-454/R-94-015. U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711. Wang, I.T. and P.C. Chen, 1980: Estimations of heat and momentum fluxes near the ground. Proc. 2nd Joint Conf. on Applications of Air Poll. Meteorol., AMS, 764-769. 13a ------- TABLE 5 Comparison of ISCST deposition flux estimates based on the old and new algorithms for selected averaging times, release heights, and particle diameters as indicated. Fractional Difference: FD = 2 (NEW - OLD) / (NEW + OLD) 1-HOUR EXTREMES (g/m'2) Diameter (microns) 10 20 50 80 100 OLD 352 418 1096 1906 1296 Surface NEW 230 727 2299 1987 1156 FO Sfc. -0.4 0.5 0.7 0.0 -0.1 35-meter OLD NEW 1.04 0.62 1.94 0.71 3.99 4.86 20.83 32.67 69.86 75.88 FD 35-m -0.5 -0.9 0.2 0.4 0.1 OLD 0.43 0.79 1.48 2.39 7.75 100-meter NEW 0.06 0.07 0.52 3.63 14.05 FD 100-m -1.5 -1.7 -1.0 0.4 0.6 OLD 0.27 0.49 0.93 1.51 2.48 200-meter NEW 0.01 0.02 0.15 1.43 4.41 FD 200-m •1.9 -1.8 •1.4 •0.1 0.6 3-HOUR EXTREMES (8/nT2) Diameter (microns) 10 20 50 80 100 OLD 357 647 3125 5338 3388 Surface NEW 549 1799 6464 5533 2913 FD Sfc. 0.4 0.9 0.7 0.0 -0.2 35-meter OLD NEW 2.19 1.40 4.10 1.64 8.55 14.20 59.99 92.35 198.90 211.25 FD 35-m -0.4 -0.9 0.5 0.4 0.1 OLD 0.59 1.10 2.21 6.51 21.99 100-meter NEW 0.14 0.16 1.46 10.49 38.80 FD 100-m -1.2 -1.5 -0.4 0.5 0.6 OLD 0.34 0.63 1.23 2.17 5.52 200-meter NEW 0.03 0.04 0.34 4.18 10.94 FD 200-m -1.7 •1.8 -1.1 0.6 0.7 24-HOUR EXTREMES (g/m'2) Diameter (microns) 10 20 50 80 100 OLD 990 2267 11659 14414 12256 Surface NEW 2626 5957 27498 19622 14292 FD Sfc. 0.9 0.9 0.8 0.3 0.2 35-meter OLD NEW 6.24 6.07 12.00 7.99 29.50 41.95 172.50 254.91 563.50 583.65 FD 35-m 0.0 -0.4 0.3 0.4 0.0 OLD 0.95 1.77 3.44 17.15 56.73 100-meter NEW 0.42 0.61 4.07 25.74 78.85 FD 100-m -0.8 -1.0 0.2 0.4 0.3 OLD 0.43 0.76 1.45 3.80 13.44 200-meter NEW 0.08 0.13 0.89 8.18 21.68 FD 200-m -1.4 -1.4 -0.5 0.7 0.5 ANNUAL EXTREMES (g/m'2) Diameter (microns) 10 20 50 80 100 OLD 111539 243405 955851 1258514 905441 Surface NEW 503090 1318515 4041833 3773714 2679361 FD Sfc. 1.3 14 1.2 1.0 1.0 35 mater OLD NEW 855.00 530.00 1677.00 1107.00 4838.00 6697.00 26225.00 36422.00 81102.00 83232.00 FD 35-m -0.5 -0.4 0.3 0.3 0.0 OLD 62.00 120.00 343.00 2625.00 8589.00 100-meter NEW 40.40 92.50 695.00 3928.00 9344.00 FD 100-m -0.4 -0.3 0.7 0.4 0.1 OLD 19.00 35.00 80.00 583.00 1990.00 200-metor NEW 8.60 20.50 159.00 989.00 2548.00 FD 200-m •0.8 -0.5 0.7 0.5 0.2 14 ------- TABLE 6 Comparison of ISCLT annual deposition flux estimates based on the old and new algorithms for release heights and particle diameters as indicated. Fractional Difference: FD = 2 (NEW - OLD) / (NEW + OLD) Surface Release Diameter (microns) 0.1 1 10 20 50 80 100 OLD 0 0 1199 3202 12376 21440 28738 NEW 2831 1093 67358 117317 239284 235098 194290 FD 2.0 2.0 1.9 1.9 1.8 1.7 1.5 35-Meter Release OLD 0 0 96 187 505 1794 5483 NEW 2 1 67 103 412 2182 5136 FD 2.0 2.0 -0.4 -0.6 -0.2 0.2 -0.1 Diameter (microns) 100-Meter Release OLD NEW FD 200-Meter Release OLD NEW FD 0.1 1 10 20 50 80 100 0 0 9 17 41 149 488 0 0 5 8 37 210 511 2.0 2.0 -0.6 -0.7 -0.1 0.3 0.0 0 0 3 6 13 34 114 0.0 0.0 1.0 1.8 7.5 51.6 135.0 2.0 2.0 -1.0 -1.1 -0.5 0.4 0.2 15 ------- TABLE 7 Comparison of ISCST concentration estimates based on the old and new algorithms for selected averaging times, release heights, and particle diameters as indicated. Fractional Difference: FD = 2 (NEW - OLD) f (NEW + OLD) 1-HOUR EXTREMES (micro-g/m'3) Diameter (microns) OLD 0.1 21711142 1 21712652 10 20370922 20 18579514 50 14046924 80 5646102 100 1504833 Surface NEW 21608136 21625196 20260306 16512426 8455137 2841743 1032195 FD Sfc. 0.00 0.00 -0.01 -0.12 •0.50 •0.66 -0.37 OLD 2836 2836 2658 3403 20355 48452 59623 35-meter NEW 2824 2832 2839 3414 17875 47026 69928 FD 35-m 0.00 0.00 0.07 0.00 -0.13 -0.03 0.16 OLD 897 897 836 787 2208 5386 6895 100-meter NEW 894 895 895 865 1931 5235 12947 FD 100-m 0.00 0.00 0.07 0.09 •0.13 -0.03 0.61 OLD 439 439 410 386 648 1626 2230 200-meter NEW 437 438 438 424 563 2063 4072 FD 200-m -0.01 0.00 0.07 0.09 -0.14 0.24 0.58 3-HOUR EXTREMES (micro-g/m'3) Diameter (microns) OLD 0.1 17028218 1 17029042 10 15960174 20 14728003 50 13971580 80 5475584 100 1407853 24-HOUR EXTREMES I Diameter (microns) OLD 0 1 6225974 1 6226221 10 5832943 20 5392277 50 6232613 80 2300707 100 836814 Surface NEW 16957118 16972472 16089264 13625220 7921412 2639824 875861 |micro-g/m"3) Surface NEW 6197444 6205344 5893759 4966362 4292841 1241776 530638 FD Sfc. 0.00 0.00 0.01 •0.08 -0.55 -0.70 •0.47 FD Sfc. 0.00 0.00 0.01 -0.08 -0.37 -0.60 -0.45 OLD 2490 2491 2353 3278 19687 45792 57704 OLD 1777 1778 1687 1661 7885 18098 22210 35-meter NEW 2484 2488 2491 3270 17414 44300 64895 35-meter NEW 1770 1775 1521 1621 7277 17401 25557 FD 35-m 0.00 0.00 0.06 0.00 -0.12 -0.03 0.12 FD 35-m 0.00 0.00 •0.10 -0.02 •0.08 -0.04 0.14 OLD 486 486 454 429 2119 5122 6734 OLD 151 151 142 148 832 1832 2319 100-meter NEW 484 485 484 472 1797 5034 11918 100-meter NEW 149 150 150 141 705 1756 3453 FD 100-m 0.00 0.00 0.06 0.10 -0.16 -0.02 0.56 FD 100-m -0.01 0.00 0.06 -0.05 -0.17 -0.04 0.39 OLD 186 186 174 165 525 1242 1675 OLD 56 56 51 48 182 388 509 200-meter NEW 185 185 186 179 416 2008 3362 200-meter NEW 55 55 55 51 142 559 950 FD 200-m 0.00 0.00 0.07 0.08 -0.23 0.47 0.67 FD 200m -0.01 •0.01 0.08 0.06 -0.24 0.36 0.60 ANNUAL EXTREMES |micro-g/nT3) Diameter (microns) OLD 0 1 2259888 1 2260180 10 2138979 20 2081605 50 1971727 80 886039 100 342726 Surface NEW 2252064 2256171 2174798 2028710 1534105 594997 274974 FD 0.00 0.00 0.02 -0.03 -0.25 -0.39 0.22 OLD 1020 1020 985 1025 2421 5207 6485 35-meter NEW 1015 1019 956 1017 2497 5829 8699 FD 0.00 0.00 -0.03 -0.01 0.03 0.11 0.29 OLD 85 85 82 89 266 587 756 100-meter NEW 84 84 78 88 256 627 975 FD •0.01 0.00 -0.05 -0.01 -0.04 0.07 0.25 OLD 18 18 17 19 63 143 185 200-meter NEW 18 18 17 19 58 157 266 FD -0.01 0.00 -0.05 -0.02 -0.08 0.10 0.36 16 ------- TABLE 8 Comparison of ISCLT annual concentration estimates as affected by the old and new deposition algorithms for release heights and particle diameters as indicated. Fractional Difference: FD = 2 (NEW - OLD) f (NEW + OLD) Diameter (microns) Surface Release OLD NEW FD 35-Meter Release OLD NEW FD 0.1 1 10 20 50 80 100 194905 194905 182408 170238 103686 45320 23616 193914 194452 177153 156715 83371 35293 19240 0.0 0.0 0.0 -0.1 -0.2 -0.2 •0.2 88 88 84 84 128 286 376 88 88 78 83 137 336 521 0.0 0.0 -0.1 0.0 0.1 0.2 0.3 100-Meter Release 200-Meter Release Diameter (microns) 0.1 1 10 20 50 80 100 OLD 7.0 7.0 6.6 6.6 12.0 28.9 38.3 NEW 6.9 6.9 6.1 6.5 12.4 32.2 51.7 FD 0.0 0.0 -0.1 0.0 0.0 0.1 0.3 OLD 1.9 1.9 1.7 1.7 2.7 7.1 9.4 NEW 1.8 1.9 1.6 1.7 2.4 7.8 13.6 FD -0.1 0.0 -0.1 0.0 -0.1 0.1 0.4 17 ------- 3 FIGURE 1 Comparison of highest 1-hour ISCST deposition estimates based on old and new algorithms for a 35-meter release for 50 micron diameter particles. D • OLD D D NEW D D a D • ri a a D D n 0.1 1 10 Downwind Distance (km) 18 ------- 80 60 0> .2 40 FIGURE 2 Comparison of highest 1-hour ISCST deposition estimates (g|m*2) based on the old and new algorithms for a 35-meter release for particle diameters (microns) as indicated: Perfect Agreement NEW - OLD 10 20 50 20 20 40 Deposition (old) GO 80 19 ------- 15 10 FIGURE 3 Comparison of ISCSTdeposition estimates (g|m"2) based on the old and new algorithms for selected extreme meteorological conditions and particle diameters (microns) as indicated, for a 35-meter release Stability (wind speed) Perfect Agreement NEW - OLD 50/ 10 15 Deposition (old) 20 ------- FIGURE 4 Fractional difference (FD) of highest 1-hour ISCST deposition estimates based on old and new algorithms for particle sizes and release heights as indicated FD = 2 (NEW - OLD) I (NEW + OLD) 2.0 1.0 f 0.0 1.0 2.0 0.1 10 20 50 80 Particle Diameter (microns) Sfc. H 35-m D 100-m ^ 200-m 100 21 ------- FIGURE 5 Fractional difference (FD) of highest 3-hour ISCST deposition estimates based on old and new algorithms for particle sizes and release heights as indicated FD = 2 (NEW - OLD) f (NEW + OLD) 2.0 1.0 3: - 0.0 -1.0 2.0 0.1 10 20 50 Particle Diameter (microns) Sfc. 35-m D 100-m M 200-m 80 100 22 ------- FIGURE 6 Fractional difference (FD) of highest 24-hour ISCST deposition estimates based on old and new algorithms for particle sizes and release heights as indicated FD = 2 (NEW - OLD) f (NEW + OLD) 2.0 1.0 0.0 10 1.0 2.0 0.1 10 20 50 Particle Diameter (microns) Sfc. 35-m LJ 100-m m 200-m 80 100 23 ------- FIGURE 7 Fractional difference (FD) of highest annual ISCST deposition estimates based on old and new algorithms for particle sizes and release heights as indicated FD = 2 (NEW-OLD) | (NEW + OLD) 2.0 1.0 0.0 -1.0 2.0 0.1 10 20 50 Particle Diameter (microns) Sfc. 35-m D 100-1 80 100 m 200-m 24 ------- FIGURE 8 Fractional Difference (FD) of highest 1-hour ISCST concentration estimates considering effects of old and new deposition algorithms for particle diameters and release heights as indicated FD = 2 (NEW - OLD) /{NEW + OLD) 1.00 0.50 0.00 •0.50 1.00 0.1 10 20 50 Particle Diameter (microns) Sfc. H 35-m D 100-m Wh 200-m 80 100 25 ------- FIGURE 9 Fractional Difference (FD) of highest 3-hour ISCST concentration estimates considering effects of old and new deposition algorithms for particle diameters and release heights as indicated FD = 2 (NEW • OLD) /(NEW + OLD) 1.00 0.50 01 o c 09 0.00 •0.50 1.00 0.1 10 20 50 Particle Diameter (microns) Sfc. H 35-m D 100-1 m 200-m 80 100 26 ------- FIGURE 10 Fractional Difference (FD) of highest 24-hour ISCST concentration estimates considering effects of old and new deposition algorithms for particle diameters and release heights as indicated FD = 2 (NEW • OLD) /(NEW + OLD) 1.00 0.50 0.00 •0.50 1.00 0.1 10 20 50 Particle Diameter (microns) 80 100 Sfc. 35-m 100-m H 200-m 27 ------- FIGURE 11 Fractional Difference (FD) of highest annual-hour ISCST concentration estimates considering effects of old and new deposition algorithms for particle diameters and release heights as indicated FD = 2 (NEW - OLD) f( NEW + OLD) 1.00 0.50 0.00 -0.50 1.00 0.1 10 20 50 Particle Diameter (microns) 80 100 Sfc. 35-m D 100-m H 200-m 28 ------- APPENDIX A Comparison of Highest 1-hour ISCST Deposition Estimates Based on the Old and New Algorithms for a 35-m Release ------- APPENDIX A LIST OF FIGURES Figure Page A-l Comparison of highest 1-hour ISCST deposition ... A-3 estimates based on old and new algorithms for a 35-meter release for 10 micron diameter particles. A-2 Comparison of highest 1-hour ISCST deposition ... A-4 estimates based on old and new algorithms for a 35-meter release for 20 micron diameter particles. A-3 Comparison of highest 1-hour ISCST deposition ... A-5 estimates based on old and new algorithms for a 35-meter release for 50 micron diameter particles. A-4 Comparison of highest 1-hour ISCST deposition ... A-6 estimates based on old and new algorithms for a 35-meter release for 80 micron diameter particles. A-5 Comparison of highest 1-hour ISCST deposition .... A-7 estimates based on old and new algorithms for a 35-meter release for 100 micron diameter particles. A-2 ------- FIGURE A-1 Comparison of highest 1-hour ISCST deposition estimates based on old and new algorithms for a 35-meter release for 10 micron diameter particles. 5 r • OLD c NEW 6 3 -2. E c u o ey 0.1 1 10 Downwind Distance (km) A-3 ------- FIGURE A-2 Comparison of highest 1-hour ISCST deposition estimates based on old and new algorithms for a 35-meter release for 20 micron diameter particles. 5 r • OLD D NEW -S2. E = 2 D " D D . , D DD Q ' "SB D D D D 1 10 Downwind Distance (km) A4 ------- FIGURE A-3 Comparison of highest 1-hour ISCST deposition estimates based on old and new algorithms for a 35-meter release for 50 micron diameter particles. I 3 01 D D • OLD 0 NEW on n _l I L_ 0.1 1 Downwind Distance (km) 10 A-5 ------- FIGURE A -4 Comparison of highest 1-hour ISCST deposition estimates based on old and new algorithms for a 35-meter release for 80 micron diameter particles. 100 80 £ 60 I- 09 1 I E a 40 20 n • OLD D NEW D 0.1 1 Downwind Distance (km) n nuo-D-g 10 A-6 ------- 40 20 FIGURE A-5 Comparison of highest 1-hour ISCST deposition estimates based on old and new algorithms for a 35-meter release for 100 micron diameter particles. 100 80 1 GO OJ E • OLD D NEW 0.1 Pan n -D-D-O-d 1 Downwind Distance (km) 10 A-7 ------- APPENDIX B Comparison of Highest 1-Hour Concentration Estimates as Affected by the Old and New Deposition Algorithms for Elevated Releases ------- APPENDIX B LIST OF FIGURES Figure Page B-l Comparison of ISCST concentration estimates as ... B-5 affected by the old deposition algorithm for C stability with a 3 m/s wind speed for a 35-m release for particle diameters as indicated. B-2 Comparison of ISCST concentration estimates as ... B-6 affected by the old deposition algorithm for C stability with a 15 m/s wind speed for a 35-m release for particle diameters as indicated. B-3 Comparison of ISCST concentration estimates as ... B-7 affected by the new deposition algorithm for C stability with a 3 m/s wind speed for a 35-m release for particle diameters as indicated. B-4 Comparison of ISCST concentration estimates as ... B-8 affected by the new deposition algorithm for C stability with a 15 m/s wind speed for a 35-m release for particle diameters as indicated. B-5 Comparison of ISCST concentration estimates as ... B-9 affected by the old and new deposition algorithms for E stability with a 1.5 m/s wind speed for a 200-m release for 100 /im diameter particles. B-2 ------- The results presented in Appendix B reflect diagnostic analyses of selected extreme 1-hour concentration events for elevated releases. Figures B-l through B-4 show results for a 35-m release for C stability with a 3 m/s wind speed [Figure B-l (old) and Figure B-3 (new)] and with a 15 m/s wind speed [Figure B-2 (old) and Figure B-4 (new). Effect of Wind Speed on Concentration For the 3 m/s wind speed, the maximum concentration increases, and the distance to the maximum decreases with increasing particle size. This is the case for both the old (Figure B-l) and new (Figure B-3) algorithms. These results reflect the influence of gravitational settling and its affect on plume tilt. Gravitational settling of a 100 /xm particle (0.3 m/s) is significant relative to a 3 m/s wind speed. Under these conditions, an elevated plume of 100 /xm particles descends more steeply and intersects the surface closer to the source than a plume composed of smaller particles. By comparison, gravitational settling and plume tilt are not as significant with a 15 m/s wind speed. In this case, concentrations with the old deposition algorithm (Figure B-2) actually decrease with increasing particle size (note that deposition increases with particle size). With the new algorithm (Figure B-4), concentrations appear to increase with particle size (at least to a downwind distance of 1 km.) - this reversal may possibly be due to the affect on deposition of a higher friction velocity. Effects of Deposition on Concentration Concentration estimates were also computed with the deposition option turned off. These estimates are shown by the solid curves in Figures B-l and B-3 for the 3 m/s wind speed, and by the solid curves in Figures B-2 and B-4 for the 15 m/s wind speed. For both old (Figures B-l and B-2) and new (Figures B-3 and B-4) algorithms, the no deposition curve coincides with the concentration estimates for the 1 /xm particle size category. This is as expected for the old algorithm since it considers effects due to gravitational settling only and assigns a reflection coefficient of 1.0 (no deposition) to this size category. For the new algorithm, the 1 /xm particle size does not have a noticeable effect on concentration estimates. For larger particles, the maximum concentration increases with particle size. B-3 ------- Results for a 200-m Release of 100 /on Particles Figure B-5 shows the estimated concentrations for a 200-m release of 100 /im particles for the extreme 1-hour meteorological conditions associated with this case (E stability at 1.5 m/s). Concentration estimates as affected by the new deposition algorithm are shown for two cases (with and without plume depletion). It is seen that the effects of plume depletion become significant at downwind distances beyond the location of the maximum (about 5 km). The results show that the concentration estimates as affected by the new deposition algorithm are about a factor of 2 greater than the corresponding estimates for the old algorithm. B-4 ------- FIGURE B • 1 Comparison of ISCST concentration estimates as affected by the old deposition algorithm for C stability with a 3 m/s wind speed for a 35-m release for particle diameters as indicated 4000 3000 - 1 micron 50 microns 100 microns no deposition 2000 1000 0.1 1.0 Downwind Distance (km) 10.0 B-5 ------- FIGURE B-2 Comparison of ISCST concentration estimates as affected by the old deposition algorithm for C stability with a 15 mfs wind speed for a 35-m release for particle diameters as indicated 4000 3000 1 micron 50 microns 100 microns no deposition C"J « E 2000 1000 u 0.1 1.0 Downwind Distance (km) 10.0 B-6 ------- FIGURE B -3 Comparison of ISCST concentration estimates as affected by the new deposition algorithm for C stability with a 3 m/s wind speed for a 35-m release for particle diameters as indicated 4000 3000 - * * 1 micron 50 microns 100 microns no deposition 2000 o o 1000 - 0.1 1.0 Downwind Distance (km) 10.0 B-7 ------- FIGURE B-4 Comparison of ISCST concentration estimates as affected by the new deposition algorithm for C stability with a 15 m/s wind speed for a 35-m release for particle diameters as indicated 4000 r 3000 - D 1 micron 50 microns 100 microns no deposition u 'i 7 2000 o I e U 1000 0.1 1.0 Downwind Distance (km) 10.0 B-8 ------- FIGURE B-5 Comparison of ISCST concentration estimates for a 200-m release of 100 micron particles as affected by the old and new deposition algorithms. Meteorological Conditions: Stability E at 1.5 m/s 5000 4000 - 3000 !- u 'i o o 2000 1000 10 Downwind Distance (km) 100 B-9 ------- TECHNICAL REPORT DATA (Please read Instructions on reverse before completing) 1 REPORT NO EPA-454/R-94-018 3 RECIPIENT'S ACCESSION NO 5 REPORT DATE 4 TITLE AND SUBTITLE Comparison of ISC2 Dry Deposition Estimates Based on Current and Proposed Deposition Algorithms July 1994 6 PERFORMING ORGANIZATION CODE 7 AUTHOR(S) 8 PERFORMING ORGANIZATION REPORT NO Desmond T. Bailey 9 PERFORMING ORGANIZATION NAME AND ADDRESS 10 PROGRAM ELEMENT NO U.S. Environmental Protection Agency Office of Air Quality Planning and Standards Research Triangle Park, NC 27711 11 CONTRACT/GRANT NO 12 SPONSORING AGENCY NAME AND ADDRESS 13 TYPE OF REPORT AND PERIOD COVERED Director Office of Air Quality Planning and Standards Office of Air and Radiation U.S. Environmental Protection Agency Research Triangle Park, NC 27711 14 SPONSORING AGENCY CODE EPA/200/04 15 SUPPLEMENTARY NOTEi 16 ABSTRACT The ability to accurately estimate deposition of particulate matter is of special concern in assessing environmental impacts from a variety of sources including Superfund sites, municipal waste incinerators, and surface coal mines. The current deposition algorithm in ISC2 simulates deposition as the movement of particles toward the surface by the combined processes of atmospheric turbulence and gravitational settling. The method was developed for applications to large particles dominated by gravitational settling; these are typically particles greater than 20 /*m diameter. The current algorithm was not intended for use with particles smaller than about 20 /tm which are often of concern in air toxics assessments. In light of this limitation, a new deposition algorithm is being considered for use in ISC2. The proposed algorithm is intended to simulate processes important over the entire range of significant particle sizes (0.1 to 100 /ttm). The proposed algorithm employs a deposition velocity based on a resistance model. In this approach, deposition flux is calculated as the product of the near- surface air concentration and the deposition velocity. The latter is computed as the inverse sum of the aerodynamic layer and deposition layer resistances, plus gravitational settling. The new deposition algorithm has been tested within the framework of the 1SC2 model and comparisons of deposition estimates using the old (current) and new deposition algorithms have been made for a range of source types and particulate emission scenarios. Similar comparisons have been made of particulate concentration estimates as affected by the old and new deposition algorithms. KPY WORDS AND DOCIMBP/T ANALYSIS IDBf/nplBRS/OPBN BNDBD TERMS Air Pollution control » (BCURITY O-A3S Unclassified 44 Release Unlimited > SBCURTTY O-A33 IFafl Unclassified KPA Font 12JO-1 (Mr* 4-T! M'S EDITION IS OBSOLETE 11 '• .1 -Jlfl/1 2iij floor ------- |