United States Environmental Protection Agency Industrial Environmental Research Laboratory Research Triangle Park NC 2771' Research and Development EPA-600/S8-84-017 Aug. 1984 &EPA Project Summary Paniculate Data Reduction (PADRE) System Reference Manual W.M. Yeager and C.E. Tatsch The Particulate Data Reduction (PADRE) system is an interactive computer program that facilitates entry, reduction, and analysis of cas- cade impactor data for particle size distributions. PADRE was developed to ensure the quality of data included in the Fine Particle Emissions Information Systems (FPEIS), a component of the Environmental Assessment Data Sys- tems (EADS). Users control the logical flow through the system in response to prompts from the program. PADRE can be used to store, review, edit, and analyze data and, through a variety of data checks, to identify invalid or suspect data. Impactor stage cut points are calculated, and cumulative and differential mass concentrations are determined and interpolated to standard diameters. This document describes how to access and use PADRE and summarizes its logic and capabilities. The program narrative describes all calculations and algorithms and all messages produced by the system. Specific user instructions are detailed in a separate PADRE User's Guide. This Project Summary was developed by EPA's Industrial Environmental Research Laboratory. Research Triangle Park. NC, to announce key findings of the research project that is fully docu- mented in a separate report of the same title (see Project Report ordering information at back). Introduction The purpose of the PADRE system is to facilitate entry of validated cascade impactor data from representative in- stackruns into the Fine Particle Emissions Information System (FPEIS). Reduction of observed data, to determine the effective stage cut diameters as well as the cumulative and differential mass concen- trations, takes only a few seconds; this capability permits test plans or operating conditions to be modified to optimize data quality. Data entered through PADRE are not automatically included in FPEIS; the test contractor is expected to designate representative runs after he has validated the data. Data Reduction is based on the Cascade Impactor Data Reduction System (CIDRS) computer programs developed by South- ern Research Institute (SoRI) under contract to EPA. PADRE allows data reduction from a wider variety of impactor types and configurations than does CIDRS. Additionally, PADRE is interactive, which facilitates data entry and allows rapid analysis of impactor data by novice computer users. FPEIS was developed in the mid- 1970's and became fully operational in January 1977. EPA's Industrial Environ- mental Research Laboratory at Research Triangle Park (IERL-RTP) then develped a series of data bases (modeled on the FPEIS concept) that would be capable of reporting multimedia discharge informa- tion uniformly and consistently. This work resulted in EADS, which is both a resource for producing evaluated infor- mation and an environmental data repository (see Figure 1). Its structure permits full characterization of all data with minimum redundancy. Overview Basically, the system consists of data organization and data processing. ------- Data Organization Data organization and terminology are as consistent with EADS/FPEIS as possible. Thus, for PADRE, several impactor runs (samples for EADS/FPEIS) are logically connected to one test, with specification of the stream and operating levels embedded in the run data records. In particular, all runs for a given test share a common site and particle density and begin on or after that test date. PADRE utilizes user-specified keys (described below) to access all stored data. Three data categories may be entered by PADRE users: weight data, test data, and run data. A fourth category, impactor data, must be supplied to the EPA Project Officer before data can be reduced. • Weight data include a description of the substrates, pre-weights for each substrate, and post-weights for each substrate. Net weights are calculated from pre-weights and post-weights. Substrate pre-weights are entered into the system independently of run data. PADRE will assign a substrate's ID number when the pre- weights are first entered; users must specify this ID number when they enter run data. • Test data are the same for all test runs. These include the test com- ments, (assumed) particle density, site code, and test date. • Run data include run comments, operating conditions, identification of impactor used, and ID number of the set of substrates used in the impactor. • Impactor data include brand name, model, serial number, calibration date, and number of stages. For each stage, the number of jets, average jet diameter, Stokes' number, and cumulative fractional pressure drop are required. Data are entered and stored in units where they are commonly observed to facilitate data entry and correction. Data Processing PADRE involves four processes: data entry, data reduction, data comparison, and data designation for inclusion in FPEIS. In the entry process, observable data are read from an interactive compu- ter terminal operated by the user. These data may then be reviewed and modified. In the reduction process, effective cut points (stage Dso's) and mass concentra- tions for each impactor stage are com- puted under actual operating conditions. Mass concentrations are interpolated to standard intervals. In the comparison Administrative Data Bases Requests Status Reports •» Waste Stream Data Bases Fine Particles (FPEIS) Gases (GEDS) Wastewater (LEDS) Solid Waste (SDDSJ Fugitive Emissions (FEIS) "Continuous Monitoring (CMOS) ^- Reference Data Bases Standard Terminology (TERMS) Software Reports Data Analysis "Planned or proposed EADS components. Figure 1. Environmental Assessment Data Systems (EADS) overview. process, data from several user-specified runs are reduced and statistics at each standard diameter are computed. In the designation process, observable data from user-specified runs are printed in the format of EADS data input forms. The user may submit these forms to EPA to include data in the EADS. Prerequisites Prerequisites of using PADRE include EPA approval, data, and a terminal. Approval of EPA The PADRE system is operational at EPA's National Computation Center (NCC) on the UNIVAC 1100/83 computer. The system may be made available to all users of cascade impactors: EPA person- nel, EPA contractors. State agencies, and commercial and consultant impactor users. EPA provides this resource tool to analyze cascade impactor data with the understanding that users will share validated data from representative runs by submitting them for inclusion in the EADS/FPEIS data base. Data While new users are encourged initial- ly to practice with PADRE using artificial or sample data, it is assumed that most users will be using actual data from in- stack impactors. PADRE is no substitute for a careful record of all operating conditions and observed data. New users may wish to transcribe these data to PADRE data entry forms, as described in the PADRE User's Guide. This procedure will help users become familiar with the variable names and brief descriptions used in the prompts, prepare users for the order of the prompts, and ensure that all necessary data have been recorded. Calibration constants for the user's impactor are required in order to calcu- late the Dso's. These may be submitted to the EPA Project Officer before the PADRE system is used, so they will be available to the system when the impactor configura- tion is specified. Calibration information for a specific impactor is sought by the impactor type (e.g., Anderson and Brink), the impactor model (e.g., Mark III), and a unique number assigned to the set of plates used, in the calibration run. If a user does not know the calibration constants for his impactor but wants to conduct preliminary analyses, he may specify the impactor type and model with plate set ID zero to access "generic" calibration constants. The validity of the Dso's will, of course, depend on how closely his impactor corresponds to the mythical impactor the generic calibration constants describe. Terminal Access to PADRE requires an interac- tive computer terminal. All program printed lines contain less than 80 charac- ters. Any terminal that can access that NCC in Research Triangle Park, NC, may be used, but one that produces a ------- —Introduction to the PArticulate Data Reduction (PADRE) Srtltm— An outline for logical flew of the program fellow*. Your comments and suggestion* will be (elicited a* you exit fro* PADRE. ••ADD* PADRE BULLETIN BOARD INTRODUCTION TO PADRE? <- You arc herel I — >—> * * ENTER/MODIFY WEIGHTS? ENTER/MODIFY TEST DATA? I - ft post-samplin» weiahts test key* ft comment** particle density ENTER RUN DATA? UN ENTER MORE DATA? -<-YES-<-* *-<—<--< ~NO-<-. RETRIEVE A RUN? . » — >— YES~> — >-. operating condition*! repeat thi* section impactor id NO ,-->-->—-* mutt provide RUN ker* (SITE. te«t DATE. RUN NUMBER) LIST DATA? I MODIFY RUN DATA? REDUCE RUN DATA? DESIGNATE RUN FOR FPEIS? RETRIEVE ANOTHER RUN ! -<-YES-<-* +-<—<—<-NO—<-. all data associated with RUN correct error* in RUN data calculate DSO's. mass concentration* print EADS forms 6 ft 7a repeat this section 3. RETRIEVE A TEST? NO — >--> •» REVIEW ALL RUNS? must provide TEST ker* (SITE and DATE) Summarr with one line per run Calculate summary statistic* COMPARE RUNS? I CALCULATE PENETRATION/EFFICIENCY? RETRIEVE ANOTHER TEST? repeat thi* section *—<--<—<-NO-<-. RETURN TO BEGINNING OF PADRE? -<-YES—+ ENTER COMMENTS? TERMINATE SESSION You can follow several paths thpouah this proaram. You will be offered a series of options at decision points. At each decision point you will be asKed a Question of the form "DO YOU WISH TO ...? ENTER Y OR N." You may choose either to accept the option (e.a. enter data for a new run) or to continue to the next decision point, where xou will be offered another option. Indicate your decision by entering Y or NT then depres* tne return Key on the terminal. Any response other than Y or N will be rejected and you will asain be prompted to enter Y or N. There are three basic sections or IOOPS. At the end of each section you will have the option either to return to the beainnina of that section or to proceed to the next section. After the last section, you will have the option to return to the beainnina of the first section. There are no dead ends in PADRE. If you take a wrona path, you can aluays drop throuah to the bottom of the section, or of the proaram, and return to the beainnina. When you indicate that you are ready to terminate the proaram, you will be asked for comments about the proaram or its documentation. Please enter your comments while your experience with PADRE is still fresh in your mind. NOTE: rou can always abort & restart PADRE by! 1> gf>x tie This brings EVERYTHING to a screeching halt! •?> ffln This closes all the data files properly. This restarts PADRE from the beainnina. 3) Brun 43KXX.63022.... 4) Padd PADRE etc. Once you've done the "00x tio," the data files are left in unpredictable states of assianment in the system, so steps 2 ft 3 are necessary to reinitialize everything. Figure 2. PADRE introduction. ------- permanent "hard copy" to document the PADRE session is recommended. The basic requirement is that the terminal emulate a teletype at 30 or 120 characters per second and handle computer com- munications with even parity. Use of PADRE Use of PADRE includes user authoriza- tion, logical control, data entry and review, retrieving stored data, and sample sessions. User Authorization Users should check with the EPA Project Officer, Gary L Johnson, (9197 541-7612) to obtain the proper authoriza- tion to access the EPA/NCC. PADRE is readily availble to users holding a valid NCC account number; however, special arrangements may be made for other users to access the NCC (and PADRE) on a case-by-case basis through the National Technical Informa- tion Service (NTIS). All users will be expected to bear the costs of the computer use. PADRE Logical Control Users can follow several paths through this program. For example, the user may wish only to enter pre-weights for several sets of substrates, or may wish either to update or to reduce previously entered data. The user will be offered a series of options at decision points. At each decision point, the user will be asked a question of the form "DO YOU WISH TO...? ENTER Y OR N." The user may choose either to accept the option (e.g., enter data for a new run) or to continue to the next decision point in the program, where another option will be offered. The user indicates a decision by entering Y or N and then depressing the return key on the terminal. Any response other than Y or N will be rejected, and the user will again be prompted to enter Y or N. Figure 2 is a guide to the options this program offers. There are three basic sections or loops. At the end of each section, the user has the option either to return to the beginning of that section or to proceed to the next section. After the last section, the user has the option to return to the beginning of the first section. No "dead ends" exist in PADRE. If the user takes a wrong path, he can drop through to the bottom of the section (or of the program) and return to the beginning. Data Entry and Review Eventually, the user will reach a point when he should enter observed data. The computer will prompt him with a brief description of and the name associated with the data item. The computer will then wait for the user to enter the data. After all items have been entered, there will be a review of the data, listing both the names and values. The user will then have an opportunity to mark which, if any, items he wishes to change. After the marked items have been changed, there will be another review and opportunity to make further changes. When the user has been through the entire list of varia- bles and has made no changes, the system will proceed to the next decision point. Retrieving Stored Data PADRE manages all data storage and retrieval for the user, but certain user requirements are basically matters of good housekeeping. These requirements are also essential for retrieving stored data. • Substrate weights are assigned an ID by PADRE. • Run data are saved by PADRE using multiple keys. To retrieve stored data, the user must specifiy these keys exactly as originally specified or assigned. Whenever PADRE stores data, it prints a message with the keys that must be used to retrieve those data (see Figure 3). These keys should be recorded for further reference. If the user wants to recall data for an old (previously entered) run, he must specify the same site ID and test date as when the data was originally entered, aswellasthe run number assigned by PADRE. The user will be prompted for each of these items at the appropriate point in the session. If the user wants to recall data from an old test in order to modify the test data, he must specify the same site ID and test date as he did when the test data were entered. He can then change any of the test data; if any of the key items are changed, the system will alter them for all runs in that test. Sample Sessions Figures 3 through 5 show actual PADRE dialog for basic operations. A pre- viously entered set of data is modified and reduced. Both the terminal and the expanded output are reproduced. ------- 00 YOU WISH TO LIST THE DATA FOR RUN? ENTER Y OR N >N YOU WISH TO MODIFY ANY RUN CONDITIONS? ENTER Y OR N ENTER A HARK UNDER VARIABLE(S) YOU WISH TO MODIFY OR RETURN 1RUN RCOnn JDATE I START DUR RTYPE ««« ANOTHER RUN WITH THIS ID ALREADY EXISTS THE NEW RUN NO. IS 3 ENTER RUN COMMENTS >Thi$ run propaaated fro* II ENTER NEW VALUE FOR ISTART >!200 ENTER NEU VALUE FOR DUR >60 ENTER A MARK UNDER VARIABLE (S) YOU UISH TO MODIFY OR RETURN PAMBNT DPSTK VSTACK TSTACK TIMP DMAX CYCL DNOZZ CUTNOZ CMASS vou T0 ENTER A MARK UNDER VARIABLE(S) YOU WISH TO MODIFY OR RETURN BF FLARGE IMPNAM IMPMOD I PLATE FLOH METHOD ENTER A MARK UNDER VARIABLE(S) YOU UISH TO MODIFY OR RETURN IRUN RCOMM JDATE ISTART DUR RTYPE ' ««* MASS ON NOZZLE IS LOU DO YOU UISH TO MODIFY THE INVALID VALUES? >N ENTER Y OR N DATA FOR RUN 3 BY RTI AT RTP ON 32183 HAS BEEN SAVED IN THE TRIAL DATA FILE DATA FOR TEST BY CONTRACTOR RTI , AT SITE RTP , ON DATE 32183, RUN NO. 3 TCOMM- Demonstrate PADRE with sample Andersen data fro* CIDRS TEST COMMENTS manual PARTICLE DENSITY (GM/CC): RHO* 2.400 RUN COMMENTS: DATE OF RUN: START TIME: -JRATION RUN OUR RUN TYPE: X CARBON DIOXI X CARBON MONOX x NITROGEN: x OXYGEN: x MATER VAPOR: RCOMM* JDATE" ISTART* (MIN)I DUR* RTYPE* o This run propaaated fro* fl 32383 AMBIENT PRESSURE (IN HG) §TACK PRESS URT AMBIENT (IN TACK GAS VELOCITY (FT/SEC): STACK TEMPERATURE (DEG F): IMPACTORTEMPERATURE ( W H20) N2* 02* H20= 14, BO! 6, 8, OR TEMPERATURE (DEG F)J MAX. PARTICLE DIAM. (MICRONSI: CYCLONE CUT POINT (MICRONS): NOZZLE DIAMETER (INCHES): NOZZLE CUT POINT (MICRONS): BACKUP FILTER USED? BF* YES LARGE PARTICLES ON FILTER? FLOM RATE DETERMINATION! METHOD-UNKNQUN IMPACTOR TYPE: IMPACTOR MODEL: IMPNAM=ANDERSEN IMPMOD=UNKNOMN IMPACTOR FLOU RATE (ACFMK IMP. PLATE SET 1DMPLATE* 3 SUBSTRATES ID! NO. IMP. STAGES USED NS* 8 DESCR-GREASED NO. JETS DIAM. CUMDP CALIB NOZZLE 264 .1671 .0000 .3050 STAGE 1 264 .1281 .0000 .4300 STAGE 2 264 .0953 .0000 .4100 STAGE 3 264 .0780 .0000 .3850 STAGE 4 264 .0547 .0000 .3410 STAGE 3 264 .0359 .1760 .3200 STAGE 6 264 .0269 .2940 .3310 STAGE 7 156 .0253 1.0000 .2740 STAGE 8 IMPACTOR DP SCALING FACTOR 1.287 FILTER MANT TO RETRIEVE ANOTHER RUN? ENTER Y OR N >N DO YOU HISH TO RETRIEVE A TEST? ENTER Y OR N >N 1 'FOIL .000 MG .000 1 580 .180 1.530 4.390 1.900 .390 .540 MG PAMBNT* IDPSTK* VSTACK* TSTACK' TIMP- DMAX* CYCL* DNOZZ* CUTNOZ" 30.06 280.0 2 FLARGE* NO FLOH* IDSUBS* 1.000 .00 .358 0 Figure 3. Modifying run data. ------- HANT TO RETRIEVE ANOTHER RUN? ENTER Y OR N >N DO YOU WISH TO TO REDUCE THE DATA? ENTER Y OR N >Y PARTICULATE DATA REDUCTION AND ENTRY SYSTEM DATA FOR TEST BY CONTRACTOR RTI , AT SITE RTP , SAMPLING AT 109.4 X OF ISOKINETIC RATE WANT TO SEE RESULTS USING CLASSICAL (STOKES) DEF. OF DIAM.? >Y 03/21/83 16:48:06 ON DATE 32183. RUN NO. 1 Demonstrate PADRE's capability to accept free for* comments TOTAL MASS 7.9245-003 1.1979-002 1.8134+001 2.7412+001 GR/ACF GR/DNCF MG/ACM MG/DNCM ENTER Y OR N DMAX= 49.0 MICRONS STAGE NOZZLE 1 2 3 S 6 I FILTER *** *»* MASS MASS CUM. MASS (MG) (MG/DNCM) (MG/DNCM) .01 .02 27.39 1.22 3.03 24.36 .29 .72 23.63 l'.53 3 '.80 17i94 4.39 10.91 7.03 '•B 4-B Hi • w3 • 9 / 1 • «J^ .54 CUT POINTS FOR MEASURED DATA CUT POINTS FOR MEASURED INTERPOLATED TO HANT >Y DATA STD. TO PLOT CUM. X 1.34 STAGES -1 FOR STAGE, STAGES 1 FOR STAGE, INTERVALS: 0 J 2 ARE WITH ARE D50 GEO. MEAN (MICRONS) (MICRONS) 38.80 43.60 9.16 18.85 8.66 8.91 H5 J'S «9 • D£ ^ • wD 1.83 2.58 .86 1.26 SIMILAR 050= SIMILAR WITH 050* (VALUES 1 3 9 13 20 22 24 25 26 MASS? ENTER .47 .56 .72 .28 . 15 .76 .69 .95 .59 Y OR OR 38 OR 8 AT DIAH. • • 1. 1 2. 6. 10. 15. 20. N 30 63 00 25 50 00 00* 00* 00* .19 DM/DLOGD DN/DLOGD (MG/DNCM) (NO/DNCM) 2.45-001 2.35+003 4.84+000 5.74+005 2.97+001 3.34+007 K29+001 5 '.98+008 3.33+001 1.33+010 S-2I+8Si 5-95+9J8 3.24+000 4.46+010 4.46+000 4.89+011 INVERTED .80 OMITTED INVERTED .66 OMITTED MARKED WITH FROM INTERPOLATION FROM INTERPOLATION * ARE EXTRAPOLATED) .00 2.22+001 3.99+001 3.35+001 1.34+001 9.59+000 8.37+000 5.98+000 4.37+000 Figure 4. Reducing run data (sheet 1 of 2). W. M. Yeager and C. E. Tatsch are with Research Triangle Institute, Research Triangle Park, NC 27709. Gary L. Johnson is the EPA Project Officer (see below). The complete report, entitled "Paniculate Data Reduction (PADRE) System Reference Model," (Order No. PB 84-206 929; Cost: $ 13.00, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield. VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Industrial Environmental Research Laboratory U.S. Environmental Protection Agency Research Triangle Park, NC 27711 ------- 27 INPUT POINTS* 2 HERE OUTSIDE LIHJTS \ \ \ \ \ \ \ N 1 80.00 > 2 \ 2 \ \ \ 1 \ \ \ \ 2 \ 60.00 > \ \ \ \ \ 1 \ \ \ 40.00 > \ \ \ 1 N \ \ \ \ 2 \ 20.00 > \ \ \ \ 1 \ \ 2 \ \ 21 \ .00 >1 1 -1.000 -.600 -.200 .200 .600 LOG10< DIAMETER) \ 1 \ 1 \ \ 1 \ 2 \ \ \ \ < \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ < \ \ \ \ \ \ \ \ \ < \ \ \ \ \ \ \ \ \ 1.000 1.400 0.1 0.15 0.3 0.625 1.0 2.5 6.0 10. 20. DIAMETER (MICRONS) 1 INTERPOLATED POINTS 2 MEASURED POINTS 3 NEARBY POINTS WANT TO SEE RESULTS USING THE TGLD DEF. OF AERO. DIAM.? ENTER Y OR N >N WANT TO SEE RESULTS USING MERCERS DEF. OF AERO. DIAM.? ENTER Y OR N >N DO YOU WANT EXPANDED OUTPUT MAILED TO YOU? ENTER Y OR N. DO YOU WISH TO PRINT FPEIS CODING FORMS? ENTER Y OR N >Y PRINT DATA ON CODING FORMS FOR TEST 8Y RTI , AT RTP , ON 32183, RUN NO. 1 WANT TO RETRIEVE ANOTHER RUN? ENTER Y OR N •igure 4. Reducing run data (sheet 2 of 2). •&U.S. GOVERNMENT PRINTING OFFICE: 1984/759-102/10631 ------- 'ARTICULATE DATA REDUCTION At« ENTRY STSTEN Ol/tl/SI 14:401*4 DATA FOB list ir com*ACTO» FJTI , AT sue RTP . ON DA-I 11101, RUM NO. i TIST COItttllTS: OtllOOSlRATE MOPE MITH SAMPLE AlOtPSfH DATA FPON C10RS MANUAL •UN CCitlENTS OEI10IISTRA1E PADRE'S CAPA6ILIT* TO ACCEPT FREE FORM COt DENTS IHPACTOR DESCRIPTION »« C»LIB»»TIOH TtPE: AIAEPSEN M.ATI «T ID' I rtoti: I*M«VN inPACTOR DP SCALING FACTOR' ST/5E NO. JITS CUM. CUMDP CALID .i»7i .«ooe .im .0000 .*«S1 .toot .I7*o .0000 .01*7 .0*0* .*»* .17»» l.UT t44 144 (44 (44 ti« tt» is* .»os* .43*0 .4106 .its* .141* .MO* .nit .t7*» .«*» j.ooce .ss« Acre inPACTO* PHESSURI OP.OP « .!»• IN. OF H6 IMPACTO* TCHPtRATUIE • tM.t F HIT.* C) STACK TtnPERATVIRE • tM.t F <1J7.C Cl OPT CAS COMPOSITION IPIRCIMT) COt « 14.11 CO • CALC. HA'S LCAOIN6 • 7.«t«S-Ot) CR/ACF AT 10*.* X OF ISOKIHETIC RATE SAMPLING DURATION * 40.00 PUN 3TACK PRESSURE * 10.04 INCHES OF MB Nt « 00.00 Ot « «.00 HtO > 0.00 i.oi3««ooi DIAMtTEPS ARE CAICULATED HERE ACCOPDIN6 TO THE CLASSICAL I STOKES I DEFINITION ASSUMED PARTICLE DENSITY • t.»OSn/CU.CM. MAX. PARTICLE DIAMETER. • **.0 MICRONS IMPACTOR STAtt STACE INDEX NO. Nome i si t St 1 DSO IUSS IBSI us/one n/STASE CUM. X OF MASS < Oft CUM. russ I:«/ACNI < OM CUM. russ iiie/Ducm « DS* CUM. mSS IC'/ACFI < DfO MISS ICR/DIICFI < DSO MEA-I DIA.iniCPONSI II1G/BIICMI 011/010:0 II.VSER/OHOO ••• CUT FO:i"S F0» STAGES -1 MC'fCPtO O.'TA FO* STAGE, ••• CUT POINTS F01 STUGES 1 DATA FO* STAGE CUM. CEO. 16.60 «.14 6.44 .01 l.tt .t* t.4*-00( 3.0X090 7.11-0*1 »».»! 66.CS •*.(( 1.614*01 1.41*001 1.S44001 t.744*01 t.444001 2.144001 7.«(-001 7.04-001 4.61-001 1.10-OOt 1.06-001 1.01-00! 4.144001 1.6*4001 6.*l400« (.45-001 4.6*400* t.*740*l (.154001 5.7*4005 1.1*4007 0 A»E SlnlLAP OR INVERTED 0 MITH 050' 16.60 dimtO FROM INTERPOLATION t APE SIMILAR OR INVERTED t WITH OSO« 6.44 01111110 FROM INTERPOLATION SI S* SS St S7 SO 4 1 « 7 • * (.M 1.4* 1.01 .U .SS .17 .SO .!• l.S) *.!« l.*0 .1* 1.4*4000 4.47-031 S.00<000 l.Of'OOl 4.7Z.OOO *.4*-001 •0.«4 7*. IS 4S.44 IS. 44 1.41 *.«• .47*001 .it'OOl .4Z-001 .70-00} .7S>000 .4S.OOO .7i<007 .*»4001 l.lt'OOl .174C01 1.794001 .l*-t01 .SO-101 .344000 .774900 .4S4007 .H-001 .64-001 .5*4001 .2*4001 .»8«OOB .45-000 .OJ'OOO .03-001 .07-003 .(44000 .134001 .134010 .SJ'OOO .114000 .46-004 .01-001 .84-001 .174001 .644010 .66-001 .144000 66-004 64-004 64-001 £44000 444010 FILTER 10 .17/t. 0 .1* 1.14* 00* l.*4-001 4.444000 4.0*4011 INTEtPMATEO VALUES IVAtUCS AT DIAMETERS HARKLD WITH • ARE EXTRAPOLATEOI S'O. DIA. IMICROIISI .100 .425 1.000 l.tSO ».§00 4.00* Cin. MASS IMI/Dllcni 1.471 1.S41 0.7tt 11.176 tO.lSS ll.TS* 10.006* t4.4 1S.OOO* 1S.»SO 10.000* !4.S*S IMG/DIO1) .*• t.t(4001 l.*»4*01 S.lS'OOl 1.144001 *.S*4«M 6.>7<0*» S.«««000 4.174000 WPHAL ItMSIKtERIMG STAIIDAROI COXDITIOIIS ARE tl DEC C AID 740t*1 W.I Figure 5. Expanded output: reduced data (at NCC). United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 Official Business Penalty for Private Use $300 po U b\ ------- |