SPTP MESO MODELING
SPTP MESO MODELING ATMO 689
Popular in Course
Popular in Atmospheric Sciences (ATM S)
This 225 page Class Notes was uploaded by Eduardo Lowe on Wednesday October 21, 2015. The Class Notes belongs to ATMO 689 at Texas A&M University taught by Staff in Fall. Since its upload, it has received 56 views. For similar materials see /class/225951/atmo-689-texas-a-m-university in Atmospheric Sciences (ATM S) at Texas A&M University.
Reviews for SPTP MESO MODELING
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 10/21/15
Chlorine reservoir distribution HALOE I Pressure mm 3 HO Mixing Ratio ppm i i i i i i i i wow 2 36 431 40 ago a 20 4a 513 Latitude Plate 20Ei b Observed zonally amraged dish1quot bution from 8003 to 60 N ofmetheme above and hydrogen chloride befow by the HALOE instrument on board the Upper Atmosphere Reseamh Satellite UARS fmm September 21 to October 15 1992 courtesy 0f J M Rume III Hampton University Virginia Brasseur Orlando Tyndall OUP 1999 Altitude km Typical Antarctic Ozone Loss Arctic Ozone Loss in Winter 19992000 ozone loss destruction by chemical i I I I i 2 4 6 8 Ozone concentration 1012 cm393 Altitude km ozone loss by chemical destruction 6 8 Ozone concentration 1012 cm393 Stratospheric ozone depletion connections Plate 21 Lower stratospheric 111ing ratio of CM ppbv HNO3 ppbv and ozone ppmv on the 465 K ieeutropio surface as well as ozone ooiumu abundance Dobeon Unite measured by the UARS Micronewe Limb Sounder MLS on Januaijy 30 1996 in the Nomileru Hemisphere The location of the polar vortex is indicated by White hues Chlorine activation is I39isibie in the pow 1311quot regime where the temperature is low and poiar stratospheric Clouds P305 are formed Within the black contour inside the poiar vortex Denitri cation is wisihie Within the PSU region The ozone column is low inside the poiar 7 vortex courtesy oi39J Wquotaters Jet HN03 ppm 39 Propulsion Laboratory Pasadena 30 50 m 90 110 I10 so 200 24x1 230 320 ca fm njaL Brasseur Orlando Tyndall CUP 1999 GOME NRT QCIO003O201 SC OCID mm quot molec cmquot L m Ln mquot 50 vol 75 mquot s a you 45mquot smw GOME NRVTVOCIQ g0030227 SC OCIO 5 on molec crrrii if mm G 0mm IZ0 up Handbag VJ mp Humequot deudmnm iupchya kun bramende Ln mu 90m UP Had wa yup amm Mknumnmupnirya39kuu bramande SC OCIO SwagIE NRT OQI 20020917 SC OCIO moleccm H 3 cm Lilo 9 I210 man men 90 w 90 w 75m 75m sum 5w 45 w i 45 w39 i g A 43 man sumquot GOME NRT OCIO 20020923 SC DCIO Sc OCIO 39 ch zj moleccmzl LEWquot Lzmu mm mm 90mquot 9m 75m 7m 50mquot sumquot 65mquot 4W mam 3010quot m Ism LSw DLR I ml 00U 00 W39 up madam MP 3mm Falkan aidebk mum Mkuuvnnmmu L J manyMimi bramende swam 39 ESADLRHUP Emmaquot GOM E 03 20020930 GDP 3 ESADLRHUP Bremen 39 10 e cnmannmup physt unirhremen c2 Ozone depletion potential ODP Definition Ratio of global loss of ozone from X per unit mass emitted at steady state compared to loss from CFC1 1 per unit mass reference Can be chemically modeled or semiempirically derived Solomon et al 1992 GDP FXFCFC11 X MCFc 11MX x nX3 X a X TT t F instantaneous photochemical chlorine production n number of Cl atoms a enhancement factor for Brl T1 t atmospheric lifetime function Ozone depletion potential ODP TABLE 133 Atmospheric Lifetimes and Steady State Ozone Depletiun Potenti als ODP Predicted Using Either a TworDilnensional Model or a Semiempirical Method b potential Almosph ric Steadystate ozone depletion trace gas lifetilne years Model Semiempiricalquot CFCell 50 10 10 CFC12 102 082 09 CFQ113 85 090 09 CHqCC13 54 012 012 HCFC722 133 004 005 HCFC123 14 0014 0102 HCFC7141b 94 010 0 1 HCFC142b 195 005 06 HFcl a 14 lt15 x 10quot lt5 X 10 HIGHS 36 lt3 gtlt 1m5 C 3Br 13 07 004 0570139quot H4301 12 13 H1211 51 5 CHIClBr 02370136 0098 019 CHEBrCHZCHJ u 29quot quot Based on Eq A as r so quot From World Meteorological Organizatiun 1995 C From Wuebblcs er a1 1998 1999 me lifetimes and ODPs given for cngcusx are the range with and without an ocean sink quot Valu s 1 0m K0 6 a 1998 estimated using the more recent kinetic and pholachemical data er DsMore at 111 1997 Phase of an Electromagnetic EM wave Sins 01 Angle l a A I 39 I 1 O 39 939 t t t 1 O 90 180 270 560 Angle deg Figure 61 Elna wave solid curve and a ascend eignal 50 out of pirates with the ret wave daahaal curve Rinehart 2004 Simple graphical example of a 30 0 phase shift Phase I a the fraction of a full wavelength angle in radians or degrees a particular point on an EM wave is from the nearest reference point Doppler Frequency Shift and Doppler Velocity 1 d D fD 27r dt fD Doppler frequency shift D Doppler phase shift VD Doppler velocity 7 wavelength 2VD A where ffrequency of transmitter Doppler Shift Frequencies in Hertz for Various Radar Wavelengths and Target Velocities Wavelength Cm fD ltlt f 1 XL3938 LS 32 55 1UU For typical meteorological m ll 6 4 2 i 7 velocntles Doppler radar receiver A 1 322 263 mUSt be very carerHY deSigned mop i I i 39 11111 6236 3636 2000 in order to detect such small Doppler phase and frequency shifts fD relative to much larger transmit frequency f Battan 1973 Table 81 Radial Velocity Vr VF Vcos 05 Note A Doppler radar only detects the radial component of the velocity ie towards or away V target velocity 0L angle between target motion and radar pointing directions Figure 62 Geometric reletionehip ota target located on the center of the antenna beam axle moving with velocity V at an angle cc relative to the pointing direction The radar eleteete the radial component of velocity v Rinehart 2004 Simplified Block Diagram of a Doppler Weather Radar STALO stable local oscillator maintains very stable frequency fO COHO coherent oscillator locking f Ampifies and 5mm oscillates at an rderuuinrsw frequency fc that has same phase relationship as transmitted wave intermediate or 7i Figure 635 Biock diagram of a simple Dappisr f f 0 radar iF amp intarmcdiata frequency f0 STALO frequency amplifier STALO tabl i focai oasiiiator COHO fci 39 termed39at frequency cahsnsnt osciiiaizon Th5 top display shows f0fc transmitted frequency 2 t 1 Tc t m h b foHcHB received frequency r3 5i he 56 M y as or w it t 5 attain fD Doppler frequency shift di piay 5h0w Doppirsr radial vaiocity Rlnehart 2004 The Doppler Dilemma 61 8 rmax V max It is typically desirable for both max and rmax to be as large as possible to minimize data artifacts and misinterpretation But Vmax and rmax are inversely proportional the Doppler dilemma as rmax 9098 UP Vmax goes down and viceversa connected via the pulse repetition frequency PRF Maximum Unamblguous Valacity We lOOO FR F H7 00 lo 100 lOOO Maximum Unamlaiguuua Range km Flgurs 54 Summary of conditions for ram 5 and valocity folding lieu tlm Doppler dilamma The numbers near the bottom are wavelength frequency Notice that rm at FKF and i5 independsnt 0f wavalength After Goeaard and Strauch 7955 Rinehart 2004 Recognizing range aliasing or folding r1 range of actual storm causing range folding r2 range of actual storm NOT causing range folding rmax maximum unambiguous range r1 gt rmax range folded r1r max Figure 65 5 ulatcd FFl display ahowihg a real echa loz ted to the northeast To the a dietahee of r Fram the radan lt alao real ahd the alia ed poaitiahe but ite aliaeed azimuthal width i much harrower r eah be dieplayed at he Wrong range Two real aghaea exist The Hrat l lees than r awayahd From i5 displayed at the corrst range The aecond i5 Rinehart Range aliased echo beyond r 39 it is dieplayd at a range of r rm The faiht dashed atar hear the radar l5 where 2004 has a wedge lke the radar would diaplay he diataht storm appearance to It because of geometry of beamwidth r2 lt rmax no range folding SMARTR1 SMARTR1 6 Aug 2005 222030 UTC 6 Aug 2005 223759 UTC Rmax 150 km Rmax 60 km 15quot PPlon 15 PPlon No range folding evident Range folding evident Recognizing range aliasing or folding Recognizing velocity aliasing folding Example N 15 m s1 maxl Figure 664 The quotDoppler epeedometer The abaeieea i5 thedirectlon theantenna ie pointing and the ordinate ie the apeed detected in all three 521585 the radar is detecting a uniform wind blowing from the weat made detectable For example by the presence 01 ineecte For the caee at top right the velocity ie about a third of Vmax For the middle ease the velocity Ea about two thirde of ijA For the bottom oaee the apeed ie juet over Vmand 15 aliaeed when the radar looke both upwind and downwind 6d a 885 J Speed WIquot5 Vrriax 1 5 if C Uniform westerly wind looks like sine curve Azimuth away 13 Vmax toward N 5 W N 23 Vmax N E e w M Aliased itmax g e m 2 6quot E N Adapted from Rinehart 2004 Recognizing velocity aliasing folding SMARTR1 06192004 MCS squall line 510 PPI aliased Radial Doppler Velocity Radar Reflectivity Facto r Chromatography Chromatographic methods are based on equilibrium partitioning of an analyte between the a mobile phase and a stationary phase K conc in stationary phase I conc in mobile phase K f type of phases type of interaction T Increasing the partition coefficient K leads to increasing retention on the column later elution K is achieved 102 to 104 times in column chromatography high resolution power GC is one of the most versatile and widely used analytical methods available Chromatoqraphic techniques in atmospheric analytical science Gas Chromatography GC Mobile phase is a gas He N2 Ar Air H2 carrier gas Stationary phase is polymeric liquid or solid Liquid Chromatography LC Mobile phase is liquid water solvents mixtures Stationary phase is solid polymeric or mineral GC used in atmospheric science for permanent gas analysis N20 CH4 C02 CO etc VOC analysis major tool in air pollution research LC used for selected VOC analysis generally low volatility andor high polarity compounds GC schematic Thermostats Recorder IOYOYOYQI Detector Flow Column Controller Enlarged Chromatogram Cross Section mommaas M Figure 117 Schematic of a gas chromatograph The sample is injected into the carrier gas di erent components pass through the heated column at different rates and thus are detected by the detector at di erent times from Okamura and Sawyer 1978 Brasseur Orlando Tyndall AP 1999 39fz laqueaxa 1 0 Ksaunoo Jge 1119qu a smuggle qdm og ewmqg 9 911185 SEIJJ39INIIN WdPEG39GLS39H Ewasoc m 4 C3 8 c o0 PROPENE ACETALDEHYDE 1BUTENE lSOBUTENE METHVL ACETATE METHYL PROPANAL 4 II METHYL BUTANAL DIMETHVL FUHAN TOLUENE N O BENZONITRILE N D m 9 Z I Brasseur Orlando Tyndall AP 1999 UJBJbollBUJOJuQ GLLL GC Detectors used in Atmospheric Science Flame Ionization Detector FID Detects the small current induced by ions created in a flame inside an electric field Highly sensitive to CH bonds in VOCs Thermal Conductivity Detector TCD Detects differences in thermal conductivity with respect to the carrier gas low sensitivity used for permanent gas analysis Electron Capture Detector ECD Detects reduction in current from radioactively generated electrons inside an electric field Expensive highly sensitive to electronegative gases such as N20 and all halogenated trace gases Mass Selective Detector MSD Uses separation of masses in a quadrupole mass spectrometer Detects and amplifies counts sensitivity depends on count rate r WWW The Chromatoqram Gradient oo so W A so u3A 40 L 20 t Acetonitril o 5 10 15 Zeitmin Sample llliiIiliiiiIilll I O 5 10 15 20 25 Time elapsed in min A typical chromatogram showing airborne aldehydes as their DNPH Dinitro Phenylhydrazine derivatives separated by reversed phase HPLC Conformations Newman projection of butane THE CH CH H H H c H CH3 Jig Hm H CH3 H H H E E mm W m39 Overlap leads to 3 steric hindrance amp 39E 1 a m u z a 3 E f 3 2 a i 2 u m a o W m cunfurmer pmenuals a bums abom central cc band Alkenes and Alkvnes carry one of more Trbonds instead of obonds only in alkanes sp2 hybrid for alkenes at triangular qorbital structure remaining perpendicular porbital sp hybrid for alkynes linear qorbital structure remaining two perpendicular porbitals increased reactivity relative to alkanes due to high electron density in the n bonds the OH radical is electrophilic Coniuqation Aromats Trbonds can be alternating with obonds in a chain is there a preference in orientation evidence for localization vs deocaization Delocalized double bonds localized double bonds delokalisierte lokalisierte n in ungen 71 in un 154 pm2 134 pm 2 u 148 pm1 137 pm1 keine Uberlappung Uberlappung kein Doppelbindungscharakter 1 experimental evidence 2 data from individual bonds Molecular Properties Symmetry axial rotation plane mirror examples for highly symmetric molecules ethene benzene tetrahedral for saturated C single bonds can be freely rotated gt conformations double bonds increase rigidity rotationally challenged conjugated double bonds further planarity gt aromats how do these come about MO theory for orqanic molecules Knoten O antibonding Eu Hr bonding 0 c m U E 9 E n l E u 5 8 n u g E 71465 kJmol GcHgJcc EV A IGSO kJmo 76ml 2050 kJmol E7042 111 organic molecules norbitals double bonds TriEnergie Eioktvoncn xpWeIIen A ahl Hivedux beselzung funktionen Knoten K 72 77547171618i0 3 antibonding E3 333338 O 521 Oa osf bonding O E1u16 18 H Knotcn E 754 kJmol E 343 kJmol Conjugated double bonds 1 3 butadiene n Energie Eu gt E 963 kJmol E 71256 kJrnol Aromatic rings Benzene Stratospheric NOE Direct input from aircraft is small Natural nitrous oxide N20 major source N20 hv gt N2O1D Alt220 nm N20 O1D gt MO or N202 11 Catalvtical Ozone destruction NO03 gtN02O2 03 hv gt O 02 O N02 gt NO 02 Net 03 03 gt 3 O2 Stratospheric Chemistry I NOX ALTITUDE km 1 39 39 010 50 100 300 10 so 100 300 MIXING RATIO nrnol molI l39IGURE 38 Vertical pru les of the N70 mixing ratio at low and high latitude From measmemenls of Tyson a al 1978a Vedder et al 1978 1981 Fabian et ul 1979 1981 Goldan at all 1980 l981 The sohd lines are results of calculations by Gidel et ul 1983 based on a modimensienal model Stratos heric Chemist NOM chemistm Figure 71 Schematic dia gram of oddnitrogen inter actions in the stratosphere APPROXIMATE ALTITUDE km Stratospheric Chemistry IV 60 02 56 03 52 5 4e 0397 13 10 m 44 2 g 40 3 g 36 5 U 7 m 32 10 g 29 20 E 24 30 2 3 16 woo 1039 10399 10397 ODD NITROGEN VOLUME MIXING RATIO Wva NOZ distribution Figure 76 Vertical distribution of NOy and its constituents in the servatians are for May 1 at sunset from 8 1985 Russell st 211 198 N01 distribution COLUMN DENSiTY 1o 9molecuie m2 EOU 20 0 DEGREE LATITUDE FIGURE 31 Total column densities of NO2 and HNO3 in the stratosphere as a function of latitude in the Northern Hemisphere 0 summer winttr Adapted from Coffey et a1 1981 SON Stratospheric ozone chemistry summary ALTITUDE Km O 90 gt60 0on I WWImm 30 0 30 LATITUDE deg Percentage Ozone Loss Rates June HOX 6O ALTITUDE km lt2 60 90 90 60 60 90 m o I o I ALTITUDE km N m A o o I I 8 r A 9 60 LATITUDE deg Figure 148 Relative contribution of various chemical families NOX r x and x to the destruction rate of odd oxygen in the stratosphere calculated by the NASAGSFC 2D model as a function of latitude and altitude for June conditions courtesy of C H Jackman NASA AWW i M W o 0 0 30 60 ALTITUDE km 90 lt1o Che 20 ALTITUDE km 90 60 90 o 30 0 so LATITUDE deg 11551 12 1 QEIJOH39OSIB 305 VSVVNJG iTS39aiJHOJ J 5551 39qJI tqvug Buying If 331111 aignamg IIDJESSBH aqus39mmv Jedd arr pmoq rm mamrugs39nvr SHE1 elf4 ifq pamstaom Aqdd UL mm Ef pcyn ppm Hymn an Jm Imunqpr 1er pa mm39e immag QI Squid 53m aanLIJm 05 m 08 03 m a m Liza us 01 ns PRESSURE ALTIi ALTITUDE mm TUDE mbar 2661 HQJE W u grsm 10 HIBEIHWE M mm mm 5505 IlWI u39 THEM gigffawg qJJEQSQH aJarlrdsmmV Iadd am 1112th Ho mamnnsm 333 any IL pamgmm 1qdd m mam fingan EIIIGQ I39HJUJOHHOJOITID amt um nognqlmgslrp 933113113 A39H Hnog gI agal qd 59 P SCI iliV l mam EIGHJLUJV aanssaad um aumuw Stratospheric Chemistry V Haloqens halogenated hydrocarbons generally longlived in Troposphere gt penetrate to Stratosphere natural halogencontaining HC CH3CI CH3Br CH3I HCCI3 etc oceans are major source anthropogenic compounds CFCs HCFC Halons CHgCClg CH3Br etc refrigerants propellants flameretardants solvents etc stratospheric impact through photolysis eg CCI3F hv gt CFCIZ Cl A lt 250 nm Sources of Chlorine 235 550 1 0130612 230 540 2m 7 520 EN 265 500 E 280 quotE 43039 1 255 450 25a 7 D 245 m5 393 44D f 1 DU 42 t A SW10 24D 7 1 U SPO 235 u 400 1 1 1985 1990 1995 2000 1985 1990 1995 2000 1110 150 r 141 r 35 120 39 H 100 100 r E 31 E 95 CHsCCIa CCI4 4D 35 20 0 r 7 an 1985 1990 1995 2000 1985 1990 1995 2000 Stratospheric Chemistry V Haloqen chemistry Halogencatalyzed ozone destruction CI 03 gt CIO 02 CIO O1D gt CI 02 Net 03 O gt 2 02 mostly higher elevations CI 03 gt CIO 02 CIO H02 gt HOCI 02 HOCI hv HO 39Cl 39OH 03 gt H02 02 Net 2 03 3 02 high HOX regions Stratospheric Chemistry V Haloqen chemistry ChlorineBromine connections CIO BrO gt Br CIOO or OCIO Br 03 BrO 02 CIOO A gt CI 02 OCIO hv gt CIO O Halogen temporary reservoir species CI CH4 gt HCI CH3 CIOlBrO NO2 M gt ClONOlerONO2 M Br H02 gt HBr 02 HCI HBr OH gt H20 CIlBr CIONOlerONO2 hv gt CIlBr NO3 CIO mixing ratio 55 1 45 7quot L E E as r a g s 35E g m 5 9 3 BL 100 39 39 39 39 39 39 15 90 m 30 n so 60 so Latitude Plate 22 Zona y averaged daytime C10 mixing ratio ppbvj between 15 and 55 km for A ugust October 1992 observed by the MLS instrument on board UARS The Illaxjmum concentrations of 165539 than I ppbv are Visible in the upper stratosphere high latitudes High CEO abundances seen sou t1 01 60quot S in the lower stratosphere result from fluorine activation on polar stratospheric doudo from Jackman et af 1996 Stratospheric ozone chemistry Summm Percentage Ozone Loss Rates June HOX I I l I I 60 50 7 i 40 i E E a 3quot a E 20 r E lt lt 10 0 D 90 60 60 90 90 50 0 30 0 so LATITUDE deg 0on BrOX l l Sl 50 7 m 0 k E 40 e E E E 3 3 D l I 393 20 E 393 lt lt 10 L o 0 90 760 30 0 30 60 90 90 3960 30 0 30 60 90 LATITUDE deg LATITUDE deg Figure 148 Relative contribution of various Chemical families NOX HO 7 Br x and x to the destruction rate of odd oxygen in the stratosphere calculated by the NASAGSFC 2D model as a function of latitude and altitude for June conditions courtesy of C H Jackman NASA lt10 ALTITUDE km 60 90 o 30 o 30 LATITUDE deg Brasseur Orlando Tyndall CUP 1999 Ozone DU Global Stratosgheric Ozone Trends Ozone at Arosa Switzerland since 1926 I I one year smthed Irena 1926 71973 300 7 0 Idecade trend 197371997 72 9decade I I 1990 I I I960 I970 Ye ar I I I I930 1940 I950 1980 2000 ATMO 689 Lecture 8 102604 Polarimetric Radar Data Processing BC Ch 66 Elimination of nonhydrometeor radar echo eg ground clutter anomalous propagation clear air returns non meteorological targets using polarimetric techniques Apply simple threshold to the correlation coefficient phv Apply simple threshold to the standard deviation of the differential phase 0 Pdp Estimation of the specific differential phase de Finite difference formula and standard deviation of de given presence of measurement noise Two techniques for reducing the effects of noise Filtering or smoothing the range profile of Pdp Linear regression fit to the range profile of I dp Elimination of nonhydrometeor radar echo Statement of the problem For hydrometeorological applications it is desirable to isolate hydrometeors ie cloud and precipitation particles from nonhydrometeors eg ground clutter and socalled clearair returns which is actually insects and sometimes birds Nonpolarimetric radar techniques Analyze elevation or height variation in echo structure Problems with shallow systems Subjective Create a clutter mask by statistically characterizing ground clutter at a site using long periods of nonraining data Does not account for anomalous propagation Doppler clutter filters typically eliminate radar echo with nonzero Doppler velocity andor nearzero Doppler spectrum width Works reasonably well but can eliminate precipitation echo Where s the ground clutter Radar Reflectivity Z i x x d BZ an no 40 m Elevation Angle 050 Ryzhkov and Zrnic 1998 Where s the ground clutter Radar Reflectivity z dBZ dBZ Elevation angle 050 Elevation angle 150 Ryzhkov and ZrniC 1998 JTECH 15 13201330 Elimination of nonhydrometeor echo Polarimetric radar technique Rainfall and other hydrometeors are typically characterized by a correlation coefficient near unity while clutter and clear air have very low correlation coefficient Eg Rainfall phv gt 097 clutter and clear air phv lt 06 typically Therefore it is possible to chose a simple threshold in phv that separates most hydrometeor echo from most nonhydrometeor echo Even when the correlation coefficient is lowered due to the presence of large wet hail large melting aggregates or mixtures of hydrometeors rain and hail values are still typically in excess of 08 Value of phv for a given hydrometeor type is dependent on the quality of the radar system design The better the polarimetric purity eg good matching of sidelobe patterns at HN and good crosspol isolation the higher the phV Must consider quality of radar design and corresponding data in order to choose the appropriate threshold Know polarimetric properties of rainfall for your radar and then conduct sensitivity tests Z dBZ EIO5 Applying phV threshold Z dBZ E15 V l i k Hana l v 39 3 dedegkm EIO5 Fxc 1 3H0 Composite plot of radar x e ecm xty fade z and d spem c di erennal phase Km for the 5mm of 7 October 199539 a E o 32 lt El 5 and c El 0539 data mm p 07 are lotted The radar is socaxed m we lower ngm come on mm Comm oil are 1 cu apzm starting from 20 dBZ Numbers 111 c indicale the Oklahoma Masouu gauge locations The tune is 1 UPC Ryzhkov and Zrnic 1998 Elimination of nonhydrometeor echo Polarimetric radar technique continued The standard deviation of the differential phase 6 Pdp in light rain or drizzle is typically a few degrees or less amp g 2 to 3 Again depends on quality of radar system design The value can be larger smaller for a poorly well designed system For moderatetoheavy rain 0 Pdp increases proportionally to specific differential phase de and the rain rate R but rarely ever exceeds 9 over a path length of several km eg 34 km For ice hydrometeors whether ice crystals aggregates graupel or hail 0 Pdp is typically small 2 to 3 or less unless the particles are in the Mie scattering regime ie not small relative to the wavelength Mie effects are typically not a problem for Sband but can become an issue at higher frequencies Cband Xband etc Simp gx Swppmgg m f mwmd thtw A a Z 1 quot Thresho dg Ugmg PQ ammcgmc Radam T KEWWUQUQ gt pm 9 SWdPJ mrzslsala sna SPOL SUR L1deg 2 DZ leEIBSHI En SPOL SUR 11 deg 2 DZ NCAR SPOL radar gt 08 and 3 UP lt 12 TRMMLBA Experiment phv dp 23 July 02 N 01 DZ unedited 2 12 UTC nmannz 2nnss quoter PPI n5leg 1 DZ 23 July 02 N pol DZ edited 2012 UTC 45 Thresholds phVgt 07 GCPdp lt 18 Summary for Nonhydrometeor Rejection by Polarimetric Methods Advantages of polarimetric approach Simple empirical technique thresholds in phv and 0 I dp Solid foundation in theory Thresholds determined from theory quality of radar design and data and sensitivity testing Effective 999 of the time Disadvantages of polarimetric approach The O1 failure rate can be troublesome especially if systematic Some subjectivity in determination of thresholds No exact perfect thresholds raise them higher to delete more nonhydrometeor echo and you will end up removing hydrometeor echo Lower them to keep more hydrometeor echo and you will allow more non hydrometeor echo to pass through Future work investigate variable thresholds as a function of Z or power Current method does not eliminate 2nd trip echo need to look to other variables possibly value of LDR or I dp relative to typical Estimation of Specific Differential Phase de Recall that the differential phase po is defined as where ltgtdp differential propagation phase Cumulative phase shift associated with fonvard scatter Use to estimate de and hence rain rate 8 backscatter differential phase Mie effect negligible at Sband except in very large hail deg CD noise Random measurement and other sources of error CIoffset Constant system offset phase Known engineering quantity Does not affect calculation of de since it involves phase differences RADAR LPdp dp5 I noise of et 9 de dpr2 dprl z L1161111702L1Japr1 2r2 r1 2r2 r1 when 5 and 1310158 are very small lt1 For above s tan a ara deviation 039 is SDLPdp Jim n 0de Choral DPr2l r1 I392 Fig 517 Differential phases corresponding to resolution volume locations at r and r2 needed for Doviak and Zmic 1993 BSUDJHIIOI of spec1 c dxllerentlal phase over the instance r1 r1 The problem is that statistical fluctuations associated with the measurement result in a gatetogate variabilityaccuracy in Pdp of 2 3 in rain Further 6 associated with Mie resonances is not uncommon at X and Cband large raindrops melting hail and even at S band in very large hail lf 0po 3 and r2r1 3 km 0de 0710 which is too large for use in quantitative rainfall estimation or even qualitative microphysical applications Need strategy for mitigating measurement noise Method 1 Range filteringsmoothing Method 2 Linear regression Method 1 filtering in range is required to estimate the true trend of ltde and iterative filtering is required to eliminate large 6 say at C or X band or higher frequency Since ltde is a rangecumulative quantity it should vary smoothly with range From Bringi and Chandrasekar 2001 as adapted from Hubbert et al 1993 and Hubbert and Bringi 1995 IIR Infinite Impulse Response Filter FIR Finite Impulse Response Filter Filter Response n m a a 39 5 39 t 5 75 5 375 a 25 1575 Spatial 13ewavslength scale km l l l r u o Magnitude Response 18 l u bi lflllgl lWl 75 5 Spatial ewavelength scale km 39 4 632 a Magnitude ul39 39 39 V 39 Iall eylulil u 39 150mnpanW 39 HI v H t pd at ma a p impulse response lIR thlrdeordcr Buttcrwonh lters Filter quotAquot IS a light lter designed to lter out the rapid gatcetoegulc uctuations while B is a quotheavyquot lter designed a preserve the mean increasing trend 0t wdp with range 1 as m a except the magnitude response nl n twenllethllrder rlmte impulse response ller FIR is shown a From Hubbert et 41 m3 and h from lluhhm and Bringi I995 Range Filtering Method 1 Result of IIR FilteringSmoothing 725 F a quotrawquot wdp w deglael la a Filtered using IIR lter quotAquot In Fig6323 45 l l 22 24 26 25 Range km 25 U Filter quotAquot we degree J m 45 Filtered using IIR lter quotBquot Fig63211 15 20 22 2A 25 28 Range Km Flg 631 4 Example of filtering raw Ild data using the lter markcd A in Fig 632m Radar data from the C band DLR lddl39 located nem Munich Genmtny 12 As in 1 except the ller marked quotBquot in Fig 632m is used The input to this lter is the ltered curve tn 1 From Hubbert et al 1993 Filter severely dampens gateto gate gate 100300 m statistical fluctuations while maintaining mean trend over 24 km in order to better estimate de and R over that 24 km path length Dislzlncc N ul C pol lezu km Distance N S of Cpol radar km Example of Differential Propagation Phase dp and Specific Differential Phase de Estimation in rain at Cband 55 cm 1413 CAREY ET AL r l 770 7610 750 in 730 720 4390 0 IO 20 3390 40 5390 0390 7 0 30 9010 Distance E of Cpol radar km I II I ZhuBZ 0 10 20 30 40 45 50 55 60 Lincnrrtctcd 18 Nm 95 4 l6 ETC 2 inn AG Zh dBZ contoured wt m Zdr dB shaded Dixumut N UFCltpol mdarkmi 4 4 4 u is S b 4 391 l 7 9 l H l5 7 19 DNan E in Cpnl numka AlumJ 2 l 4 5 I 05 I I5 2 J 4 5 Distance EW of Cpol radar km Unpll gi m l 0 10 20 30 40 50 60 70 80 90 100120 W E E 5 I dp deg dashdot BO a 45 50 4 40 g g bdp deg dark solid 339 v 30 if E 20 E Using FIR filter 10 E 5 0 39 8deg light solid quotquot3060 50 10 3390 2010 0 10 2390 3390 40 5390 60 70 80 90100 E de degkm dot Carey et al 2000 Linear Regression Use linear regression over a sample collecting in range r over some of samples N Typically over some path length L usually 24 km for convection or 812 for stratiform rain LNAr where N of phase samples in the line fit and Ar is the gate spacing that is assumed uniform eg range of 30300 m with 150 m typical Method 2 See BC 662 N len we 7gtl de 11 N 2201 FY i1 where Standard Deviation of de for method 2 For the above method the 5VTquot quot standard deviation 6 of de K de km can be estimated from G dp 9 amp W 2 Arrow 1 mm31 15 2 2 5 3 35 4 l path length L in km L N Ar Flg 635 Standard deviation of de estimate obluined as the slope of n straight line t to he Wan pro le shown as a function of path length The various curves show the effect of different range sample spacings for a xed path length For reasonably accurate debased r e 39 estimation of rain rates in moderate to heavy rain need 6de 025o km1 or so For example at 150 m gate spacing you would grab sample from a path length of about 34 km in range 1 8 a m JAJA e 1 m Bringi and Chandrasekar 2001 Ozone isopleths HC NOX 100 300 Urban suburban 10 E 1 100 3 1 gao 3 3 ppb hr 05 Rural 0 3X 1 3 10 0 3 Z 3 1 01 E Remote tropical Net 03 loss forest 1 3 001 I I I I I 01 1 10 100 1000 OH reactivity adjusted VOC ppr FIGURE 1638 Observed NOX and OHreactivity adjustcd VOC expressed as propene in various regions of the troposphere Isopleths shown are midday rates of 03 production ppb hquot calculated using a box model adapted from Chameides et al 1992 FinlaysonPitts and Pitts AP 2000 NOz formation RH m OH Ro2 0 1m Ho2 R0 NO H01lt 02 R02 0 N02 Figure 2 Tropmpheric 0 production mechanism com pum uruuupicu catalytic HO and Nox cycles 0 average one loop around he HOX cycle drive the NOx cycle twice producing two new 0 molecules Formation ol an alkyl nitrate RONOE terminates 3 production by removing one NOX and one HOx molecule 39om each cycle Rosen et al JGR 2004 Ozone production ef ciency How often is NO2 converted into ozone before it is oxidized to NOy 03 NO2 MIXING RATIO 0 2 4 6 8 1O 12 14 NOy NOX MIXING RATlO ppbv 16 18 20 nm 9351mm Figure 734 The relation between 03 N02 versus NOy NOX from a ground site in Schauinsiand Germany Voiz Thomas et al 1993 The slepe of this plot is interpreted approx imater as the ozone production e ciency the number of 03 moiecuies produCed for each N OX molecule emitted or oxidized to reservoir components In these studies the N02 mixing ratios were a signi cant fraction of the 03 mixing ratios and thus their sum is used as the ordinate scale Brasseur Orlando Tyndall CUP 1999 Ozone production ef ciency Houston 220 200 V 180 g 160 D D xquot O 140 120 7 y428 0x 51 098826 100 y5595x F o9896 6765x moss 321 80d I y I I lt 5 1o 15 20 25 N01 ppbv Daum et al JGR 2004 Ox ppbv Ozone production ef ciency as function of VOC 200 180 160 140 120 100 80 60 40 81210x If n 9m 0x 30 157chol 2 a l 8IZ9 Ox H2 098 64 16 OIHCHO 42 54chol HCHO ppbv Daum et al JGR 2004 901 Nighttime chemistry O3NO gt 02N02 ozone titration in dense urban areas NO titration in rural and remote areas a 03 N02 gt 02 NO3 b 03 alkenes gt carbonyls OH slow No3 No2 M N205 M slow N205 H20 2 HNo3 slow HNO3 HZOsurfaces gt nitrate NO3 hv gt NO 02 or NO2 03P morning 2500 0 2000 Q m m D 79 1500 Z 3 5 O E It NE 1000 6 7 a O 2 500 H 9 o x n 600 620 640 660 70 mm FIGURE 416 Absorption spectrum of NO3 at 298 K adapted from DeMore e1 IL 1997 based on data from Ravishankara and Mauklin 1986 Sander 1986 and Canosa Mas et al 1987 FinlaysonPitts and Pitts AP 2000 NO3 photolysis N02 O Fluorescence 04 Quantum Yield I l I I l quotquotquot 590 600 610 620 630 640 7L nm FIGURE 418 Quantum yields for N03 photolysis dotted line NO3 gt N02 0 solid line N03 gt NO 02 dashed line uo rescence quantum yields adapted from Johnston m ul 1996 FinlaysonPitts and Pitts AP 2000 Term symbols for molecules Syntax 281ML ML 2 ml S 2 3 m O o from s or pZ orbitals m i1 11 from px pyorbitals Simplified by the selection rules no electron with same quantum number setll Spectroscopy AbsorptionEmission of energy in the form of a waveparticle duality happens at UVvis and lRwavelengths most often involves the ground state Boltzman distribution Drives Chemistry in the Atmosphere solar input is converted to higher chemical potential entropy gt radicals higher kinetic energy Some ground rules an electronic transition describes a change of the electron configuration of an atom or molecule usually cause by UV vis light transitions usually occur between 80 and S1 a rovibrational transition describes a change of the quantum mechanical sub states of rotation or vibration usually caused by visible or lRlight Boltzman t TmnslatiOns Rotations Schwingungs Etektmnische I I I d b O n Amegung Anreguug Anregung Anregung hi c d E 12 3091 9 iiig L 2st sag 2 10 E m 5 m 4 mm D mm m mmcc m c a Rotation b Vibration 0 Electron excitation from Atkinson Physical Chemistry Solar Spectrum Absorbing molecule CO co CO CO CO H O 2 2 2 2 2 I ZHL H20 1 03 H20 03 H20 1 m i 1 Jul IL 100 I I I l I I I E 80 E 8 60 5 E 40 I C 3 20 0 I I l I I I I I I I I I I I O 1 2 3 4 5 6 7 8 9 1D 11 12 13 14 15 Wavelength11m 39ddle I a r I near mI I 8 Infrared Infrared 39 Infrared from Boeker and van Grondelle Environmental Physics Oxyqen electronic states b 2M5 L L L 4 4L 32 Ag 2 FIGURE 45 a Molecular orbital diagram of ground X322 state of 02 1 Comparison of highest occupied prgquot M0 for the ground state X32 and the electronically excited NAB and b1 2 states 3 from FinlaysonPitts and Pitts Chemistry of the upper and lower Atmosphere Oxvqen electronic states v 4 iii 03P OPP 579 tnergy KJ FIG Schu 1955 URE 41 Potential energy curves for ground and rst four ed slzlte of 02 SeR SchumanneRungc system H berg conhuuum AeA atmospheric bands adapted from Gay 1968 from FinlaysonPitts and Pitts Chemistry of the upper and lower Atmosphere Quantum numbers Main quantum number n M n 1 2 3 equals the period or row in the periodic table orbital quantum number I M O 1 2 n1 angular momentum of electron M I O gt sorbital I 1 gt porbital magnetic quantum number m M m l l1 l1 l not all p orbitals are created equal spin quantum number s M s 12 12 up T and down i IA Periodic Table of the Elements WM 1 N Iorglc 2 urn El H 1 He WW II A T3 quotI A IV A V A VIA VII A 40026 3 4 mm 4 5 s 7 s 9 10 Li Be 399 B C 0 F Ne 694 90m 31121 J lU8ll 12ml 14007 5999 5993 mm a 11 12 13 14 IS 16 17 18 Na M VIII I3 AI Si P S CI Ar 22990 24105 III B IV E V B VI B VII B I B II B 26902 28086 30974 32066 35453 3994 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr 39wa mmx 44956 47867 50942 51996 54938 52145 5893 5869 63546 6539 6972 72m 74922 7896 79004 xmu 37 38 39 40 41 42 43 44 45 46 47 48 49 so 51 52 53 54 Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te Xe 85 465 8762 83906 1224 92906 9594 98 H117 12906 111642 l787 1241 4112 1187 12176 Z762 Il 90 13119 55 56 57 72 73 74 75 76 77 7a 79 30 a1 82 83 84 as 36 Cs Ba iLa Hf Ta W Re 05 lr Pt Au Hg Tl Pb Bi Po At Rn I319 11733 I333 I lXAQ 18095 1384 IX62 190 21 WIZZ 19508 19697 211059 211438 2072 20893 2119 2 III 222 87 as 99 104 105 Fr Ra Ac Rf Ha I223 I225 227 I261 I252 electron configuration is commonly listed in periodic tables Syntax IIIquot with x of electrons Carbon1s22s22p2SuWur3s23p4 Homework IZI write down the electron configurations of N 0 Cl E why do halogens X form X2 in the gas phase B why do the alkali metals Li Na do so too Selection rules Pauli exclusion Principle No two electrons in an atom or molecule can have the exact same quantum numbers Hund s Rules Orbitals with the same potential energy get singly occupied first Angular momentum conservation is valid for electrons in atoms and molecules The photon has a spin ofl RusselSaunders Terms RusselSaunders Terminology describes the ground state orbital and total electron spin via 281 L spin multiplicity orbital angular momentum S 23 L max2m Hund s 1St rule the ground state is the state with the maximum spin multiplicity Hund s 2nOI rule For the same spin multiplicity the state with higher L is more stable Examples C gt 3P O gt 3P MOTheorv Molecular Orbital MO Theory is an easy to apply method of creating molecular orbitals that helps us to characterizeunderstand simple multi atom molecules uses linear combination of atomic orbitals LCAO x atomic orbitals combine to x molecular orbitals is mathematically imprecise but chemically highly successful and graphically simple because intuitive Syntax radially symmetric molecular orbitals are named a sigma orbitals rotationally symmetric molecular orbitals are named 11 pi orbitals a b stands for bonding a for antibonding H H nonbonding molecular orbitals are named n LCAO of s and p orbitals a Uberlappung von 25Valenzorbitalen Abb 212 Uberlappung von zwei ZpXOrbitalen in AzMolek len LCAO of s and p orbitals b x a 5 und wry sind 151 und 1 Equivalent Abb 214 a und 39 Mola 39 kiil aus s und p Valenzelektmnen gebildet werden Reminder Rate laws the elementary reaction aA b8 3 cC dB is described by Rate k A61 Bb k CC Dd the exponent a b gives the order of reaction for each species and the total for the reaction example is first order in 02 and second order in NO Composition of Earth s atmosphere Carbon dioxide 002 Helium He 00005 Krypton plus highly variable amounts of water vapor Oxygen and Ozone Photochemistry 1 02 hv gt O3P O1D Alt24O nm 2 O3 hv gt O2 O1 D Alt325 nm 3a O1D M OZN2Ar gt O3P M 4 O3POZM gt03M a termolecular reaction 3b O1D HZO gt 2 OH Convention Only radical molecules will carry a for clarification exceptions Oz NO N02 Bi and Termolecular Reaction Rate Constants Collisions between two molecules are more likely than 3 molecule collisions so 2molecule reactions are generally faster example NONOO2 gt2NO2 is VERY slow bimolecular reaction rate constants depend strongly on temperature but not pressure termolecular reaction rate constants depend on temperature AND pressure hence should proceed faster with decreasing elevation termolecular reactions can be described by a pseudo bimolecular reaction rate constant for comparisons Termolecular reaction rate properties Reaction rate of the 802 OH M reaction 500 400 402 Torr 202 Torr 10 300 200 Decay rate 5 100 04 08 1 2 10 15SOZ0 molecules cm a Sgt IRE 51 Plots of the OH decay rates against the initial 502 f nLCntrzition at tntai pressures of Ar from 50 to 402 Tom adapted rum Atkinson at 51 1976 Termolecular reaction rate properties 03 1013kbicm3molecue 1s 1 N 0 200 400 800 800 Total pressure of argon Torr FIGURE 52 Plot of k bl against total pressure for M Ar for the reaction of OH with SO2 adapted from Atkinson et 4 1976 How do termolecular reactions proceed twostep mechanism 1 A B gt AB 2a AB M gt AB 2b AB gt A B Rate for intermediate species AB dABdt k1 AiiB kga M lob AB 0 gt AB k1 AiiB k2a M k2b Resubstitution into rate for 2a diABildt k1k2aiMilltk2b k2aM AiiB with pseudo first order k k1k26Mk2b kZaM Lifetimes SS Fate of OH Lifetimes are defined as the efolding time of a species concentration over its loss rate in the atmosphere Example CO OH gt H 002 dCOdt 39kCOOH CO39OH z 39k OH CO assuming OH is constant ie in Steady State 88 Lifetime 11 CO kOHOHCO 1 I kOHOH The lifetimes of most molecules in the atmosphere are determined by their kOH 88 can be assumed for all shortlived species ie essentially all radicals 88 production rate loss rate OH production and loss Main production pathway in 90 of the atmosphere 3b ooo H20 2 OH Intensity arbitrary units 30800 30805 30810 30815 wavelengthm Main loss pathways in 95 of may the atmosphere 4 00 OH gtHC02 Emma 1g CH4 H20 5 511 5 1390 1392 1394 1396 18 gt knowledge of OH CO and Time of day UT CH4 sources and sinks in the FIGURE 42 a Folded long path differential absorption spectra of the OH radical in ground I I I level air A original spectrum B After subtraction of absorption spectra of interfering O S p h e I 39 39 compounds mainly HCHO 502 CwHS The thick line is a superimposed OlI reference spectrum C Residual after subtraction of all known trace species b 39 nal variation of OH Diur concentration oints with 1 sigma error bars superimposed on the ozone photolysis frequency line to produce 01D Adapted from and courtesy of Dorn at all 1996 ATMO 689 Lecture 10 111604 Polarimetric Radar Properties of Hai BC Ch 72 Hailstone properties Size shape composition fall mode Hail Signatures in Differential Reflectivity Zdr Rain line in ZhZdr space Hdr Linear Depolarization Ratio LDR Correlation Coefficient pHV Specific Differential Phase de Some slides adapted with permission from W A Petersen UAH Hailstone Properties Size and Shape r 395 m 5 Hailstones range in size from Deq H 5 mm to 55 mm and larger v p quot Record is now 7 178 cm over Aurora NE on 22 June 2003 Hailstones tend to have oblate spheroidal shapes but can also have very irregular shapes eg lumps and protuberances depending on growth mode Hubbert et al 1998 Plate 7 Hailstones collected during luzulstorms In Switzerland From Levi at n1 197011 by courtesy of Amer Matter 5012 and the authors Pruppacher and Klett 1997 Hailstone Properties Shape composition fall mode Oblate spheroidal with axis ratio 05 3 am 3 095 Axis ratio of hailstones Slight tendency for axis ratio to decrease with increasing hailstone 110 Oklahoma Hailstones size up to about b 3 cm 1400 Nquot N 1790 Hailstones are typically composed of solid ice with associated bulk density and dielectric given earlier Spongy ice is possible 5 I Waterice mixture in which air q OF E39E39 glgrado Hallstones inclusions within soft ice have N been replaced by water during wet growth Higher dielectric than solid ice Early studies suggested that hailstones fall with their major axis Ha lstc r e Max39mum Dimension mm Fig 741 Axis ratios nfha slones versus their longest dimensions The number ofhm lsmncs in each size interval and the 95 con dence interval are marked top for Oklahoma hailstones a motion N 1790 bottom I m39nonhcast Colorado hailstones N 2675 From Knight 1936 More recent studies found that hailstones s in around their minor p Bringl and Chandrasekar 2001 axis which are approximately horizontal but also wobble causing a precession Differential Reflectivity hail Zdr is the reflectivity weighted Zdr rain VS ha measure of the particle axis ratio On average hail is more spherical than raindrops We ve also seen that Zdr depends 2 on dielectric and fall mode 39 1 behavnor I f X Hail wobbles gt effective sphere Ice has Relatively low dielectric gt effective sphere Zdr is near zero with a tendency for larger hail to have negative Zdr Small hail 05 s Zdr s 05 dB Large Zdr lt 0 dB negative FIG 9 Summary of typical Zn values for raindrops of various sizes and hail These 20 values are based on the median volume V Zdr diameter from Bringi at a 1986a The black arrows on the bail particle represent the tumbling motions as it moves in a thundcrstorm Related to fall mOde mlnor aXIS Raindrop pictures courtesy ofCloud Physics Group at UCLA hail In horizontal AN DOR threebody mam comm of Nancy Knight at NCAR scattering in which negative Zdr comes from ground reflection H Wakimoto and Bringi 1988 W ab shh P L 9 h Modeled Zdr for vertically oriented hailstones V 39 H Sband Obate spheroid with major axis in the vertical Varied axis ratio ab 08 06 Varied waterice arrangement and hence dielectric dry wet spongy Varied size diameter up to 6 cm Mie scattering effects D 05 A 6 m l l I I I o N 4 1quot Dry ab08 p35 0 Wet abo8 3 CI Spongy ab08 3 2 A Spongy ab06 A Cgt0 o 9 g a 0500 A g E u gg 3 i g giiA 0 g 5 2 alumnus A a 99 E A a A 9 4 A E A A 5 6 l I l 1 l 0 10 60 Hall Diameter mm Fig 754 Calculations of Zdr at SVGHZ frequency versus hall diamctcr The hailstones arc assumed to bc oblate in shape with 1 xed axis ratio of 08 with orientation similar to Fig 744 except thcrc is no conical gyration Le Fig 90 it is uniform in 0 to 27 Spougy hail rcfcrs to dielectric constant with 40 water fraction From Balakrishnan and Zrnit39 1990b BC2001 from Balakrishnan and Zrnic 1990 Hail Hole Large Zh and near zero or negative Zdr A Zh ORIGIN 000 000KM XAXIS 900 DEG B I IIIIIIIIIII 70 54 81812 MDT I x 44 bm wmmumu wazumr dag 3 4 g 30 7 A I u I V g i IaoIIn 120130 ADISD 160170 180190 200 am 220 230mm I f Y KM g 3 4a a z I 1399 Zdr ORIGIN 000 a 0mm xrAXIs 900 DEG 1 I I I I I I I I I I I I 50 5 4 E x 44 N 34 n I 1 50 2 A 39 I en 9mm 1 J 45 a 25 5 7 5 39n n 4 100110 I20 130 140150 150170 180190 200 210 220 2301 Wirce m I Y K by Hubbert et al 1998 Fig 822 3 Vertical cmss section of re ectivity actor and b Zn in chtqu39shalftunes through I I II I I I II I II I I II II I II III II II I I IIII I I I I I 15 he core ol39 a Convective cell on 28 May 1983 Conloun or re ecumy sun at l0 dBZ in steps 0f lb 1312 MDT 39 i 10 dBZ differential re ecliv 39 contours start at 1 dB Note the depression of 1th DR eld in he ZmhlI mum m I 4 5m quotMEG 555 I 11 regiun of high re ectivin 18 2 l9 km indicaling hail from BnngI kl IIL I986 in d Inn in mmng nag1013 413 DOVIak and 2mm 1993 adapted 3 g 39 39 0 1 From BrIngI et al 1986 lt 00 E 39 E 25 a 15 10 WW I f 1 5 fin 399 20 mi r 0quot 25 Q mm Fla 6 I 131 MDT 4 3 shown In fig at up new gnd ongm IS Herzegh and Jameson 1992 g1LImdg Ig pa 9 Huh slwamzes are walked 3 39In1lIn11ranH InIs mg lb 1b And d The IIIcIaIIII tampemuxe ml 15 Is In PIE 5 Cemun mlm mg feamm III IocIIIw quotaquot Ibraugh IIIIIch me IemIed III Table I The xquot Iccam III peak ZI While 11 I quot HDR Aydin et al 1986 The rainline idea Similar to using ZDP to delineate regions of rain and ice eg recall fice Combine Zh and ZDR pairs into scatter plot ZDR on horizontal axis and ZH on vertical axis ZHldEZl Dots are disdrometermeasured ZH and ZDR Sband assumed Lines are modeled rainlines with differerent drop axis ratio assumptions Define HDR as HDR Zh fZDR dB And 27 ZDR s 0 dB aZDR 270 5 ZDR s b dB 60 ZDR gt 174 dB fZDR a b are functions of axis ratio model used in scattering calculations to obtain rain lines from disdrometer DSD data as input Equilibrium axis ratio model a 165 b2 dB Oscillations included a z 19 b z 17 HDR 2 10 dB or so indicates hail conservatively Potential ambiguity if wet eg melting hail is oblate enough to yield a ZDR similar to raindrops HDR and rain line rain mixed with hail L Rain line Hdr is essentially the highest expected value of horizontal reflectivity Zh dBZ for a given value of differential reflectivity Zdr dB in rain ZhldBZ Notice that some rain points have same Zh as hail but much larger Zdr value added information from Zdr 1 0 1 2 3 4 5 Z DR 15 Fig 821 Scattergram of 2 versus ZDR fur a rainfall on 10 June 1986 from the elevation of 1 The curve for an Nu 105 mm 39 m 3 is plotted Stemhorn and Zmi t 1988 this the rain hail boundary proposed by Leitao and Watson 1984 and Z is lhe boundary rum Aydin 1 al 1986 Doviak and Zrnic 1993 as adapted from Aydin et al 1986 HDR as a function of hail size Brandes and Vivekanandan 1998 found a weak positive correlation between maximum hail diameter and HDR Results for Sband By eye fit scatter is large so we should look elsewhere for additional and hopefully better information about hail size 802001 as adapted from Brandes and Vivekanandan 1998 40 on CD I 20 July 1993 o 13JMy1993 Hail Signal Hdr dB l O O l I O 0 n l l i I O l O 20 30 4O 50 Maximum Hail Diameter mm Fig 752 Radarbased values of Hdr correlated with surface observations of maximum hail diameters in two hailstorms in northern Colorado Radar observations were made at 3 GHz frequency with the NCAR CP 2 radar From Brandes and Vivekanandan 998 Linear Depolarization Ratio in Hail Brief Review Recall from the Covariance Matrix LDR is written as LDR 10Log LDR is similar to ZDR in terms of its mathematical geometric dependence on particle shape increases with decreased axis ratio its reflectivity weighting and dependence on particle phase dielectric strength 2 Orientation dependence is a bit different though For measurable LDR particles must be canting wobbling tumbling andor irregularly shaped so that some fraction of the incident horizontally polarized wave is depolarized upon backscatter lsvhl2 lshhl2 Send out a horizontally polarized wave and listen for a depolarized return in cross pol channel eg Svh in addition to copol returned power eg Shh about zaxis Canting or tumbling zaxis Svh l2 gt O IShh l2 gt 0 LDR Finite lt 0 LDR in Large Hail LDR examples In hailstorms STEPS 29 June 2000 a HID 2H 420 km Region of high Zh gt 60 dBZ elevated LDR 26ltLDRlt21 dB aloft indicates tumblingwobbling hail in wet growth Note large melting hail signatures 2 3 Altitude km MEL dBZ colors EDR conto red near ground b H y 420 km u Large Zh Zh gt 60 dBZ Zdr hole Zdr lt 0 dB 21 dB lt LDR lt 18 dB Note LDR lt 26 dB in rain I Consistentwith output from NCAR 0 a 1 Altitude km MSL allthZey tlacifgtrcgmhyd rometeor ID output LDR Cglgglg z gggggured LH Large hail LDR tends to increase with hail size SH Small hail LHr Large hail and rain mixed m 30 NCAR Hydrometeor ID Courtesy K Weins CSU LDR example in hailstorms Fla 5 As in Fig 3 except tor a severe hailstorm observed over Denver Colorado on 13 June 1 984 Color scales at bottom of a b and cl indlcaie values of 4 2m and LD Fl respectively a Z dBZ showing the preclpltatton core at 224 range in 2D 18 showing a nail signature at the surface at 22km range ml 3 region of supercooled water drops in the storm In ow aloft mm 23 13 29km range 2 LDR dB showing significant propagation effects through the rain near the surface and beyond the precipitation core aloft Recall Herzegh and Jameson 1992 Recall hail hole as surface indication of hail However Zdr gives little or no information aloft Enhanced LDR 20 dB lt LDR lt 15 dB aloft colocated with elevated Zh gt 50 dBZ is an indication of wet growth hail Even larger LDR 20 dB lt LDR lt 12 dB near surface with near zero Zdr and elevated Zh are indications of melting hail See also Hubbert et al 1998 which was reviewed in class for excellent LDR hail signatures Correlation Coefficient Hail Correlation Coefficient pHV can be lowered lt 098 in some situations associated with hail See Balakrishnan and Zrnic 1990 13 15251540 1 Melting Hail mixture of heavy rain and hail will lower pHV 2 Very Large Hail say D5 cm at Sband can cause minima in pHV that are associated with Mie scattering CORRELATIDN COEFFICIENT pmm 1 55 Hail and Rain Mixture HEPLECTIVITY FACTOR 2quot dB 80 65 70 39LUU CORRELATION COEFFICIENT pmm I I Jr 14 09 quot39quotww18 39 a quot39 1 03 I u quot Rain 175 mm nquot and Wet Hall I I I l 40 20 an HAIL arr m hquot 2 Mie effects in Iar hail o l V 50 BO l g V I Annual 3 O S A E C 095 A a l A AA 090 I D A Dry aha05 A 035 0 Wet abO S g Spongy aibDB A Spongy abOB 0 50 J l l I D 10 20 30 40 50 60 HAIIL DIAMETER mm Correlation Coefficient Hail 3 Irregular shape lumpiness 3 Irregular eg lumpy m shape of hail as the lumpiness or protuberance to diameter ratio increases 048 the pHV decreases o 05 39 CORRELATION COEFFlClENT p 0 0 01 02 03 PROTUBEFIANCE T0 DMMETEH RATIO uDID FIG 7 Correlation coef cient for Rayleigh scatterers with protuberancca Specific Differential Phase Hail The differential propagation phase dp and specific differential phase de are immune to the presence of hail Notice in the figure to the right that de is only a function of rain rate and is essentially independent of the hail rate ie de are nearly vertical lines Because de 0 degkm in hail Could look for de holes near surface to differentiate heavy rain and hail This property makes de a desirable parameter for estimating rain rates in the presence of hail more later 1000 HAIL RATE Rh mm n 391 45 D ID 100 1000 RAIN RATE H mm h Relationship between rain rate Rr hail rate Rh horizontal reflectivity Zh and specific differential phase de From Fig4b in Balakrishnan and Zrnic 1990b JAS 47565583 Polarimetric Measurements in the Hail Cascade of a Supercell put it all together 49 47 4539 r 1l 43 Jl Distance le of C i39lLL radar km quotml 7 2amp2 In M 40 SI 0 5 1 D83 Figure 4 East west vertical cross section of CSUCHILL hnriznmal re ectivity 132 at y v 23905 km see Figure 3d Note the forward overhang and bounded weakecho region BWER lmrnx 2 7 km tox quot3 km The path uflhe hail cascade which is analyzed in Figures 5 5 is marked by crosses 3 25 2 15 1 05 8486889092949698100 K pHV dp Spoclflc Dl amnllll Phase Dormlllian Gnalfleianl 1 1 12 1 08 06 CI4 02 0 Dl nnntlal Reflectivity dB ZGlr LDR Figure 5 quotVertical pro le of CSUC IIILL mulliparamelcr radar variables through the hail cascade depicted 139 n Figure 4 The melting level is at 24 km agl Ila The correlation coef cient mm and differential re ectivity 2d dB b The speci c differential phase KW km and linear depolar ization ratio LDRI dB 05 728 726 3924 22 20 18 16 Unur Depolurlutlon Hula dB 14 Carey and Rutledge 1998 or see BCZOO1 Figs 756 and 757 PM removal ratesaerosol lifetime Coagulation for ultrafine particles Wet deposition for fine and coarse PM well studied depends on rain rate and elevation in troposphere higher removal rates in mid compared to low latitudes due to stronger updrafts in the tropics dominant for CCN in the range of 0110 pm Dry deposition less well studied sedimentation for larger particles gt10 um impaction for ultrafine particles PM removal ratesaerosol lifetime hours 39 39 39 39 duys a coagulation wet precipitation sedimentation l0 10 1 0l lt 001 r p t 10 3 10392 toquot 1 IO 102 RADIUS um FIGURE 731 Combined residence lifetimes of aerosol particles as a function of size Adapted from Jaenicke 1978c 1980 Important removal processes active in various size ranges are indicated Coagulation and sedimentation time constants were calculated the time constant for Wet removal 15 the residence time derived from the 210Bi2 DPb and 222Rn 210Pb ratios Martell and Moore 1974 Curves l and 2 represent the background for 7th equal to 12 and 3 days respectively Curve 3 represents the continental aerosol with 7W 6 days The dashed line was calculated or sedimentation equilibrium as described in Section 761 Warneck AP 1999 PM removal ratesaerosol lifetime q s 1 0quot3 1 0394 DEPOSITION VELOCITY I m S I I I 01 1 1 O PARTICLE RADIUS urn FIGURE 73 Dry deposnion velocmss for particlcs as a Funcliun or size Left Deposition m grass solid syrnhols and wzlcr surface Open symbols data are from 39hambe ain 1953 96 Mb llzr and Schumann I970 Sehmel and Suller 1974 Clough 1975 Litt39Ie and I7 and Wesley Pl c117 1977 Wind speed triangles z 2 m s I Circles 3 111 s 1 diamonds 1 m s 1 ight Deposition L0 oresls Inainly sprucc adapted from Gallagher a and references thereln Diamonds rf gtr to cluud and fog drops The solid lines in both graphs indicate the scdirnentauan Veloclty E A s u o Warneck AP 1999 Atmospheric Aerosols Summarv Direct and indirect effects on Earth s radiation budget igt Climate Scattering UVvis absorption Cloud condensation nuclei CCN formation Of natural and anthropogenic origin Complex multiphase multicomponent matrix including inorganics and organics Major analytical challenge Major international research effort H 1 Chemical Conversion of Gases to Low ap r VolaIiIin Vapors Condensation Homogeneous Nucleatmn Wind Blown Dust Emissions Sea Spray Plant Panicles I I l l l l I I I I I I I I I Volcanos l I I I I I I I I l Sedi mentaiion I I l l J 0002 001 2 10 100 0 1 l PARTICLE DIAMETER pm Fllfgii 39d cliiia iLZEHl39i GEL Fine Parlicles gt Mechanically Generalte Aerosol Flange Coarse Particles Brasseur Orlando Tyndall CUP 2000 Atmospheric particle sizes Figure 51 Ranges of equivalent diameters for some types of aerosol and hydrosol particles For perspec tive the diameters of molecules are also shown Particle Radius L7 nm 0 4 c l W Molecules L 0le W Metallurgical fume Colloids W m Viruses Combustion nuclei Accum mode particles Bacteria Windblown dust Sea salt Algae VIIIIIIIIIIA Pollen VIIIIIIA Graedel and Crutzen WH Freeman amp Co 1994 Table 1 Characteristic aerosol data for urban rural and high alpine air in central Europelal Urban Rural Alpine Munich Hohenpeissenberg Zugspitze PM25pgm 3 20i10 10i5 4i2 TC in PM25 40i20 30110 ZOilO ECinTC SOiZO 30110 30in OCinTC 40i20 70110 70i10 XSOC in TC ZOi 10 40i 20 60i20 MWSOC in XSOC 30i10 SOi 20 40i20 a Rounded arithmetic mean valuesistandard deviation determined from about 30 lter samples collected at each location during 20017 2003 Poachl Angew Chem Int Ed 2005 Table 2 Prominent organic aerosol components Substance Classes Proportions Sources aliphatic hydrocarbons 10 2 biomass fossilfuel combustion aliphatic alcohols and carbonyls IO392 biomass SOAaging levo lucosan 1039I biomass burnin polycyclic aromatic hydrocarbons PAHs 10 ossil uel com ustion biomass burning nitro and oxyPAHs 39IO 3 fossilfuel combustion biomass burning SOAaging proteins and other amino compounds 10quot biomass cellulose and other carbohydrates 39IOquotz biomass secondary organic oligomerspolymers and 10quot SOAaging soildust humiclike substances a Characteristic magnitudes of the mass proportion in fine 0PM Poschl Angew Chem Int Ed 2005 CH oxidation CH4 OH gt CH3 H20 CH3 02 M gt CH302 M CH302 NO gt CH30 NO2 CH30 02 gt HOz HCHO H02 NO gt N02 OH 2x NO2 hv gt 03P NO 2x 03P02M gtO3M CH4402 gtHCHO203H20 Net production of ozone The net reaction does not occur as such in the atmosphere CH oxidation CH4 OH gt CH3 H20 CH3 02 M gt CH302 M CH302 H02 M CH302H 02 M CH302H hv gt CH3O OH CH3O 02 H02 HCHO CH4 02 HCHO H20 No net production or loss of ozone Net loss of radicals when CH302H H20 gt deposition Peroxide Photolvsis 70 1 11 1 a 60 E e 0398 T 50 30 05 w 8 40 04 H 02 g 02 CH300H E 30 n u i i i NE 290 300 310 320 330 340 350 350 3 20 CD I g 10 CH3OOH I 1 4 i 0 190 210 230 250 270 290 310 330 350 9 nm FIGURE 425 Absorption spectra of HZOZ and CHSOOH at room temperature data for H202 from DeMore at 111 1997 recommendation and for CH3OOH from Vaghjiani and Ravishankara 1989 Pitts amp FinlaysonPitts 1999 Lifetime of days Significant removal via wet deposition possible CH4 oxidation CH4 OH gt CH3 H20 CH3 02 M gt CH302 M CH302 NO gt CH30 No2 CH30 02 gt H02 HCHO 1 HCHO OH gt HCO H20 2a HCHO hv gt HCO 2b HCHO hv gt co 3 HCO 02 gt H02 co 4 H02 NO gt N02 OH Photolysis in the UVA Branching ratio for 1 vs 2 50 HCHO photolysis 10 0 8 e m 0 0 To 3 6 E13 3 TABLE 425 Recommended Quantum Yields E for Photolysis of HCHO m 4 E Wavelength 8 11111 H HCO H2 C0 b o 2 7 240 027 049 393 250 029 0 49 260 030 0 49 270 038 0 43 280 057 0 32 0 290 073 024 l l 1 I I l l 300 078 021 240 260 280 300 320 340 360 301325 0749 03251 303 75 0753 0247 30625 0753 0247 K 1 30875 0748 0252 31125 0739 0261 TIGURE 426 Absorption spectrum of HCHO at room tempera ure adapted from Rogers 1990 31875 0623 0368 321 25 0559 0 423 323 75 0492 0 480 326 25 0420 0 550 328 75 0343 0 634 33125 0259 0697 33375 0168 0739 N a o n 33625 0093 0728 33875 0033 0667 34125 0003 0602 34375 0001 0535 34625 0 0469 34875 0 0405 a 35125 0 0337 2 35375 0 0265 CO 2 35625 0 0197 2 IUPAC recommendations from 240 to 300 mm H C H O 2 O Atkinson at al 1997a and NASA recommendations 2 from 301 to 356 nm DeMore 61 IL 1997 where the latter are for 215 Am intervals centered on the indi CO O H O Gated wavelength based on Horowitz and Calvert Int 3 2 J Chem Kinet 10 805 1978 Moortgat and War neck J Chem Phyr 70 3639 1979 and Moortgat er al ibid 78 1185 1983 Formaldehyde model prediction versus measurements HCHO 0 0 P N w 1 MIXING RATIO I nmol moll 0 DEGREES LATlTUDE FIGURE 48 Distribution wnli lautudc of formaldehyde over the Atlantic Ocean Vertical bars indicate worsigma variances Adapted from Lowe and Schmidt 1983 The dashed curve shows the results of a two dimensional model calculation Dcrwcnt 1982 Warneck 1999 Formaldehyde importance toxicity source of HOX source of ozone always present Tropospheric Sinks of OH N02 3 CH4 00 am V quot H O 9 CO 81 Clean Troposphere C0 CH4 gt 95 Tg CO 10 degrees1 yr1 Tg 3010 degreesv1 yrr Atmospheric CO budget Latitude degrees CO sources E Biomass burn E Industrial Z Horiz influx l Oceanic NMHC Veget NMHC 1 CH4 oxidization CO losses 53 Surface depos B CO oxidization I Horiz outflux Figure 1412 Annual carbon monoxide budget estimate as a function of latitude The source terms are de ned as BinIan 71m CO emissions from biomass but ning in the tropics Didimm CO emissions From fossil fuel use and industrial processes Harizi In ux net gain ol CO in the atmospheric columns extending to 23 km 215 a consequence of horizontal trans port Oceanic NMHC oceanic emise sions of CD plus CO produced in the boundary layer over the oceans through photooxidation of marine nonmethane hydrocarbons Vegan NMHC vegetative emissions of CO plus CO produced in the boune tlary layer overland through photo oxidation of nonmethune hydrocate bons emitted by vegetation MelJane racialHimgt production of CO by the atmospheric oxidation of methane The sink terms are de ned as Smyizre depot surface deposition of CO CO oxidation loss oFCO through oxidation by radi cals HO H m39iz 0 IX net loss of CO in the atmosphcric columns extending to 25 km as a conse quence of horizontal transport Courtesy ofK M Valentin Max PlanckInstitut fL tr Chemie 1990 Graedel amp Crutzen 1993 Methane in the atmosphere M ethane GHQ Concentration we Tim e Con centratiun pprm 99921993 59139 59 97 9 29 911982 1 1 441934 9 1993 5N9 999 19 19993 9999995 1392 411 999 441 99991 1 459994 199 99999 Ti ne quotmuum Methane CH 9 Ceneentratiene we Latitude A m quota in to A quotm Cnncatra m mprnu a 4 Ln 39L I 4 D D 59 9 59 199 Latitude minus South httpCdiacornlgovbynewbysubjechtmatmospheric The Ozone Hole Total DU Plate 14 Evolution hem39een 1979 and 1997 of we mm 020119 mlumu abundance March RITE ng measured by the Tami 0243119 fappiug Spectrometer TOMS Dobson Units The gurc highlights the substanth reduction in Arctic 0mm during the 1990 s Courtesy r1 P Newman 139 TASAGodd wd Space Flight Canter Stratospheric ozone hole Chemistry E 1200 a23Augus 3 3000 b18eptember Q D V 3 g Q lt 392 393 n g 600 g 2000 S R E S Q m o O o 1000 62 64 66 68 70 72 62 e4 66 68 7o 72 as emanmsmm Figure 810 Arlensnreinents of highly enhanced CIO concentrations in the Antarctic Polar Vbrtex in 1987 The destruction of ozone during the period Aug 23Sept 16 is evident as is the anticorrclation of 03 and CIO concentrations adapted With permission from Anderson J D W Toohey and W H Brnne 1991 Bee radicals Within the Antarctic vortex The role of CFCs in Antarctic ozone loss Science 251 39 1991 American Association for the Advancement of Science Stratospheric ozone hole chemistry September 22 1987 N0y MIXING RATIO ppr Brasseur Orlando Tyndall CUP 1999 a A AUJdd 03H m m Aqud 90 O E 12 53 Zg 08 E S E 04 Q 0 0 an 55 50 64 A63 72 mm 1241 95 am LATITUDE Degrees Figure 714 The variation 01503 H2 0 010 and NO as the ER 2 aircraft ew across the Antarctic vortex boundary in September of 1987 Within the vortex at latitudes south of N764 S the region is perturbed chemically and a sharp increase in 010 the active constituent responsible for the observed 03 reduction is observed based on data from Fahey et 31 1989 Stratogpheric ozone hole Chemistry I Hm he Figure 86 Inorganic Chemistry involved in the interconversion of reactions that occur on H20 hat partwles 01 E 9 NO hv O CIO BrO Ohv hv OH use ve I am Stratospheric ozone hole chemistry Heterogeneous surface reactions Incorporation of HN03H20 with time Denitrification gt Dehumidification CIONO2 gt liquidsolid particles HCI gt liquidsolid particles CIONO2 HCI gt CIZT HNO3 CIONO2 H20 gt HOCI HNO3 HOCI HCI gt CIZT H20 HOBr HCI gt Brle H20 Temperature Sulfuric acid aerosols from 802 and COS oxidation Polar Stratospheric Clouds PSCS solution 9 110 cm3 Typewsc SAT Sulfuric acid tetra hydrate I Sggggggs NAT nltrIC aCId trihyd rate 188 K N gym 3610th Type 11 P80 5 3 10 3 10 2 Cm FinlaysonPitts and Pitts AP 2000 FIGURE 1221 Schematic of polar stratospheric Cloud PSC formation m probabilities on liguidsolid surfaces 4 I 1 90 1 95 200 205 Reaction probability y increase with decreasing T faster on liquid or ice than solid SATNAT surfaces FIGURE 1225 Typical measured reaction probabilities for a CION02 HCl b ClONOZ H20 and C HOCI HCl for dif ferent surfaces that can be present and promote heterogeneous chemistry under typical stratospheric conditions adapted from Rav ishankara and Hanson 1996 and references therein Dominatinq ozone loss cycles for polar winter chemistry need sunlight quotCIO dimer cyclequot CllO3 gt CIO 02 10 gt ClOJrO2 CIO 39lM 39ltXK I l ClOOClhv gt ClOOvC ClOOl M a Cll OzM Net 03O3 gt32 All cycles depend on CIOX and sunlight bromine cycle a Br 03 a 1310 02 C103 CIO PO Bit L 10 Brt l BrCl hv a BrC1 100 M a Cl OZM Net 0303 gt302 Net 03O3 302 bromine cycle b Br03 a BIG 03 C1O3 C10Oz Bm I L loi Br Red quotrate limiting stepquot the reaction with the smallest rate or the quotbottleneck39 of the cycle Caution that does not tell us much about the dynamics 01 the cyde Eg under twmght conditions the C20 dimer cycle is surprismgly insensmve to KCWCIO but very sensmve to Jam shuts down during night due to a lack of CIO Stratospheric ozone hole chemistry summ In the light of the rising spring sun Formation ol polar Formation of CIOX E stratospheric clouds CI2 hv gt 2C HOCl hv gt OH Cl H20 QK Cl oa gt CIO 02 HNo3 Activation of Clx Catalytic IE HO ozone depletion CIONO2 OCi cod isolated Clo mo 0202 polasrggtex CIZO2 hv gt 2C 02 N205 Cl 03 410 02 Stratosphe e oSlihe e not Figure 1412 Schematic of the dynamical and Chemical processes leading to ozone deple gion within the Antarctic vortex Processes l and 2 occur during the polar night While processes 3 and 4 require the presence of sunlight Brasseur Orlando Tyndall CUP 1999 Antarctic ozone hole extent I Ozone layer temperature at Halley in 2005 2006 35 30 30 3 40 7 ul 25 7 quot E Aug 23 1993 50 7 5 20 E70ct12 1993 m E 5 2 eo i a lt 3 E 10 E 3 70 i 5 S 7 Z S O IiIIJ 80 Da yvalues 0 5 10 15 20 7 t 11daymean Historic extreme P03 mpa 90 i r r Historic mean FIGURE 1217 Vertical O3 pro le beforc August 23 and after October 12 development of the ozonc hole at the US 100 i i i i i i i AmundscnScotl Station South Pole in 1993 adapted from Hof i A S Oct Nov Dec Jan Feb Mar Apr May mann at 11 1994a 9 ep Antarctic ozone hole extent EPTOMS Version 8 Total Ozone for Sep 20 2005 a NUIIONOI momomo Dobson Units Dark Gray lt 100 and gt 500 DU svme m a we 2005 Southern Hemisphere Ozone Hole Area N Curran qur in mm 2005 2004 2003 9504 Mean 9504 Max 95704 N n Size million square km Antarctic ozone hole extent I N e I e I Average area of ozone hole are of Mom America A area of AmarcIIca M t avea ufuzune lt 220 DU avevage uvevSD days vemcaI IIne z mtntmum and maxtmum avea II I 1980 I 1985 I 1990 I 1995 I 2000 2005 OH production and loss 12 I Ih E I II I r II E D eventing a 59 8 Dr 48490 0 5 E a 5 4 in D 39 iquot of 2 g K i g D 33 I2 I I I 1 I 5 i0 0 0 g B 5 IVE o O a i 4 1539 g l mg a III 39 as I J I l I i i I I i r I I 02 04 US 08 TU 12 T4 15 18 ThneiUT FIGURE 1149 Diurnai variation of OH measured using MP I10 and DOAS I in a rurai area in Germany m the a 16th and in 17th of August 39i994 Adapted from IIofzumahauzs er aft 1998 Main production pathway in 90 of the atmosphere 3b 000 H20 2 OH Main loss pathways in 95 of the atmosphere 5 co OH H 002 6 CH4 OH gt CH3 H20 gt knowledge of OH CO and CH4 sources and sinks in the atmosphere is critical Other sources of OH HOE HOX is defined as the sum of OH and H02 H02 is formed in both CO and CH4 oxidation eginH02M gtH02M fast interconversion of HOX species occurs via 7 H02 03 gt OH 2 O2 8 OH 03 gt H02 02 9 H02 NO gt OH N02 10 every time OH reacts with a VOC or H2 Hence every time either a net OH or H02 radical is formed this is called a HOX source Homework Calculate the photolysis frequency of ozone in the lower troposphere near Houston TX for midday in late June and September Use the provided tables on the webpage Derive the root mean square velocity of moleculesparticles in the gas phase Hint Consider a fixed volume and the momentum change in a wall collision then combine with F ma and the ideal gas law Calculate the OH reaction driven lifetime of CO and CH4 in the boundary layer kco 144E13 1N24E19 kCH4 64E15 cm3 molec1 s1 calculate the OH radical lifetime as a result of reactions with co and CH4 CH4 18 ppm CO 100 ppb Next week atmospheric CO and methane radical chain oxidation mechanisms CO and methane distributions in the atmosphere and their budgets ozone production and loss in the troposphere atmospheric ozone distribution and budget Required readings ATMO 689 Lecture 3 092104 Radar Waves Polarization and Sca enng Electromagnetic Spectrum Electromagnetic Waves Brief Mathematical Description Polarization Backscattering Matrix Covariance Matrix Radar observables Several slides courtesy of W A Petersen of UAH Electromagnetic Spectrum c h c speed of light c3gtlt108 m s391 f frequency 7t wavelength For radar meteorology we are concerned with the microwave frequencies roughly 1100 GHz tooquot low HF lot lonMHz VHF lm p lon Radio UHF L l0cm 2 s z c 7 EIOGHI SHF 5 X 5 5 3 lg E IE1 5 KO u IOOGHz EHF lmm lTHz Fur 0lmm IR 39039 Infrared low Th r F s l TH 00 1 Near Um lR Vlstble lOOOTHz ulirnwalei alum k f Figure 23 The electromagnetic spectrum The diagram shows these parts of the electromagnetic spectrum which are important in remote sensing together wit the mnvencional names of the various regions of the spectrum The letters P L S etc used to denote parts oi the microwave spectrum are in common use in remote sensing being standard nomenclature among radar engineers in the United States Various terminologies are in use for the subdivisions of the infrared IR part of the spectrum That adopted here de nes the thermal hand as lying between a and 15 um since this region contains most of the power emitted by black bodies at terrestrial temperatu Stephens 1994 Electromagnetic Waves Spectrum 100 km 10km 1 km 100m 10m 1 m 10cm EF 4 VLF In LF N MF I AM HF gt TVFM VHF gt 3 kHz 30 kHz 300 kHz 3 MHz 30 MHz 300 MHz 3 GHz 30 GHz 300 GHz We are concerned with the microwave portion of the spectrum in Radar Meteorology Lband eg 30 cm Sband 10 cm Cband 5 cm Xband 3 cm Kuband 2 cm Kaband 86 cm W band 32 mm NOAA wind profilers some ATC radars Spol CSUCHILL Npol WSR88D WSR57 ASR9 ATC Cpol DLR MIT TOGA SMARTR WSR74C Xpol DOW EDOP ELDORA NCAAPB tail aircraft vvx avoidance many others TRMM PR GPMPRII dualfreq NOAA ETL GPMPRII dualfreq NASAJPL UMASS U Wyoming Precipitation Cloud Electromagnetic Waves Stephens Ch 2 DZ Ch 2 Electromagnetic waves are generated by oscillating ie time varying electric charges which in turn generate an oscillating electric field Through Maxwell s Equations we know that an oscillating Electric Field produces an accompanying oscillating magnetic field and hence proceeds to propagate outward from the original charge Figure 21 A schematic view of a time harmonic electromagnetic wave propagating along the Z axis The oscillating electric 6 and mag netic H elds are shown Note that the oscillations are in the xey plane and perpendicular to the direction of propagation Stephens 1994 Basic Review of Electromagnetic Waves DZ Ch 2 E EFq FL q1q2 q1 Electric Field Static Case 41T 0 r2 Electric Field E Electric Force F unit charge q Q2 r Field lines pomt from posmve to negative charge Flux of field through a unit area 5E Magnetic Field B A Requires electric current Biot3avart Law B 0 d s x moving charge and is J E 4 r2 Or changing electric flux eg Ampere s Law cJS B39d39s mol uosod Edt an oscillating electric field Oscillations in voltagecurrent produce oscillating B and E Part of energy is radiated as propagating oscillating electromagnetic wave in the direction of the Poynting Vector S 1uO E X B Wm2 rate of energy flow through a surface h a Faraday s Law for induction via magnetic flux Dm Faraday Ampere 5 Law tW of Maxwell s Eq governing 33 E39dg 39dwmdt EM wave propagation Changing B and E fields force EM wave propagation these are radiated bythrough wave guide and antenna feedhorn then focused by the radar antenna EM waves move at speed of light C 2998 x 108 ms recall cAf Electromagnetic Waves Ex E0XsinoatLlJ Ey EoysinoatLlJ x Y E0 amplitude of Efield in plane i LV phase of wave in plane i HORIZONTAL POLARIZATION When LlJX LlJy nTr nO123 then wave is Linearly polarized if E0y0 horizontally polarized EH if EOXO vertically polarized EV CIRCULAR AND ELLIPTICAL POLARIZATION When EOX Eoy but LlJX LlJy and LIJX LlJyTr2 right hand circularly polarized ERC E field rotates about propagation direction in a right handed fashion if deltaphase angle 39T2 left hand circular polarized othenvise eipticay polarized Scattering and the Backscattering Matrix DZ Ch 8 BC Ch 3 Define timevarying electric field E that is incident i on a particle E O Relt i em 1 cosatl9vquot where hv horizontal and vertical polarization 032ch angular Doppler shift frequency and e offset phase EZ cosat6 131 The scattered electric field E by a single particle is E 2 SE where S is the backscattering matrix A Since ES 2 E2113 Ej133 we can also express the scattered electric field for each polarization component since is m 395 quot39 fl9m 39 39 H Emilia J HI E tr Z 13971 FF lymphE jtwmIlHm J 1ij yvh jejamlifj Evian where we have assumed that E 1 and E O ie we assumed incident wave had horizontal polarization and we have summed over all particles m in the radar resolution volume Dropping joat term for convenience for now we get Elm E E W 1 marath i m Z 3th ym jej uhm in Keep in mind that we could have gone through the same exercise for incident vertical polarization Combining it all together and using the fact that hv vh for reciprocal media like precipitation we can define the backscatter BSA matrix as 513391 quotti 2 Wife Tile air if M in I Hm U MNquot I Edam 2 i39 I aim it A If we put it all together we have a relationship between the scattered electric field and the incident electric field in terms of the backscattering matrix SBSA b i E E Eh SBSAU Eh ejt V V Radar does not measure the electric field lnstead signal voltages V are processed in response to the backscattered electric field being intercepted by the antenna However voltage is proportional to the RHS of the above equation so Vh t E h VVU N SBSA t E The mean voltage is zero as you integrate over time So radar meteorologists use various secondorder moments of the voltage liij instead These voltage moments are related to scattering coefficients of the particles Without deriving here we merely provide those scattering coefficients in a 3x3 matrix called the covariance matrix The covariance matrix is a complete polarimetric characterization of the particles in the RRV 139 v What are we measuring The timeaveraged indicated by overbar backscatter POLARIZATION COVARIANCE MATRIX BC pg 134135 Wk yhhyhvejahh ahv mm 6 1qu Note the matrix is yhvmejahuahhgt 2 whvywejmmam syiinmetricrthere are Vvvrhhej quotquot gt lyuvyhuej quot hquotl Vii Egg12 2ured yhh w instantaneous scattering proportional to received voltage amplitude fn complex terms hh receive transmit horizontal vv receivetransmit vertical hvreceive horizontal transmit vertical reciprocity says hv vh just the crosspolar return vhreceive vertical transmit horizontal oct phase information Also commonly written as an Ensemble average of complex back IShh2 2ShhShv Shth v scattering terms 12ShvShh 2IShvl2 12Shvva Treat SXX Xy as proportional to the returnedmeasured signal voltage SV VShh 2SV VShV ISVVI2 with amplitude and phase information in each polarization And the terms of the covariance matrix yield PChO RCO 5fo 2ch mgy Rco Rgx ng P co copolar power received in the H channel eg Zh lt P copolar power received in the V channel eg Z 00 P crosspolar power eg LDR OX Rwyox correlation terms eg Rco or Rco used in computation of phv and the argRco contains phase information due to propagation and backscatter differential phase and the Doppler velocity complex conjugate From these we define 2 ZDR 10Log 1OLog S hh2 Differential Reflectivity dB 00 I vvl LDR 10Log Ph 10mg I Svhl Linear Depolarization Ratio dB P 00 IS HS shh R I W M Copolar correlation coefficient phv Ipwl And for the phase variables poo ISVVI ISth ej600 dp 2 2 12 IShh I lva I NOte39 cZDR will increase if Ipcol argpco po 500 de decreases pco can also be used as an overall indicator of Where noise in the system should be Pdp Differential phase Degrees Very Close to 1 39 ram ltde Propagation differential phase Degrees Sco backscatter differential phase Degrees Mie effect For Rayleigh conditions 8CD is small and the measured differential phase is essentially the Cde Doppler velocity is estimated from the sum of the total phase shift of consecutive pulses the argument of the autocorrelation between those pulses see Sec 64 in BC and also eq 5208 Finally we can define the specific differential phase de as the range derivative of differential propagation phase de Cljqdp In practice ltde is filtered prior to computing de r Hydrocarbon Sources and Sinks Anthropogenic globally 10 Incomplete combustion pyrolysis original HC from fuel source cracked HC gt shorter chain length partially oxidized HC gt carbonyl fuel evaporation vapor pressure fuel type temperature Biogenic globally 90 green leaf emissions speciesspecific larger molecules higher atmospheric reactivity Hydrocarbon Sinks Reactions with the OH radical dominant during the daytime lifetimes from months to minutes complicated mechanism Reactions with ozone only alkenes often dominant at nighttime lifetimes from days to seconds l reactions with N03 radicals practically only alkenes only at night Hydrocarbon oxidation mechanisms l 1a RCHzCH3 OH RCHzCHZ H20 1b RCHzCH3 OH RCHCH3 H20 2 RCHzCHZ 02 M gt RCHzCHZOZ M 3 RCHzCHZOZ NO gt RCHzCHZO No2 Note 1 is the ratelimiting step in all cases R1a lt R1b 2 is very fast due to a 21 O2 abundance 3 is the dominant reaction in the BL over land Fate of RCHzCHZO or RCHO CH3 important Hydrocarbon Oxidation Mechanisms l 39O C oZ HOz 0 C Probability of 12 depends on chainlengt a O b a K 2 CH3 0 CHZCH3 gtCH3 CCHZCH3 H Decomposition Probability of ab depends on stability of carbon radical Hydrocarbon Oxidation Mechanisms Isomerization H H C OH CH3 1 CHs C CCHz a CH3 IC CHziCHZ CHCH3 H H2 H 2 CH3CH2CH2CH2CH2039 02 gt CH3CHZ3CHO H02 l ngrII i fafiion Pentanal CH3CHCH2CH2CHZOH M102 00 NOUNOZ 039 J CH3CHCH2CH2CHZOH gt CH3CHCH2CHZCHZOH llsomerization HO O 2 U 2 IDH OH CH3CHCH2CH2CHOH CH3CHCH2CH2CHO 4Hydmxypentanal Hydrocarbon Oxidation Mechanisms OH addition reactions QH A i OHC C c c OH EH i M C gt C C Properties 02 1 fast negative activation energy dominant OH reaction in photochemical smog Hvdrocarbon Oxidation Mechanisms III OH 0 i CH3CH2CH CH2 OH CHSCHZCH CHZOO39 NO No2 iH OH Isomerization I 1 L641 shift CHSCH2CHCHZO M 02 Of Decomposi on H 39OOCHQCHZCHCHZQH NO 39CHZCHZCH CHZOH NO2 H ijiHo2 OCHZCHZCH CHZOH c CH2CH0HCHZOH O 3 4 Dihydroxybutanal ATMO 689 Lecture 5 092804 Polarization Dielectric Refractive Index Stephens 4143 Battan 1973 Ch 4 Recall differences in returned power for ice and water Polarization of matter Polarization ClausiusMosotti equation Frequency Dependence Debye Relaxation Microwave frequencies Refractive Index Relationship to Dielectric or relative permittivity Recall Differences Between Ice and Water Recall equation for returned power Pr as a function of the radar constant C dielectric factor K radar reflectivity factor Z and range R P CKZZ r R2 Now let s consider returned power at R for water PM and ice Pri assuming Z is same for ice and water ie no ND is same Recall z ND 06 dB 0 C IKWIZ Z P C K2 Z rw Dri I II R2 R2 Since KW2 093 K2 0176 or KW2 gt W2 PW gt Pi BUT we still don t know physically WHY Recall Differences Between Ice and Water From RayleighGans Theory 2 2 lShhl Zd 10LOG10 r SW2 pv abz quot121 pk 3 Ahvm2 11 2 12 szi2 1 1 5 W sin 1e 1 2Ah e K 9 J 10L0G10 p hz Zclr is mathematically dependent on the refractive index m Why physically 1o e Rain 39 Solid ice l Aggregate ZDR dB o 05 10 AXIS RATIO F39s r 7W A L 39 39 39 ratio tor spheroids having the effective dielectric properties of raindrops solid Ice graupal and aggregate snow As axis ratio de creases lrom 1 toward 0 particle spheroidl shape becomes more oblate Curves shown tor graupel cover bulk densities of 03 g crrr lower boundary at shaded region to 16 g cm S upper boundary Values shown or aggregate snow cover bulk densities of 1306 g cm393 lower boundary of shaded region to 012 g crn39a upper boundary Calculations use the scaltering theory of Gene 1912 Polarization Because of its geometrical arrangement with two amp hydrogen atoms grouped on H H the same side of the oxygen H215 atom the water molecule 39 quot possesses a permanent electric dipole and hence dipole moment pqs I7 3657 cmquot 273 m Figure 104 Normal modes of vibration of th free water molecule char e ie and S q 9 q q Bohren and Huffman 1983 Dipole When an external electric field E is imposed to water the permanent dipoles align themselves parallel to the electric field lines The water is said to be electrically polarized and is said to be a die901770 Figure 41 Polarization of matter under the in uence or an electric eld Stephens 1994 Note in text italics indicate a vector quantity Polarization Polarization P describes the ability of the local charge distribution associated with atoms and molecules in the material to align to an incident electric field Mathematically the polarization per unit volume of matter associated with the external field E is defined as P 8r 1 80 E where 80 is the electric permittivity in a vacuum and er is the relative permittivity or dielectric constant So dielectric constant is a constant of proportionality between the incident electric field and the polarization of the matter Higher lower dielectric results in more orientation of dipoles and larger smaller polarization for a fixed applied external electric field The polarization per unit volume of N aligned dipoles per unit volume of a water droplet or ice particle is the vector sum of all the individual dipole moments PNp The dipole moment of an individual molecule p can be related to the locally active electric field E by p or E where or is the polarizability of the material Note that E and E are not the same Polarization From physics we know that the local field E is increased above the incident field E by r P E the electric fields associated E E 8 with neighboring dipoles 380 3 r according to By combining this equation with definitions of E P and p we obtain the Clausius Mosotti equation Relates the polarizability of the material to the dielectric constant Note Dielectric factor K is related to the dielectric constant ar according to Na 2 380 g 1 8 2 81 er2 Polarization Physically speaking Frequency dependence of the dielectric constant consider three frequencies High ultraviolet Medium infrared Low microwaveradar Ultraviolet only the lightest sub component of atoms ie electrons can respond to fast or high frequency radiation Electron orbital transitions occur in response to ultraviolet radiation DE BYE RELAXATION FUNCTIONS VI SRATIONAL EL ECTRON lC DIELECTRIC m vi Infrared more dense matter ie V toe minim entire atom can respond OscHlations are more sluggish and WWW 39NFWED frequencies assomated With Wm IC I Figure 47 Schematic diagram of the frequency variation of the di M electric function of an ideal nonconductor Bohren and Huffman 1983 occur when molecular dipole moments respond andorient Stephens 1994 themselves parallel to inCIdent E field Governed by the theory of Debye relaxation Debye relaxation time Relaxation time T time from one stable orientation of molecule to another eg time to orient dipoles in response to an incident electric field 1 depends on size of particle a temperature T and viscosity n of material and Boltzmann s constant K 17 4TEna3KT It is a ratio of the viscous restoring forces which maintain alignment to the thermal forces which disrupt alignment Physical interpretation Relating the dielectric constant refractive index and the dielectric factor or how many ways can physicists say the same thing Dielectric constant Sr is a r quot complex number 8r 2 8r i8r The refractive index m can also be used to describe the polarizability of matter and is also m n k complex Dielectric constant and I refractive index are related 8 n2 k2 according to 8r m2 so r 2 So Dielectric factor K is K m 1 1712 2 I 8r real H gr imaginary n real k imaginary Refractive index and K values as a function of phase and temperature an H The Components qf the Complex Index of Refraction TAALE 42 The Components of the Complex Index of Refraction 12 and the imaginary Fan of 110 of Water as Function 1 quot a quotW magmary P3 of K 0f 1quot 35 FunC i ms or Tempcralure and Wavelength f Tempemure i Ci m 3 1 3 061 n All temperature when a 092 yncrnJ 0 353 814 615 444 p n 10 902 780 545 394 0 899 714 475 345 13 a 648 415 310 quot 39 39 39 39 39 39 39 39 quot JD 20 063 200 186 259 Kl A111 mune K 10 090 244 190 137 whenrpGI 0 147 289 277 2 04 8 55 I 77 0 1m K 10 0928 09275 09193 0892 K l IO 09313 09182 09152 0872 0 09340 09300 09055 0831 souu Gunn Ind East 95 8 08902 0791 20 000474 001883 00471 0091 ImK I 10 000688 00247 00615 01 N 39 0 001102 00335 00807 0144 S 01036 0171 Cu E a I W 7 a Ki2 0176 for Ice at p 092 Some temperature dependence In water Increased absorption With decreasmg quot5 Value can Change by as H 39 77 39 o temperaturewavelength sluggish dipoles much as 18 A dependent on the Increased attenuationabsorption at shorter Ice a m39Xture39 wavelengths but also more fonNard scattering Note that IKliz is 5X smaller For most purposes eg our Rayleigh than KW2 We should know assumption we use KW 093 Why now Lecture 10 Gas Chromatography J n llu II quot E Mn Fmf r N sti EPJEE Column Resolution Deleclar slgnnl Figure 261 Scpumlinm ul IIIrcc mmlminns Hue K R 7 WAZ wnz ZAZ WA W uolu hus I 391 IlIe number or plulcsAII vcrsc consequence nl39 Ihe added plalCS lIowcver s 5 7 2mm mm 2620 increase in he limc rc uircd For he sex II WA wB q 1 3 quot A a A B Ag In l u x N If 5 Himgq l I U l I I I 1 WAIE It 1 WHO K I l r IVAA l wn l I l I I I l l I Tim mm ZAZW 1 WII Skoog Holler Niemen Chapter 26 Summary The General Elution Problem rWT H Mb Skoog Holler Niemen Chapter 26 mu m mm ambuugmmm mumquot mum and 1 mm 5mm nfExpl mlan llmmli mummm Mum u rummamp39lml mum m1 munmnltg nm n we may mm mm mm an m A mmuc m ummury pm A K N mnlnmlm Naval w mm A w my m m mum 11 w Mmmmuv pinks mu 265 In nrlanl Derived Quavm m um l Rclmonnups m uf mm mm wammp m 1quotqu mm hum unmbwun comm Mammy mm mm lumherefphvu N my N f I Pm hm mm Skoog Holler Niemen Chapter 26 Instrumentation Basic Gas Chromatography W 27 l mm m m gnmn unlm Skoog Holler Niemen Chapter 26 Sample Injection System Liquid or Gas mportant Slow injection causes band spreading poor resolution Sample is 01 to 20 microliters typical packed column GC or0001 microliters for capillary GC 33 rlngc ll 9quot th L39L i Lizlimit I ilb39 wrist i H IL L IHE39 439 LILIIIZJ3939T rsl IL M Irw l 399 39quotur39r391uili I i39 L I39139 I q V a I I 3 3 I w 1 I ira i Lll39illl f IIIII IL 39 i l Hauntrm Iquziinui Flgilri 27 3 l 3950EdiLilian view I palm V 1113 39L J m Hm Llir39l il39liL39u lnr P l Skoog Holler Niemen Chapter 26 Rotary Sample Valve A Fill sample loop 2 1393939l 3939 rr39W1 lll rIJ39l139ll39l i r39plll39JHullll39 l39h39 39l39l I i HJJLI l ud 39 LI 3 I I39 y elml H39 Iquot u I H sumpIL I1 39 39 ul n 1 quot 35 39IL39 quot 39 lg I 39I I II 1r 1 I ssh7 4 l l w III I I w I J l H 51 it VI r7 z r I39I39 aquot 5 I r l 3 I 1 lr 1T 7 II If HJillllll k39 hulvplu Fm ugllu fl min in Ilill II nu l IIII N EST AI v ll 1 r1 III 39Ialquot3939 III39L39 MWHull i1 2I Il I39llllllg Eil39llllll39 122ml HUIl II fill inlmxlmmn a mmIiEw intn urelunzr Skoog Holler Niemen Chapter 26 Cmumn COH g LI rations Length 250 m 1 Packed 030 33 2 2 Open tubular or capillary 3 Temperature Skoog Holler Niemen Chapter 26 Temperature Programming 45 c X Wu ijwiihquot L 1450 301800 4 Wm quot I u 1 Many choices for Stationary Phase for packed and tubular columns min I 272 mm lt omnum Nationquot y Phase for m u fllmnmmgnlphy ow mm m w 1 ms nu mm sw m r gt m w Many Choices 1m mm Wm WM an m rmmmmmu x a mqu l39rnpemu Manama zsmnnmcmw Immnugmymm 01mm w M quotum l Isml lull scm39 1mth hum u my mum uunmxnu mmm nun mm WM Hm H mm m wm w LUV a m up mm ILLMH Hum Lw um LN mm m w M H mmmwmw mrl mum w m m mm m w m WWW pm we Lu W mlumn m I39lun39ul wmlm h m Hm nmdch nmlrhw mume um um Open Tubular Columns HI Westerly wind increasing in magnitude with height F mm mm m m m us 60 2 Iwccnov dst rum Ixrn All idealized imagery of Doppler radial velocity herein is from Brown and Wood 1987 Westerly Wind with midtropospheric Wind speed maximum aft FFFFFFFF v21 EIGHT FF39E l on M 5 HEIGHT EFT I quotCl 11395 Jf39El HE 359 DIRECTION 518 D m in El B39JI SPEEH ill15 Dif uent Westerly Wind constant speed with height Con uent westerly Wind constant speed with height HEIGHT IKFTI u Veering wind pro le with height 12 quot9 ng 015210 constant speed HEIGHT IKFn E 1 r o D so 2a m 50 SPEED KT a Mind prof e b Winds on PPI surface To find wind direction at given range height draw line segment starting on zero white line at w range of interest to origin Wind is perpendicular to this line with direction givenby C0101 key C Interpretation of display d Doppler ve1oc1 ty dismay Envirg iigfwmd Source NOAA FMHll Backing wind pro le constant speed 21 1 4 J HEEGH T mF39II E l r39 39 115 5a 215 am EH59 mm 35539 wigg I i I 3 All El SPEEEI Hill 395 I mm m 1 L 075 quot1 U 3953 39 W 1 1 quot Backing Backward S Veering t0 midtroposphere then backing constant speed 2 I E E u I I W I35 1 225 EN 53 umcmmu IDEGI 2 E 5 Eli 139 E quot39239 l a 39539 n a 1 an an SPEED lam ELK FFFWM Approaching cold front Figure 65 Convective Storm Display Window Location and relative size of the 27 x 27 nmi 50 x 50 km window used for displaying simulated Doppler velocity patterns within convective storms The window is 65 nmi 120 km due north of the radar located at the center of the overall display region Source NOAA FMHll Pure cyclonic rotation cg mesocyclonc Pure divergence Mesocyclone with Tornado Vortex Signature TVS embedded in center TVS small scale circulation associated with tornado seen at close range as pixeltopixel variation in radial velocity which is typically folded ATMO 689 Lecture 4 092304 Backscattering matrix coefficients and radar observables for oblate spheroids DZ Ch 852 BC Ch 2 Backscattering matrix S revisited Backscattering radar observables Zh ZV Zclr LDR as a function of S RayleighGans theory for scattering by oblate spheroids Brief physical overview Presentation of backscattering matrix elements for an oblate spheroid Backscattering Matrix Equation Revisited The backscattered electric field Eb from a Single hydrometeor Backscatter Matrix 8 for each polarization h b l horizontal and v vertical can Eh Shh Shv Eh e kor be related to the incident electric field E for each EV Svh SW EV 1 polarization as a function of range r via the backscattering matrix 8 note k0 wave number S is an intrinsic ie inherent Eb property of the hydrometeor causing the backscattering Size shape orientation El dielectric ie phase density Backscattered power and hence radar reflectivity for Horlzontal re eCt39V39tyi Zh IShh2 each polarization is Vertical reflectivity ZV SW2 proportional to the square of the appropriate 8 term Backscattering Radar Observables from Backscattering Matrix S Reflectivity factor at horizontal polarization zhzzhh Copolar h ZhdBZ1OLOG10zh Reflectivity factor at vertical polarization zVEzW Copolar v ZVdBZ1OLOG10zV Differential reflectivity Zdr Copolar hCopoar v Zdr 10LOG10zhzv ZhdBZ ZVdBZ Linear depolarization ratio LDR CrosspolarCopolar LDRhv 1OLOG10thzw LDRvh 1OLOG10zvhzhh Theory of reciprocity LDRhv LDRvh LDR In practice LDRhv e LDRvh LDR Zh 414 4le2 5 2 2V 424l7r4lelzlllSwlz 2 ZdrleLOGlo ISh llz SW 2 LDth 10L0G10 h 2 S 2 L1th 10LOG10 SV h2 hh RayleighGans Theory Stephens Ch 5 BC 1323 DZ 8524 Rayleigh scattering radiation scattered by a single dipole which you can consider a small spherical particle that is much smaller than the wavelength of the incident radiation Gans 1912 extended Lord Rayleigh s theory to oblate and and proate spheroids V V The small dipole oscillates at the frequency of the incident L h E39h Ebh electromagnetic field producing a secondary field that radiates out in all directions ie the scattered field The incident electric field has two orthogonal components Eih horizontal and EiV vertical Each component of incident electric field independently induces a dipole moment in the same polarization plane that creates that scattered electric field or ESh and ESV respectively The resulting intensity of the excited dipole in each polarization plane is not only proportional to the incident electric field strength but also to the size shape orientation and dielectric factor ofthe scattering particle For radar we are often concerned with backscattering or radiation scattered 1800 from the incident field ie back to the radar Ebh and Ebv respectively RayleighGans Scattering Geometry Simplifications we will typically make 8 00 Le particle is aligned with polarization plane 9 00 zero radar elevation angle for now P is fixed say 00 Other assumptions possible probability distribution for orientation angles to simulate LDR for hail see BC 236 and BC Fig 28c that follows KEY v vertical h horizontal Symmetry Axis Wt Doviak and Zrnic 1993 Plane of Polarization Aopagation Direction Fig 815 Scattering geometry where n is the symmetry axis ol the scatter n39 is the projection ofn onto the constant phase plane 8 is the radar elevation angle in and v are the linear polarization base vectors and u is the canting angle 0139 the scatterer The vector v denoting venical polarization is in the A planet 6 angle between the propagation direction of incident field and symmetry axis n 1 canting angle or angle between incident electric field and the projection of the axis of symmetry on the polarization plane 6 radar elevation angle Rayleigh Gans Theory Backscattering Matrix Elements The backscattering matrix elements for an oblate spheroid eg raindrop are given by p 249 DZ Shh k02Pv39PhSinz5Sin20 Ph SW kozPVPhCOSZ5COSZ P Ph phy ShV 05k02pVphsin26sin2 P where e eccentricity ab minormajor axis ratio m refractive index more later minor K ml 1 3 Ahvm2 11 1 1 e2 1 AV eZ 62 V Obate Spheroid major 1 39 12 sin1e 1 2Ah If 5 PO You too can calculate Zdrand LDR xwxwxyxx x 12 Raindrop a an am m E 3 e Graupei 7 E E 45 N S m K xx 40 E R xx 45 Snowflake x 50 A55 60 l l i t ova o 9 02 on 05 0 in t t Axis rah ba 0 7 0398 o 9 Axis ratio lba 0 mlnormajor l nae 22 up Di 39el enlinl reliccltvit 2dr of h single particle nt39ohlalc shnne Versus nxn mnh Hg 23 W0 Note isot rop I C o r Several purliclc types with different densities are illustrated 0 show the strung weighttn d on the dial wtnc constant t nr n given 171 1 Summary at typical zd Vulucs or tntnthnps nl39 ran d 0 m 1 0t xed ul39iotlh silm mid hzli The black tittows 0n the hail panicle IEprcSCnt tile tumbling Inalionn in n m I no I39ll a O r rnlh in n lhundersmrm Adapted from Wakimnto nnd Bringi 1988 L39L1n r dcpnlal izaltun 0 n e ntat on f0 r L D R mun hr 39 u g 39 39 w N 1272versut nxls t ln Fig 2xn Note slmng weighting nl LDR nllh dielectric constant on tlcttxlty l39tlru txch axis ratio rlllm Purliulu lprS zlrc 39 For Oblate Spheroids from Gans Theory next time refractive index of water ice and iceair mixtures Bringi and Chandrasekar 2001 Differential Reflectivity Ice dielectric effects Response of Z r to hydrometeor shape for Ice IS very different than for water drops Shapes of ice particles and water drops are different Also Zdr sensitivity to hydrometeor shape varies with dielectric constant of the scattering hydrometeors Since dielectric constant of ice is about 20 that of water particle shape has a much smaller effect on Zd measurement in ice than in liquid water hydrometeors Inclusion of air in ice particles of low bulk density eg snow lowers effective dielectric constant further Fig 2 Z r dB vs axis ratio ab for oblate sp eroids of varying dielectric Rain Solid ice hail ice crystals 09 g cm393 Graupel bulk density 0306 g cm393 Snow bulk density 003012 g cm393 Canting can also decrease Zdr Mean canting angle in rain is zero Hail wobbles and spins in descent 0 05 10 AXIS RATIO ab FIG2 P h A 1 1 u 5 i t 4 I 39 ratio for spheroids having the effective dielectric propenies of raindrops solid ice graupel and aggregate snow As axis ratio de creases from 1 toward 0 particle spheroid shape becomes more oblate Curves shown for graupel cover bulk densities of 03 9 UN8 lower boundary of shaded region to 06 g cm 5 upper boundary Values shown for aggregate snow cover bulk densities of 003 9 CMquot3 lower boundary of shaded region to 012 g cmoquot upper boundaryj Calculations use the scaltering theory of Gans 191 2 JOURNAL OF GEOPHYSICAL RESEARCH VOL 103 NO D17 PAGES 22425 22435 SEPTEMBER 20 1998 Observations of isoprene chemistry and its role in ozone production at a semirural site during the 1995 39 Southern Oxidants Study T K Stamf ziP B Shepson1 S B Bertman3 J S White1 B G Splawn1 D D Riemer4 R G Zika4 and K Olszyna5 Abstract Isoprene and its oxidation products methyl vinyl ketone MVK and meth acrolein MACR were measured in a semirural environment that was occasionally heavily impacted by urban emissions At this site isoprene was the most important hy drocarbon in terms of kOH hydrocarbon but the aldehydes HCHO and CH3CHO also appear to be very important The local iSOprene photochemistry appears to be occasion ally enhanced in NOx rich urban plumes that are advected to the site over intermediate forested land When 03 was being rapidly produced in urban plumes advected to this forested site isoprene was found to contribute z28 of the total ozone production We observe that many of the peaks in isoprene oxidation products atvthis surface site arise 39 from downward mixing of more photochemically processed air aloft as the nocturnal inversion breaks up in the morning We estimate that in the daytime typically 12 of the NOy at this NOXrich site is composed of isoprene nitrates 1 Introduction Research conducted over the past decade has shown that iso prene oxidation can contribute signi cantly to tr0pospheric ozone production even for urban enviromnents Trainer et al 1987 Chamez39des et al 1988 199239 Biesenthal etal 1997 For rural areas ozone production can be NOxlimited that is dependent mainly on the availability of the relatively limited precursor NOX rather than on the abundant biogenic volatile organic compounds VOCs Chameides et al 1992 Thus a forest environment that is sometimes downwind of an urban center with associated combustion NOx sources is an interesting case in that NOx lev els and thus ozone production ef ciency could be quite variable providing an opportunity to assess the sensitivity of O3 produc tion to available NOX From June 29 through July3925 1995 we participated in the Southern Oxidants Study SOS in the Nashville Te1messee area To investigate the role of isoprene in ozone production and the sensitivity of this chemistry to the availability of NOX we conducted measurements at a surface site that is 32 km southeast of the Nashville urban center In this study we focused on measurement of the isoprene oxidation products MVK and MACR as it has been shown that such measurements can be used to quantify the contribution of isoprene oxidation to ozone production Biesenthal et al 1997 1Department of Chemistry Purdue University West Lafayette Indiana 2Chemistry Department West Chester University West Chester Penn sylvania 3 Chemistry Department Western Michigan University Kalamazoo 4Rosenstiel School of Marine and Atmospheric Science University of Miami Miami Florida 5 Atmospheric Sciences Department Tennessee Valley Authority Muscle Shoals Alabama Cepyright 1998 by the American Geophysical Union Paper number 98JD01279 014802279898JD01279O900 The Nashville area is ideal for studying urban and rural pho tochemical interactions The city is fairly isolated from other major metropolitan centers Much of the surrounding area is a mixed deciduousconiferous forest In addition there are several NOx and 802 point sources from electric power plants in the Nashville vicinity Thus the distribution of ozone precursors volatile organic compounds VOCs and NOX can be highly variable at this site In this paper we describe a set of measure ments of isoprene MVK and MACR as well as several other VOCs at the Youth Inc measurement site Here we discuss the details of the measurement techniques used and the use of the data produced for assessment of the role of isoprene in ozone production and NOx removal in the southeastern United States 2 Experiment 2 1 Analytical Procedures Measurements of hydrocarbons and carbonyl compounds were conducted using an automated solid sorbent preconcentra torGCMS system The sampling system uses a dual Tenax TACarboxen sorbent trap for preconcentration as described by previous researchers Yokouchi et al 1986 Y okouchi and Sano 199139 Ciccioli et al 199239 Biesenthal et al 1997 Our system functions essentially as described by Biesenthal et al 1997 with some modi cations as summarized below A measured volume of air typically 05 to 075 L is drawn using a GAST diaphragm pump Model DOAPlO4AA through the cooled solidsorbent trap to extract the VOCs from the air sample The sample air ow rate was measured using an MKS mass owme ter Model 258C which was calibrated against an MKS National Institute of Standards and Technology NIST traceable mass owmeter Model 358C The sorbent trap was constructed from a 18 cm long by 47 mm ID 304 stainless steel tube The solid sorbent bed maintained at 20 C for sampling consisted of 0234 g of TenaxTA and 0200 g of Carboxen569 Air samples rst encountered the Tenax TA bed followed by Carboxen569 as a backup to prevent sample breakthrough Breakthrough volumes 22425 22426 were measured for each compound of interest at 20 C and were found to be 21 L for each compound as a precaution we util ized a maximum sample volume of 075 L The concentrated 39analytes were desorbed from the trap at 250 C with the He car rier ow reversed so that the analytes that were trapped on the Tenax do not encounter the more adsorptive Carboxen during the desorption step The temperature of the trap block was varied by either cooling it to 20 C using chilled water from a Neslab RTE111 constant temperature bath or by heating it to 250 C using heater cores embedded in the block The trap block and the gas flow switching valves were housed in an oven maintained at 100 C The oven and block temperatures were regulated with Omega CN9000A controllers Valco E60 electric valve actuators which resided outside of the 100 C oven were used to change valve states 39 39 The descrbed analytes were transferred into a 30 m x 032 mm PoraPLOTQ capillary GC column contained in a Hewlett g Packard 5890 Series 11 Plus GC Although the desorption is a relatively slow process all analytes are collected at the head of the column which is maintained at 20 C After desorption is complete the column was programmed from an initial tempera ture of 20 C to a nal temperature of 200 C Selected analyte mass fragment detection was achieved with a HewlettPackard 5972 MSD that scanned mz 45 to 170 at a rate of 10 scans squot For isoprene MVK MACR acetone propanal and toluene quantitation was done through integration of the selected ion chromatographic peak areas for mz 67 55 70 58 58 and 91 reSpectively A sample selected ion chromatogram from an air sample obtained at the Youth Inc site during the study is shown in Figure 1 As shown in the gure for isoprene and MACR the combination of the capillary column and mass selective de tection provides for very selective measurement of these analytes All functions of the GC MSD and autosampler are controlled through a desktop computer running HP Chemstation and in house software written with LabVlEW 31 National Instru ments Grade 01 clean air and ultrapure helium Grade 50 were supplied to the autosampler and GC from compressed gas cylin ders Airco Oxygen and moisture traps ltered the helium be fore reaching the GC and the clean air line had hydrocarbon and moisture traps in line The carrier gas supply exiting the GC electronic pressure control system EPC was rerouted through 1 the autosampler so that the helium ow rate through the trap during desorption was maintained by the GC A separate helium line was used for purging the trap after desorption prior to sub sequent sampling Calibration for each analyte was conducted by measurement of the mass selective detector response relative to that for acetone A response factor for each hydrocarbon of interest relative to acetone was determined from gas phase standards made in F EP Te on bags The absolute instrument response to acetone was then measured at least twice daily using a calibrated acetone permeation source The output of the acetone permeation source maintained at 40 C was independently determined by high per formance liquid chromatography HPLC using the 24 dinitrophenyl hydrazine DNPH method as outlined by Sirju and Shepson 1995 The permeation source was supported in side a stainless steel tube positioned in an oven controlled to 40 C using an Omega CN9000A temperature controller The permeation source output was then diluted with a known ow rate of clean air using a calibrated mass ow controller MKS Model 1259C to low ppb mixing ratios This sample is then STARN ET AL OBSERVATIONS OF ISOPRENE CHEMISTRY Retention Time min 8 3910 a 12 14 16 18 14000 2 A 12000 10000 8000 6000 4000 2000 71495 Youth Inc Signal total ions 1600 67 1200 d 800 4 12 ppb lsoprene 056 ppb MACR Signal mz no 000 000 Signal mz70 8 Amenrs O O O O O m l 10 12 14 16 18 Retention Time min Figure 1 top Total ion chromatogram and selected ion chro matograms for middle iSOprene and bottom MACR for a sam ple obtained July 14 1995 sampled in a manner identical to that for ambient samples under control of the Labview software routine Measurements for 03 CO NO N02 and NOy were con ducted using TEII Model 49 03 TEII Model 488 CO and TEII Model 428 NONOx analyzers respectively The NOx and NOy measurements were conducted via 03 chemiluminescence after conversion to NO The molybdenum converter which was used to convert NOy into NO was removed from the TEII Model 428 and placed at the sample inlet The N02 measurement system consisted of a photolytic cell to convert N02 into NO The NO mode of the 11311 Model 428 was time shared 5 min cycles for the ambient NO and N02 measurements Further details on the ambient air measurements for 03 CO 802 NO N02 and NOy are given elsewhere K J Olszyna et al unpublished manu script 1998 Peroxyacyl nitrates were measured as described in Nouaime et al this issue 22 Site Description and Sampling Procedures Ambient measurements were conducted as part of the 1995 Southern Oxidants Study at the Youth Incorporated Ranch in LaVergne Tennessee This site was roughly 32 km southeast of Nashville Tennessee at 36 037739N 86 305839w on a peninsula of the Percy Priest Reservoir at an elevation of 171 m above sea level The instrumentation was set up in a small trailer located in a roughly 16000 m2 clearing Vegetation immediately sur STARN ET AL OBSERVATIONS OF ISOPRENE CHEMISTRY rounding the site consisted primarily of osage orange trees Maclura pomz39fera along fence rows Within a half mile of the site the forested areas consisted of a variety of deciduous hard woods To the east and north of the site across the reservoir stood the Cedars of Lebanon State Forest To the south were the small communities of LaVergne and Smyrna about 97 km away V Ambient air at the site was sampled through a 5 cm 1D Pyrex glaSs manifold the entrance of which was 20 m above ground level Air was pulled through the manifold at a rate of 2000 Lmin Our instrument sampled this manifold air through 4 m of PFA Te on tubing 6 mm ID The sample passed through a potassium iodide ozone scrubber before entering the trap Tests showed that the scrubber effectively removed ozone but did not 7 signi cantly affect the analyte concentrations 3 Results and Discussion 31 Data Quality and Characterization In this study a large data set has been produced covering a 5 week period during which several key reactive organic com pounds were measured along with ozone components Of NOy CO and meteorological parameters As such this data set is potentially very useful for model evaluation and some discussion of data quality is important During the measurement period we 39 conducted periodic measurements from certi ed compressed gas cylinder standards that were diluted to nal low ppb mixing ratios with clean air This was done with both commercial stan dards Scott Marin and from standards prepared at the National Center for Atmospheric Research NCAR Apel et al this is sue This was done for the analytes acetone isoprene MVK and MACR On average our quantitative determinations dif fered from those certi ed or independently determined at NCAR for the cylinders ie actual by 33 15 5 and 14 for ace tone isoprene MVK and MACR respectively over the whole study Ape et al this issue Although it is extremely difficult to judge the quality of ambi ent measurement data for these compounds without conducting an independent intercomparison study some assessment can be made from examination of ratios of concentrations for species given a known relative production rate This has been done pre viously by Montzka et al 1993 who measured MVK and MACR at a forest site in Alabama The yields for production of MVK and MACR resulting from OH reaction with isoprene in the presence of NO are 32 and 23 respectively T uazon and Atkinson 1990 Miyoshi et al 1994 Thus in the absence of subsequent photochemistry the observed ratio MVKMACR would be expected to be 14 However as discussed by Montzka et al 1993 OH reaction increases this ratio While 03 reaction decreases this ratio as a result of the differential reactivity of MVK and MACR with these oxidants The calculated diurnal average MVKMACR ratio for all data obtained during this study is presented in Figure 2 top As shown the maximum in the average ratio was 14 This is considerably lower than that observed by Montzka et al 1993 and Biesenthal et al 1997 However in this environment NOx levels are relatively high and OH levels could be suppressed see Discussion below The MVKMACR ratio is highly dependent on the OHIO ratio As suming MVK and MACR production via OH and O3 reaction with isoprene and destruction via OH and 03 reaction the ratio will approach steady state values ranging from roughly 20 for 22427 O3OH 5x105 100quot ppb 03 5x106 moleculescm3 OH to as low as 10 for 03OH 2x106 40 ppb 0 5x105 mole culescrn3 OH Thus the average values shown in Figure 2 are not surprising but do indicate that on average OH radicals may be suppressed at this site Given the results of the cylinder analyses and the MVKMACR ratios and our own uncertainty analysis based on the uncertainties in the standards we esti mate the overall uncertainty for the isoprene MVK and MACR data as 20 39 During the week July 1216 1995 there was an oxidant epi sode throughout much of the eastern United States see Discus sion below and isoprene MVK and MACR concentrations were relatively high indicating actiVe isoprene photochemistry As shown in Figure 2 bottom the MVKMACR ratio during this period exhibited a strong diurnal variation resulting from rapid isoprene oxidation in the daytime along with relatively faster consumption 39 of MACR and relatively greater production of MACR at night from ozonolysis during these relatively high 03 conditions Thus the pattern observed in this photochemi cally active period is essentially the same as previously observed by Montzka et al 1993 and Biesenthal 39et al 1997 We note that although both MVK and MACR have anthropogenic sources Biesenthal and Shepson 1997 based on the observed CO mixing ratios ie typically 4OO ppb the contribution of auto motive sources to MVK and MACR at this site is relatively very small The largest contribution occurred for the early morning of PurdueYouth Inc MVKMACR ratio data 20 13 Diurnal average 1 full study 390 14 a l MVKlMACR 900993 OthmON 111 1 l f I F T l T l l 396 81012141618202224 Time of Day 0 NM 225 200 l 175 I 150 125 39 l a 075 050 025 000 w I 0 12 O 12 0 12 0 12 0 Time of Day Figure 2 Diurnal average MVKMACR for top all data and individual MVKMACR for the period July 1215 1995 MVKMACR July 12 39 July 14 July 13 July 15 r T 22428 Day of Year Day of Year Figure 3 Ozone isoprene and MACR data for the entire study period July 15 when CO reached 750 ppb and the calculated contri bution to MACR from automotive sources see Biesenthal and Shepson 1997 was 55 ppt or 19 of the observed MACR of 294 ppt 32 Characterization of the Youth Incorporated Site In Figure 3 we present the measurement data for 03 isoprene and MACR which enables us to summarize and highlight sev eral features of the measurement data over the course of the study period In Figure 4 we present the NOy and NOx data for the study period In Figures 3 and 4 the dates are shown in num bered day of year format where day 183 equals July 3 1995 Daytime ozone concentrations ranged between 30 and 100 ppbv for rst 6 days of the study with a clean air mass entering the region on day 184 The period July 917 was characterized by a stagnant high pressure system that began as a large high pressure ridge over the southwestern United States and moved eastward leading to very high temperatures with daytime temperature maxima in the 35 37 C range and low wind speeds in Tennes see These conditions are conducive to photochemical ozone production as described in detail by McNider et al this issue As shown in Figure 3 relatively very high 03 gt120 ppb was observed for the period July 1316 at Youth Inc This system remained until about July 17 Throughout the period July 417 the daytime ozone mixing ratios at this surface site in creased fairly steadily in part as a result of recirculation around 140 183 186 189 192 195 198 201 204 B0 M q 120 Ozone Data 1101 L i 100 a 381 3 mi 2 401 39 30 q i i 20 1 lO 0 1 Q 7 L lsoprene Data l 9 6 Q P39I 5 4 a 3 8 2J 01 399 12 Methacrolein Data 2 L0 FA 08 a L 06 OA quot39 02 00J v f r I 1 183 186 189 192 195 198 201 204 STARN ET OBSERVATIONS OF ISOPRENE CHEMISTRY the stagnant high pressure Presumably ozone precursors and other photochemical products increased as well Temperatures were highest during the period July 1216 when ozone was ob served ito exceed the National Ambient Air Quality Standard NAAQS at this site Figure 4 shows that this is a relatively high NOx environment with total odd nitrogen mixing ratios ie NOy ranging from lt1 ppb to 40 ppb During the period July 1215 the NOy levels steadily increased at this site with daytime NOy levels ranging between 10 and 25 ppb in the afternoon of July 14 Surface ozone at Youth Inc reached a maximum concentration of 39137 ppb quotat 1600 on this day During this week daytime isoprene mixing ratios were typically 23 ppb The maximum isoprene mixing ratio observed at this site actually occurred in the evening of July 13 at 2130 when the isoprene mixing ratio reached a value of 8 ppb This event and associated nighttime isoprene chemistry is discussed by Stam et al this issue Given the lev 39 gels of isoprene at this site the relatively high ozone levels ob served and the variability of NOx there is an opportunity with this data set to evaluate the contribution of isoprene chemistry to ozone production in urban plumes that impact on forest environ ments and the sensitivity of the forest isoprene chemistry to the availability of NOX I It is clear that isoprene is a very signi cant contributor to lo cal scale 03 production at this site This can be demonstrated with a comparison of the values of kiVOCi for various VOCs where ki is the rate coef cient for OH reaction with VOCi since this is proportional to the rate of production of peroxy radicals via OH reaction with VOCi Such a comparison for the data from July 14 at 1400 LT when the O3 mixing ratio was 122 ppb is Odd Nitrogen Data Youth Inc Day of Year 183 186 189 19239 195 198 201 204 No Noyt ppb NOXL ppb 183 186 189 192 195 198 201 204 Day of Year Figure 4 NOy and NOx data for the entire study period STARN ET AL OBSERVATIONS OF ISOPRENE CHEMISTRY 22429 Relative VOC Oxidation Rates 1400hrs July 14 1995 Youth Inc TN 39 VOC benzene pentenes pentanes butanes higher alkanes toluene butenes alkyl benzenes Pinenes 39 acetone CZH5CHO CH3CHO UL uu U9 UU m MACR MVK 1 isoprene co CH4 l 2 O u 3 4 5 6 km x VOC squot Figure 5 Bar graph representation of the relative VOC oxidation rates for t 1400 on July 14 1995 shown as a bar graph in Figure 5 We have included estimated values for HCHO and CH3CHO for mixing ratios of 20 and 5 ppb respectively from the data presented by Apel et al this is sue The actual HCHO and CH3CHO mixing ratios during this time are highly uncertain as measured values obtained by sev eral investigators using the DNPH method varied signi cantly Apel et al this issue However we have included them in Figure 5 since as shown isoprene C0 and HCHO and CH3CHO were the most important reactive VOCs measured at this site for this episode day This gure is meant not to convey a de nitive reactivity ranking for the measured compounds39but to express the importance of isoprene HCHO and CH3CH0 as reactive VOCs in this semirural environment We note that C2 and C3 hydrocarbon data are not available and it is likely that ethene and propene contribute signi cantly 33 N0x Dependence of Isoprene Chemistry That ozone production is a nonlinear function of NOx has been previously discussed in the literature cf Lin et al 1988 1991 Chameides et al 1992 03 is produced as a result of oxi dation of NO to N02 in reactions 2a and 3 below ie fol lowing N02 photolysis OH VOC 02 R02 39 1 R02 N0 v R0 N02 2a H02 N0 0H N02 3 The chain reaction terminates via reactions 46 and 2b be low among others 0H N02 o HNOg 4 H02 H02 H202 02 5 H02 R02 gt ROOH 02 6 R02 N0 RONOz 2b 7 Competition for 0H between VOCs and N0x in reactions 1 and 4 V is partly responsible for the ozone production ef ciency e g chain length dependence on the VOCN0x ratio Competi tion between peroxy radical reaction with NO in reactions 2a and 3 producing 03 and chain propagation and with other peroxy radicals in reactions 5 and 6 leading to chain tenni 39 nation makes the ozone production ef ciency a complex func tion of the absolute N0x level Because of these competing re actions the maximum ozone production ef ciency depends on the concentration of reactive VOCs but is typically achieved when NOX is in the low ppb range Given that the Youth Inc region is forestedagricultural with abundant biogenic hydrocarbon sources and that the measure ment period was in midsummer we might expect 0H mixing ra tios and 03 production rates to be NOXlimited A plot of day time 03 versus NOy for this site shows that 03 increases linearly with NOy up to about 5 ppb NOy ie is NOXlimited up to roughly that point but r is independent of NOy for N0y mixing ratios greater than this A reasonable way to test the NOX dependence of the local isoprenedominated chemistry is with the MACRisoprene ratio Since the production rate of MACR is a function of the integrated product 0Hxisoprene if OH is dependent on NOX levels a relationship should exist between this ratio of productreactant and NOX for daytime conditions We expect this might be the case since for typical daytime MACRisoprene ratios photochemical MACR production from isoprene oxidation is several times faster than MACR destruction by OH In Figure 6 we show a regression of MACRisoprene versus NOX for data between 1200 and 1500 when NOx s 3 ppb As shown there is a tendency toward larger MACRisoprene ratios as N0x increases however the rela tionship is weak At NOX concentrations higher than 3 ppb data not shown the MACRisoprene ratios are highly variable ranging from 016 5 but several points are on the low end of this range Although there are relatively few data points to substanti ate this it appears that at high NOX levels 0H and thus iso prene chemistry is suppressed as a result of reaction 4 above 22430 STARN ET AL OBSERVATIONS OF ISOPRENE CHEMISTRY Youth Inc 1200 1500 MACRisoprene H ppb This is consistent with the modeling results of Lin et al 1988 1991 indicating that the OH radical concentrations maximize in A the low ppb range and that oxidation chemistry slows at higher NOx levels In fact much of the data at this site appears to be in conditions where oxidant production is NOx insensitive Daytime 12001800 MACRis0prene ratios were typi cally 015 and increased to 0607 on average at night be cause of the faster removal of isoprene at night Stam et al this issue MACRisoprene averaged for N0ylt3 and NOygt3 ppb as a function of time of day showed that the two data sets are not statistically signi cantly different This is likely the result of sampling air masses with varying degrees of photochemical proc essing for a particular NOX level However there are Several in dividual events during the Study in which urban plumes with as sociated elevated N05 levels impacted on the site with dramatic impacts on the extent of isoprene processing An example is shown in Figure 7 which shows the isoprene NOy and MACRisoprene data for July 3 1995 An NOyrich plume impacted on the site at 1700 with a concomitant increase in MACRisoprene by over a factor of 10 to 12 a very large daytime value Some of this change is related to the decrease in isoprene however the MACR concentration increased from 015 to 050 ppb The surface wind direction data for this day shows that the wind direction moved from southerly in the morning to westerly by about 1500 and then to 310 ie exactly from the direction of Nashville at the time of the plume impaction We interpret this type of event as resulting from an increase in OH processing of isoprene in the NOxrich air as it is transported in a plume from the Nashville urban center over the vegetation on route to our measurement site As discussed below for Figure 11 there was also a substantial increase in MPAN for this event The data for this day present an excellent example of the differences in degrees of biogenic hydrocarbon processing under varying NOX levels 0 O O I 0 gm quotq O 180pr Figure 6 Regression of MACRisoprene versus NOx for data in the time period 12001500 and for NOylt3 20 25 30 Typically high pollutant concentrations are observed after sunrise that is after breakup of the nocturnal inversion An ex ample is shown in Figure 8 for July 13 This day was the rst of a 4 day oxidant episode in the Nashville area As shown in the gure NOy increases dramatically starting at 0800 the approxi mate time of the nocturnal inversion break up Thus for this case there must have been a much more polluted layer aloft in the residual layer which mixes down at sunrise The NOZ NOy NOX data show that this is not a fresh pollution plume as over half the NOy is NOx oxidation products This situation would arise as a result of nighttime transport of relatively polluted air aloft of the inversion followed by downward mixing after sun rise Interestingly this air has a relatively high MACRisoprene ratio 05 It is also intriguing that the MACRisoprene ratio increases before the inversion breakup As discussed in S tam et al this issue this likely arises from transport of more isoprene chemistry impacted air under the inversion and relates to the heterogeneity of isoprene sources locally A perhaps simpler case can be seen in Figure 9 for July 20 in which the MACRisoprene ratio is well correlated with the morning NOy downmixing peak This morning peak in the MACRisoprene ratio is likely a result of downward mixing of the previous day s isoprene oxidation products The MVKMACR ratio at the time of the morning peak is in the range 1214 As discussed by Stam et al this issue N03 reaction can often cause rapid con sumption Of isoprene at night while producing very small amounts of MACR as the yield is only 35 Kwok et al 1996 N03 reaction with MACR is negligibly slow Kwok et al 1996 Thus some of the morning increases in this ratio re ect the much longer lifetime of MACR relative to isoprene at night It is likely that much of the N03 chemistry occurs aloft in the residual layer which is isolated from surface NO sources which would otherwise tend to suppress N03 However there are cases when the polluted air aloft contains substantially elevated concentra STARN ET AL OBSERVATIONS OF ISOPRENECHEMISTRY July 3 1995 Youth Inc 22431 25 20 g MACRMISOP 0 No o 20 y f 16 6 o lsoprenequot ltD a 2 o a 2 O 1 2 Q 0 8 E 0 lt1 05 quot 0 o 4 E o P 0 O l l I F O 8 10 12 14 16 18 20 Time of Day39 Figure 7 Plot of MACHisoprene isoprene and NOy for July 3 1995 tions of isoprene oxidation products 39An example is shown in Figure 10 for July 8 In this case we see a large increase in NOy about half of which is N02 MACR and MPAN after the in version breakup at about 0900 It is possible that MPAN is pro duced at night in a NOxrich residual layer as a result of N03 chemistry As discussed by Siam et al this issue although N03 radical chemistry cannot have a signi cant impact on MACR fractional decay at night the NO3MACR reaction can have a signi cant impact on the relative changes in the MPAN concentrations at night because of the generally low mixing ra tios for MPAN and much larger MACR mixing ratios 34 Contribution of Isoprene Chemistry to Ozone Production It is nowvwell known that isoprene can be an important reac tive hydrocarbon and can even contribute signi cantly to 03 pro duction in urban areas Chamez39des et al 1992 The current study affords the opportunity to examine the impact of urban in July 13 1995 Youth Inc 20 07 NOy 390 A NOz 06 a 16 l l Q l o soprene a MACRISOP 05 393 g c 5 12 l 9 a 04 8 g 8 I L03 3 r 39 Q 1 5 Ti 0392 a O 4 q 39Z 000 o A 039 V 0 o 00 3 01 39 O O 0 f 1 00 0 4 8 12 16 20 Time of Day Figure 8 Plot of MACRisoprene Noy No1 and isoprene for July 13 1995 22432 STARN ET AL OBSERVATIONS OF ISOPRENE CHEMISTRY July 20 1995 Youth Inc 20 H 08 NOy ll 1 390 NOZ g 6 43 Isoprene I I A 1 a MACRISOP 1A I 06 5 I ll 2 I 321 w o 12 39 1 I a 8 1 3951 O quot393 39 1 ii 04 ll a 392 i 1 A 5 39I t O O 39 l 391 lt Z 1 K 2 aquot l 1 02 39 04 z My 7 us 141 39 1 2K ng ngkfyy 0 L 4 8 12 16 Time of Day Figure 9 Plot of MACRisoprene NOy N02 and isoprene for July 20 1995 uences on isoprenegenerated ozone in a rural setting As dis cussed by Biesenthal et al 1997 this can be calculated in cases where the NOX levels are high enough that all isoprene derived peroxy radicals will react with NO rather than other peroxy radicals that is in relatively high NOx environments This is a very good assumption for the conditions prevalent at the Youth Inc site This calculation is done given a knowledge of the mechanism for isoprene oxidation and the relative yields of its oxidation products and ozone resulting from OH reaction For example OH reaction with isoprene produces MACR with a yield of 023 In this process 19 molecules of 03 are produced for each R02 that oxidizes NO to N02 2 molecules of 03 are produced however there is a 096 correction for the fact that the yield of organic nitrates from this R02 reaction with NO is 0044 Chen et 11 1997 Therefore as discussed in more detail by Biesenthal et al 1997 83 molecules of 03 are produced for each MACR produced from isoprene oxidation Thus when MACR is rapidly increasing ie production of MACR is much faster than destruction we can calculate how much of the corre sponding increase in 03 is produced from isoprene oxidation This method is applicable only when fast isoprene oxidation by OH is occurring ie in daytime so that for example ozonoly July 8 1995 Youth Inc D g 20 5 Si Noy 2 No i 6 x o MACRx1O 4 g MACRISOP v3 2 MPAN5 C a v D Q 12 E 39 a Z 8 I O O l 1 2 E 8 l lt 3 J v 39 1 E O 4 O o 75 1 Z a O o E 0 000000000 gt O 0 1 f 1 1 f 1 I 0 Z 6 8 1o 12 14 16 I8 20 1 Time of Day Figure 10 Plot of MACRisoprene NOy N02 MPAN and MACR for July 8 1995 STARN ET AL OBSERVATIONS OF ISOPRENE CHEMISTRY Youth no July 3 1995 024681012141618202224 07 i a A l A A l L so 06 80 39 o MACR 70 a 05 MPAN r 60 E Q A 04 E a 6 o3 40 lt O r 30 g 20 01 10 00 l o 80 r 18 ozone 39 7 N0y 3 6 quot I 8 o 5 12 Q 0 50 39 4o Q 39 8 E 30 I 6 20 i i a 39 4 10 2 O 1390 1392 1 4 1396 1398 20 22 24 Time of Day Figure 11 Plot of top MACR and MPAN and bottom ozone and NO for July 3 1995 sis of isoprene to produce MACR is much slower A very good execution of this calculation can be done for the July 3 1995 plume impaction case described above In Figure 11 we show the 03 NO MACR and MPAN the peroxyacyl nitrate produced from MACR oxidation data for this day As shown in the gure MACR increases by 035 ppb at the time of the plume impaction while 03 simultaneously increases by 104 ppb As described above we interpret the increase in MACR to processing of isoprene in the NOxrich plume as it is transported to the site Figure 11 indicated little shortterm vari ability in MACR before the plume impaction and we thus be lieve that the observed MACR and 03 levels were regionally rep resentative up to that time Thus for this plume we calculate that 035x83104 x 100 28 of the increase in ozone as the urban air mass advects to the measurement site results from iso prene chemistry This result is certainly not inconsistent with the results of Williams et al 1997 who calculate that on average about 25 of the ozone in the boundary layer in this region re sults from isoprene chemistry In this case we nd that even in the urban plumes that are advected over bordering forest regions isoprene chemistry is very important to 03 production We note that our calculation assumes that MACR production is much faster than MACR destruction in the air mass over the period of measured MACR increase that is that the ratio 023xlqon iso Immxisoprenekon MACRXMACR gt 1 Biesenthal et al 1997 For the plume studied this quantity is 075 However it is clear from the data shown in Figure 11 that since both MACR and MPAN increase dramatically and are highly correlated con siderable isoprene oxidation occurred upwind of the site and thus during this chemical processing of isoprene and concomi tant 03 production the isopreneMACR ratio was greater 22433 than that observed whenuthe plume impacted on the site In other words it is very likely that in the upwind plume MACR pro duction was much faster than MACR destruction That some of the MACR produced may have been consumed by OH reaction requires that this calculation provides a lower limit to the isoprene contribution to 03 production As discussed by Biesenthal et al 1997 it is also a lower limit because this calculation considers only the impact on ozone of the oxidation of isoprene itself and does not take into account the contribution from its oxidation products HCHO MVK and MACR As dis cussed above for Figure 5 it appears that HCHO is very impor tant For this site we nd that it is rare that conditions exist where there are substantial shortterm increases in MVK or MACR to enable this calculation For example and as described above during the episode wee of July 1216 much of the MVK and MACR increase at the ground each day resulted from downward mixing in the morning from more polluted layers aloft upon inversion layer breakup In this environment quite unlike the Lower Fraser Valley case discussed by Biesenthal et al 1997 the isoprene photooxidation products can recirculate in the relatively stagnant highpressure system and morning MVK or MACR concentrations and oxidation rates can be substan tial 39 35 Isoprene Oxidation as a NOx Sink Although it is well establiShed that isoprene chemistry can play a very iinportant role in regional oxidant production this same chemistry can act as a NOx sink and thus impact on the lifetime of NOx transported over isopreneimpacted continental regions This occurs through the production of organic nitrates the formation of which has been discussed in several publications Tuazon and Atkinson 1990 Shepson et al 1996 Chen et al 1998 The study of Chen et al 1998 indicates that organic nitrates are produced as a result of OH radical reaction with iso prene in the presence of NC with a 44 yield Our measure ment data for other isoprene products for example MACR along with knowledge of the relative product yields of these products allows us to estimate the concentration of the isoprene derived organic nitrates for conditions during which photooxi dation of isoprene is fast As an example organic nitrates are produced at a rate of 4423 that of MACR Although both prod ucts can be lost via OH reaction they are also both ole nic we estimate an OH radical rate constant of 13x1039ll cm3 moleculequot s 1 for a representative iSOprene nitrate Shepson et al 1996 while the OH rate constant for MACR is a comparable 31x10 ll cm3 molecule39l squot Thus we can estimate isoprene nitrates by multiplying the observed MACR by 4403 We have done this for the data shown in Figure 12 for the photochemically active period July 1216 1995 Also shown in this gure is the NOy data As shown in Figure 12 the calculated isoprene nitrate con centration is typically 005015 ppb during daytime that is comparable to the total organic nitrate concentration observed in other urbanrural environments F locke et al1991 O Brien et al 1995 1997 We also show in Figure 12 the ratio of isoprene nitratesN0y as a percentage that is the estimated percent contribution of isoprene nitrates to NO The daytime values range from about 052 Figure 12 also shows that the Calculated isoprene nitrate can be as high as 9 of NOy at night The signi cance of this is unclear given that we calculate the isoprene nitrate concentration asSuming it is all produced from daytime OHisoprene chemistry 22434 39STARN ET AL OBSERVATIONS OF ISOPRENE CHEMISTRY Youth nc July 1216 1995 30 NOy ppb 24 612182461218246121824 6 July 12 July 13 Time of Dayquot July 14 030 r 12 Q o 025 o 10 0 P 4 0 I 020 9 r 8 Z 3 m It 53 39E 9 1 quot 015 a 6 g C 139quot D C L a 010 8 4 c r I 9 03 8 005 2 9 000 L 0 12 18 24 NOy ltf estimated isoprene nitrates isoprene nitratesNOY Figure 12 Plot of the estimated isoprene nitrates isoprene nitratesN0y and NOy for July 1216 1995 4 Conclusibns There is some evidence from this study that elevated urban NOx concentrations that advect over forest environments can ac celerate isoprene oxidation and concomitant 03 production At other times the impact of the urban plume at this site which is close to the urban center results in suf ciently high NOx levels that OH radical oxidation chemistry is suppressed However we also observe that even in urban anthropogenic VOCrich plumes isoprene chemistry is an important contributor to 03 production The extent to which this urban areasource NOx contributes to 03 production downwind depends on the lifetime of that NOx This depends on its rate of conversion to oxidized nitrogen ie NOZ for example PAN HNO3 and isoprene nitrates and on the rate of deposition of those species As discussed in this paper we believe that the isoprene nitrates which have been observed in laboratory studies can be signi cant components of ambient N02 and it is thus important to better understand their atmos pheric fate Acknowledgments This work was supported by Purdue University the National Science Foundation grant ATM9520374 and an EPA subcon tract with Georgia Institute of Technology and was a part of the Southern Oxidants Study SOS a collaborative university government and private industry study to improve scienti c understanding of the accumulation and effects of photochemical oxidants We would like to thank the Jonathan Amy Facility for Chemical Instrumentation for design implementation and construction of the autosampler References Apel E et al Generation and validation of oxygenated volatile organic carbon standards for the 1995 SOS Nashville Intensive J Geophys Res this issue Biesenthal T A and P B Shepson Observations of anthropogenic inputs of the isoprene oxidation products methyl vinyl ketone and methacrolein to the atmosphere Geophys Res Lett 24 13751378 1997 Biesenthal T Q Wu PB Shepson HA Wiebe KG Anlauf and GI Mackay A study of relationships between isoprene its oxidation prod ucts and ozone in the Lower Frasier Valley Atmos Environ 31 2049 2058 1997 Chameides WL RW Lindsay J Richardson and OS Kiang The role of biogenic hydrocarbons in urban photochemical smog Atlanta as a case study Science 241 14731475 1988 Chameides WL et al Ozone precursor relationships in the ambient atmos phere J Geophys Res 97 60376055 1992 Chen X D Hulbert and P B Shepson A study of the production of or ganic nitrates from OH radical reaction with isoprene in the presence of NOJ Geophys Res in press 1998 Ciccioli P A Cecinato E Brancaleoni and M Frattoni Use of carbon ad sorption traps combined with high resolution gas chromatography Mass spectrometry for the analysis of polar and nonpolar C4C14 hydrocar bons involved in photochemical smog formation J High Resolut Chro matogr15 7584 1992 Flocke F A VolzThomas and D Kley Measurements of alkyl nitrates in rural and polluted air masses Atmos Environ Part A 25 19511960 1991 Kwok E S C S M Aschmann J Arey and R Atkinson Product forma tion from the reaction of the N03 radical with isoprene and rate constants for the reactions of methacrolein and methyl vinyl ketone with the N03 radical Int J Chem Kinet 28 925934 1996 Lin X M Trainer and S C Liu On the nonlinearity of the tropospheric ozone production J GeophysRes 93 1587915888 1988 Lin X OT Melo DR Hastie PB Shepson H Niki and JW Botten heim A case study of ozone production in a rural area of central Ontario Atmos Environ PartA 26 311324 1991 McNider R T W B Norris A J Song R L Clymer S Gupta R M Banta R J Zamora and M Trainer Meteorological conditions during the 1995 SOS NashvilleMiddle Tennessee Field Intensive J Geophys Res this issue Miyoshi A S Hatakeyama and N Washida OH radicalinitiated pho tooxidation of isoprene An estimate of global CO production J Geo phys Res 99 1877918787 1994 Montzka SA 39M Trainer PD Goldan WC Kuster and RC Fehsenfeld Is0prene and its oxidation products methyl vinyl ketone and meth acrolein in the rural atmosphere J Geophys Res 98 11011111 1993 Nouaime G S B Bertman C Seaver D Elyea H Huang P B Shepson T K Stam D D Riemer R G Zika and K Olszyna Sequential oxi dation products from tropospheric isoprene chemistry MACR and MPAN at a NOxrich forest environment in the southeastern United States J Geophys Res this issue STARN ET AL OBSERVATIONS OF ISOPRENE CHEMISTRY O Brien 1 P B Shepson K Muthuramu C Hao H Niki and D R Hasstie Measurements of alkyl and multi mctional organic nitrates at a rural site inOntario J Geophys Res 100 2275922804 1995 39 O39Brien 1 M P B Shepson T Biesenthal Q Wu J W Bottenheim H A Wiebe K G Anlauf and P Brickell Production and distribution of organic nitrates and their relationship to carbonyl compounds in the Lower Fraser Valley EC Atmos Environ31 20592069 1997 Shepson P B E Mackay and K Muthuramu Henry s law constants and removal processes for several atmospheric Bhydroxy alkyl nitrates En viron Sci T 65711101 30 36183623 1996 Sirju AP and RB Shepson Laboratory and eld investigation of the DNPH cartridge technique for the measurement of atmospheric carbonyl compounds Environ Sci Technol 29 384392 1995 Starn TK PB Shepson S B Bertman D D Riemer RG Zika and K Olszyna Nighttime isoprene chemistry at an urbanimpacted forest site J Geophys Res this issue 39 Trainer M E Y Hsie S A McKeen R Tallamraju D D Parrish F C A Fehsenfeld and S C Liu Impact of Natural Hydrocarbons on Hydroxyl and Peroxy Radicals at a Remote Site J Geophys Res 92 118975 1 1894 1987 39 Tuazon EC and R Atkinson A product study of the gasphase reaction of isoprcne with the OH radical in the presence of NOX Int J Chem Kinet 22 12211236 1990 22435 Williams 1 et al Regional ozone from biogenic hydrocarbons deduced from airborne measurements of PAN PPN and MPAN Geophys Res Lett 24 10991102 1997 Yokouchi Y and M Sano Trace determination of volatile organic com pounds in soil based on thermal vaporization followed by TenaxGC39 trapping and capillary gas chromatographymass spectrometry J Chro matogr 555 297301 199 1 Yokouchi Y Y Ambe and T Maeda Automated analysis of CBC13 hy drocarbons in the atmosphere by capillary gas chromatography with a cryogenic preconcentration Anal S011 2 571575 1986 S B Bertman Chemistry Department Western Michigan University Kalamazoo MI 49001 K Olszyna Atmospheric Sciences Department Tennessee Valley Authority Muscle Shoals AL 35660 D D Riemer and R G Zika Rosenstiel School of Marine and Atmos pheric Science University of Miami Miami FL 33124 P B Shepson B G Splawn T K Stam and J S White Department of Chemistry Purdue University 1393 Brown Building West Lafayette IN 47907 email pshepsonchempurdueedit Received August 27 1997 revised April 9 1998 accepted April 13 1998 mo 0va TABLE 3 I Actinic n VIII 11 A1 21 Ah Eath39w Sur n i7 FmIIm m VquIgIh Imuval and wIIQIInuIIIIImI 5mm mmg amw II Ewanv III I an III so 70 7E K6 I I I I I I I 1 7 2 2 2 I I I I 7 z z 2 I I I 7 2 2 2 2 I I I 7 a 1 I I I 7 2 2 2 I I 7 2 2 v 2 I I I r 2 V 2 I I III 39m mullmn In WW Iquot a gm Mmm I Wmme WWW mm minImam quotm mIIIIIIII Vim In vuwar of III III Wm AH mm would be InuIIIIIIm m examplzI II II 7 MIIIMIII IIIIemI me me III we I Is I oz x InH photons cm 4 39Im IIIIII Mm Iqu II III