Estimation of...Location Data_Qi LI.pdf

Estimation of Lightning-Generated NOx in the Mainland of ChinaBased on Cloud-to-Ground Lightning Location DataQi LI1,2,Fengxia GUO*1,Xiaoyu JU3,Ze LIU1,Mingjun GAN1,Kun ZHANG1,and Binbin CAI11Key Laboratory of Meteorological Disaster,Ministry of Education(KLME)/Joint International Research Laboratory ofClimate and Environment Change(ILCEC)/Collaborative Innovation Center on Forecast and Evaluation ofMeteorological Disasters(CIC-FEMD)/Key Laboratory for Aerosol-Cloud-Precipitation of China MeteorologicalAdministration,Nanjing University of Information Science&Technology,Nanjing 210044,China2Huangshi Meteorological Bureau,Huangshi 435002,China3Anhui meteorological disaster prevention center,Hefei 230061,China(Received 18 August 2021;revised 19 April 2022;accepted 19 May 2022)ABSTRACTLightning-generated nitrogen oxides(LNOx)have a major influence on the atmosphere and global climate change.Therefore,it is of great importance to obtain a more accurate estimation of LNOx.The aim of this study is to provide areference for the accurate estimation of the total LNOx in the mainland of China based on cloud-to-ground lightning(CG)location data from 2014 to 2018.The energy of each CG flash was based on the number of return strokes per CG flash,thepeak current of each return stroke,and the assumed CG breakdown voltage.The energy of intracloud lightning(IC)wasbased on the estimated frequencies of IC and the assumed energy of each IC flash.Combining the energy of lightning andthe number of nitric oxide(NO)molecules produced by unit energy(no),the total LNOx production in the mainland ofChina was determined.The LNOx in the mainland of China estimated in this study is in the range(0.1570.321)109 kg peryear Tg(N)yr1,which is on the high end of other scholars works.Negative cloud-to-ground lightning(NCG)flashesproduce the most moles of NOx,while positive cloud-to-ground lightning(PCG)flashes produce the least total moles ofNOx.The breakdown voltage of PCG is greater than that of IC or NCG,while the latter has a greater output of LNOx.Key words:cloud-to-ground lightning location,lightning peak current,lightning breakdown voltage,nitrogen oxide(NOx)Citation:Li,Q.,F.X.Guo,X.Y.Ju,Z.Liu,M.J.Gan,K.Zhang,and B.B.Cai,2023:Estimation of lightning-generatedNOx in the mainland of China based on cloud-to-ground lightning location data.Adv.Atmos.Sci.,40(1),129143,https:/doi.org/10.1007/s00376-022-1329-6.Article Highlights:Based on cloud-to-ground lightning(CG)location data,the energy of each CG flash and the frequencies of intracloudlightning(IC)were estimated.The lightning-generated nitrogen oxides(LNOx)estimation for the mainland of China is between 0.157 Tg(N)yr1 and0.321 Tg(N)yr1,which is on the high end of other studies.Compared to positive cloud-to-ground lightning,intracloud lightning and negative cloud-to-ground lightning havegreater output of lightning-generated nitrogen oxides.1.IntroductionNitrogen oxides (NOx)are important to troposphericchemistry due to their role as a precursor of ozone (O3)(Crutzen,1974).It should be noted that NO2 is the chemicalexpression for nitrogen dioxide.In addition,NOx can controlconcentrations of tropospheric hydroxyl radical (OH),thereby partially controlling the level of oxidants.NOx canarise from both anthropogenic and natural emission sources.The anthropogenic sources of NOx are largely known.How-ever,lightning-generated nitrogen oxides (LNOx)are stillthe source with the greatest uncertainty.In the troposphere,lightning is the largest naturalsource of NOx(Bucsela et al.,2010);however,estimates ofLNOx and its efficiency(LNOx per flash)are still quite uncer-tain(Beirle et al.,2010).Many researchers have conductedstudies focused on improving estimates of the total amount*Corresponding author:Fengxia GUOEmail:ADVANCES IN ATMOSPHERIC SCIENCES,VOL.40,JANUARY 2023,129143 Original Paper Institute of Atmospheric Physics/Chinese Academy of Sciences,and Science Press and Springer-Verlag GmbH Germany,part of Springer Nature 2023of LNOx(mainly LNO).By measuring the NOx generated inlaboratory long-gap discharge experiments,researchershave been able to estimate LNOx and extrapolate globalLNOx(Peyrous and Lapeyre,1982;Wang et al.,1998;Cooray et al.,2009).There are several problems with thismethod.First,the peak current and the energy of laboratorydischarges are much smaller than those of natural lightning.Second,water vapor,relative humidity,and other environmen-tal parameters also influence lightning development,but inlaboratory long-gap discharge experiments,these influencesare usually neglected.These factors bring great uncertaintyto the estimation of LNOx,and the results obtained by differ-ent researchers vary greatly.Most estimates of global LNOxproduction range from 28 Tg(N)yr1(Schumann andHuntrieser,2007).At present,there are few studies on LNOx estimationfor the mainland of China.Zhou and Qie(2002)extrapolatedthe total LNOx emissions in the mainland of China to be0.384 Tg(N)yr1 using the cloud-to-ground lightning(CG)location data from Guangdong,Longdong,Beijing,and north-east China.The energy of a CG flash was assumed to be6.7 109 J in the theoretical extrapolation calculation pro-posed by Price et al.(1997).However,Wang et al.(1998)sug-gested that this value was the upper limit of the energy gener-ated by a single lightning flash,which might contribute tooverestimation of the overall result.Furthermore,the light-ning location data did not cover the whole of China,and thelightning data in the four regions may not be sufficiently rep-resentative for the whole country.Zhang et al.(2002)andDu et al.(2002)used the CG location data in northeastChina,Beijing,and Guangdong to estimate the energy ofeach lightning flash.Then,using the positive cloud-to-ground lightning(PCG)and negative cloud-to-ground light-ning(NCG)frequencies,they estimated the LNOx productionin the three areas.Compared with Zhou and Qie(2002),Zhang et al.(2002)and Du et al.(2002)obtained more param-eters by observations.However,the CG data only includedthree local regions of China.In addition,only the amount ofCG-generated NOx was estimated,while the intracloud(IC)-generated NOx was ignored.Zhou et al.(2003)analyzed the NOx generated by naturallightning,which was observed by a ground-based NOx ana-lyzer in Datong county of Qinghai province.They estimatedthe energy of a lightning flash as(1213)106 J.In theirresearch,only part of air entering the NOx analyzer is hot air,that is,a considerable part is ambient air,so the calculationof LNOx has very high uncertainty.In addition,the heightof thunderstorms in the QinghaiTibet Plateau is lower andthe channel length of a lightning flash is shorter,so theLNOx amount of a lightning flash is smaller than in othersareas.Based on the total lightning data obtained by OpticalTransient Detectors(OTD),assuming the energy of a light-ning flash to be 4.5 107 J,Sun et al.(2004)estimated thatthe total LNOx in China and the adjacent area is only 0.016Tg(N)yr1,an order of magnitude less than the estimate ofZhou and Qie(2002).The lightning light energy measuredby OTD is the lightning radiation light energy at the top ofthe cloud,which is absorbed and scattered by the cloud,sothe measured value is smaller than the actual value.SinceSun et al.(2004)neglected the reduction of optical energycaused by the optical thickness of clouds and the inversionfrom the optical energy to the total energy,they may haveunderestimated the energy of a single lightning flash,thuscausing underestimation of the total NOx.Based on the data obtained by Lightning Imaging Sensor(LIS)and Global Ozone Monitoring Experiment-2(GOME-2),Ju et al.(2015)analyzed the correlation between lightningand NO2 vertical concentration column (NO2VCD)dataover the central Tibetan Plateau.They found that the linear fit-ting correlation coefficient between lightning density andNO2VCD was 0.84.Based on the fitting results,the totalamount of LNOx in the mainland of China was estimated tobe 0.15(0.030.38)Tg(N)yr1.Guo et al.(2016),using a sim-ilar method,obtained a mean LNOx production efficiency of330 mol flash1.Guo et al.(2017)analyzed the total lightningdata from OTD and LIS and the NO2VCD data from theRoyal Meteorological Institute of Holland.They proposedthat because human activity is minimal on the QinghaiTibetan Plateau,lightning is the main source of NOx anddetermines the NOx concentration in this area,and thus estab-lished a fitting relationship between the NO2VCD and light-ning on the QinghaiTibetan Plateau.The amount of NOx pro-duced by a single lightning flash can be derived from the fit-ting relationship.The total amount of LNOx in the mainlandof China was estimated to be 0.07(0.020.27)Tg(N)yr1.This method did not completely exclude the NOx generatedby other sources,such as soil emissions.In addition,because of the characteristic differences of lightning in differ-ent areas,using the amount of LNOx in the QinghaiTibetanPlateau for extrapolation to the whole mainland of Chinamay cause some error.Moreover,the lightning satellite dataused did not distinguish between different types of lightn-ing;the IC to CG ratio in the QinghaiTibetan Plateau maybe different than in other parts of China,which also addsuncertainty to the results.In summary,different researchers have adopted variousmethods and parameters or parameter values in the estimationof LNOx in China,leading to considerable variation in theirestimation results.It is therefore necessary to further discussthe methods of estimation.Based on CG location data forthe mainland of China,the energy of each CG flash was calcu-lated from the number of return strokes per CG flash,thepeak current of each return stroke,and the assumed CG break-down voltage.The total energy of intracloud lightning(IC)is calculated using the estimated frequency of IC flashesand the assumed energy of a single IC flash.Consideringthe total energy of lightning and the number of nitric oxide(NO)molecules produced by unit energy (no),the totalLNOx production in the mainland of China was determined.The data provided can be used as a reference for more accurateestimation of the total LNOx in China.130ESTIMATION OF LIGHTNING-GENERATED NOx IN CHINAVOLUME 402.Mechanism of LNOx productionWhen lightning occurs,the air in and around the lightningchannel is instantaneously heated as high as 30 000 K ormore(Uman et al.,1964),and the pressure can reach severalto dozens of atmospheres(Hill,1971).During this process,O2 and N2 in the air are completely ionized and a series ofchemical reactions continue to take place,resulting in prod-ucts dominated by NOx.After discharge,the channel coolsrapidly.The reactions begin with the thermal decompositionof O2:O2 O+O,(1)followed by the production of NO via the reaction chain:O+N2 NO+N,(2)andO2+N NO+O.(3)Meanwhile,NO consumption occurs as follows:NO+N N2+O,(4)andNO+O N+O2,(5)where NO is converted back to O2 and N2.NO itself decom-poses as follows:NO N+O.(6)The NO will also transform into N2O as follows:NO+NO N2O+O.(7)The previous seven equations were taken from Borucki andChameides(1984).In addition,there are quick photochemi-cal/chemical inter-conversions between NO and NO2 in thefollowing reactions (Levine et al.,1984;Verma et al.,2021):NO+O3 NO2+O2,(8)O2+O+M O3+M.(9)Here,M in the equation represents the third body to receiveextra energy.NO2 can be rapidly photolyzed back to NO dur-ing the daytime:NO2+O2+h NO+O3(k 420 nm).(10)Here,h is planck constant and is frequency of light waves.It has been suggested that in an electrical discharge,NO ismainly produced by a series of high temperature reactions(Cooray et al.,2009).The experimentally determined NO/NOx ratios vary considerably depending on the presence ofexcess O2,O3 and the residence time(Cooray et al.,2009).However,the total number of molecules of NOx is equal tothat produced by NO(Cooray et al.,2009).Therefore,wefirst estimated the LNO production,followed by LNOx.The time taken for NO to establish thermochemical equi-librium increases as the temperature decreases.It takes onlya few microseconds for NO to reach equilibrium at 4000 K,but approximately 1000 years at 1000 K.Thus,as the light-ning-heated air cools,the“freeze-out”temperature isreached,below which the reactions that produce and destroyNO become too slow to maintain NO in equilibrium,andthe NO density remains at the value characteristic of equilib-rium at that temperature.The rate of NO production at thattemperature can then be used to estimate the LNO yield.Each lightning process contributes a different amountof NO.Return strokes generate the most NO during a CGflash because most air molecules are ionized in the returnstroke process and NO-dominated chemical substances areproduced(Sisterson and Liaw,1990).The NO produced dur-ing corona discharge is several orders of magnitude lessthan that produced during the return strokes(Sisterson andLiaw,1990).IC flashes generally begin with streamers thatpropagate continuously.When the streamer encounters acharge source with opposite polarity,a recoil streameroccurs.A recoil streamer is similar to the return stroke ofCG,and the accompanying electric field it causes is knownas a K-change.The average current of a K-change is 14 kA,and each IC flash has about 6 K-changes(Ogawa andBrook,1964).During each K-change,the average chargetransferred is about 1.4 C(Brook and Ogawa(1977).Theaverage charge of each IC flash during K-changes is about8.4 C.Most NO is produced during the K-change stage ofIC flashes.3.Data and methodThe China automatic CG monitoring system began opera-tion in the 1980s.Initially,it mainly imported foreign CGlocation equipment(Chen et al.,2008).By 1990,a domesticCG location system had been put into operation(Chen et al.,2008).The domestic CG location system is mainly used forlocal CG monitoring and has not yet covered the whole coun-try.It is used in the fields of meteorology,electric power,telecommunications,civil aviation,and military.From 2009to 2017,the number of sensors managed by the China Meteo-rological Administration(CMA)increased from 265 to 406,with the largest increases occurring in Qinghai,Inner Mongo-lia,Tibet,and Xinjiang,where 24,29,24,and 43 stationswere added,respectively (Fig.1)(CMA,2010,2018).Itshould be noted that the Xinjiang stations were installed in2013,while most of the Qinghai and Inner Mongolia stationswere installed in 2016.In addition,as seen in Fig.2 aredrawn figure from Zhu(2016),the sensors cover a largeportion of China;however,there are more sensors in theeast than the west,and more in the south than the north.Assoon as a regional lightning network was established,theCMA began to digitize CG lightning strokes on a dailybasis(Xia et al.,2015)and provide access to the data.JANUARY 2023LI ET AL.131The ground-based Advanced Time of Arrival and Direc-tion(ADTD)system,used in the CMA CG monitoring net-work,was developed by the National Space Science Center,the Chinese Academy of Sciences.This system uses theTime of Arrival(