atmosphere Article Seasonal Changes in Urban PM2.5 Hotspots and Sources from Low-Cost Sensors Lorenz Harr 1,* , Tim Sinsel 1 , Helge Simon 1 and Jan Esper 1,2 1 Department of Geography, Johannes Gutenberg-University, Johann-Joachim-Becher-Weg 21, 55128 Mainz, Germany; t.sinsel@geo.uni-mainz.de (T.S.); h.simon@geo.uni-mainz.de (H.S.); esper@uni-mainz.de (J.E.) 2 Global Change Research Institute of the Czech Academy of Sciences (CzechGlobe), 60300 Brno, Czech Republic * Correspondence: l.harr@geo.uni-mainz.de; Tel.: +49-6131-39-29-803 Abstract: PM2.5 concentrations in urban areas are highly variable, both spatially and seasonally. To assess these patterns and the underlying sources, we conducted PM2.5 exposure measurements at the adult breath level (1.6 m) along three ~5 km routes in urban districts of Mainz (Germany) using portable low-cost Alphasense OPC-N3 sensors. The survey took place on five consecutive days including four runs each day (38 in total) in September 2020 and March 2021. While the between- sensor accuracy was tested to be good (R2 = 0.98), the recorded PM2.5 values underestimated the official measurement station data by up to 25 µg/m3. The collected data showed no consistent PM2.5 hotspots between September and March. Whereas during the fall, the pedestrian and park areas appeared as hotspots in >60% of the runs, construction sites and a bridge with high traffic intensity stuck out in spring. We considered PM2.5/PM10 ratios to assign anthropogenic emission sources with high apportionment of PM2.5 in PM10 (>0.6), except for the parks (0.24) where fine particles likely originated from unpaved surfaces. The spatial PM2.5 apportionment in PM10 increased from September (0.56) to March (0.76) because of a pronounced cooler thermal inversion accumulating fine Citation: Harr, L.; Sinsel, T.; Simon, particles near ground. Our results showed that highly resolved low-cost measurements can help to H.; Esper, J. Seasonal Changes in identify PM2.5 hotspots and be used to differentiate types of particle sources via PM2.5/PM10 ratios. Urban PM2.5 Hotspots and Sources from Low-Cost Sensors. Atmosphere Keywords: OPC-N3; particulate matter; personal exposure; mobile measurement; PM2.5/PM10 ratio 2022, 13, 694. https://doi.org/ 10.3390/atmos13050694 Academic Editors: Esther Hontañón and Brigida Alfano 1. Introduction The global perception of air quality and air pollutants such as particulate matter (PM) Received: 9 March 2022 Accepted: 25 April 2022 has increased partly due to the COVID-19 pandemic [1,2]. While coarse particles with an Published: 27 April 2022 aerodynamic diameter between 2.5 and 10 µm (PM2.5–10) are inhalable, fine particles with diameters <2.5 µm can reach the bronchial system and cause airway inflammation, lung Publisher’s Note: MDPI stays neutral disfunction, and chronic obstructive pulmonary disease [3–5]. with regard to jurisdictional claims in However, the toxicity of particles is not only determined by their absolute concentra- published maps and institutional affil- tion but also varies between different types of PM, e.g., metallic elements of residual oil fly iations. ash have more adverse health effects than biogenic or inorganic components [6–8]. PM ele- ments can be detected via chemical analyses, though in the absence of these measures, the origin of particles can be attributed by calculating the ratio of PM2.5/PM10 [9,10]. Whereas Copyright: © 2022 by the authors. a weighting towards PM2.5 indicates emissions from combustion processes, i.e., vehicle Licensee MDPI, Basel, Switzerland. exhausts and house heating, a low ratio indicates natural emissions as sources, i.e., pollen This article is an open access article and leaf particles and/or fugitive or re-suspended road dust from tire and break abrasion, distributed under the terms and for instance [7,9,11,12]. conditions of the Creative Commons In urban areas, PM2.5/PM10 ratios can rapidly change over time due to short-term Attribution (CC BY) license (https:// variation in emission intensity, e.g., rush hour or non-rush hour, but also in response to creativecommons.org/licenses/by/ changing weather situations. Stationary anti-cyclonic weather in Central Europe is asso- 4.0/). ciated with low wind speeds and limited precipitation [13,14] as well as, particularly in Atmosphere 2022, 13, 694. https://doi.org/10.3390/atmos13050694 https://www.mdpi.com/journal/atmosphere Atmosphere 2022, 13, 694 2 of 14 autumn and winter, high convection inhibitions (CIN) causing low mixing layer heights (MLH). The vertical air exchange is thus reduced, leading to an accumulation of fine parti- cles near ground and a high PM2.5 apportionment > 60% relative to PM10, which is generally in contrast to PM2.5/PM10 ratios < 0.5 typically recorded in spring and summer [15–17]. To monitor the seasonal variability of the PM2.5 and PM10, 30 to 60 min mean data are provided by the official stationary measurement networks in Europe [18]. However, highly temporal changes < 30 min cannot be detected, and more importantly, spatial variability of particle concentrations and their sources cannot be represented due to the immobility of permanent network facilities. Spatiotemporal differences in personal exposure can therefore not be represented. In contrast, mobile measurements provide the possibility to extend the spatial coverage of stationary measurements, particularly at the pedestrian breath level [19]. A cost-effective solution for mobile measurements is the use of so-called low-cost monitoring systems [20]. These devices are also highly portable due to their small weight and size and can be easily mounted on vehicles or racks carried by a person [21]. We used Alphasense OPC-N3 sensors [22], demonstrated to perform well under laboratory conditions [23,24], to measure different types of particles at high temporal resolution of 1 s. However, in urban outdoor environments, the accuracy of these data is adversely affected by changes in particle composition and relative humidity (RH) [25–27]. The goal of this study was to demonstrate seasonal and spatial variability of PM2.5 concentrations in a Central European city (Mainz, Germany) using mobile low-cost in- struments at high spatiotemporal resolution. We (i) compared these measurements with long-term stationary data, (ii) identified PM2.5 hotspots and their source, and (iii) inves- tigated seasonal changes in source regimes throughout the study area. We expected to find (i) similar peak PM2.5 values in March and September, (ii) highest polluted locations nearby streets with high traffic intensity and close to anthropogenic sources, and (iii) higher PM2.5/PM10 ratios in spring than in late summer due to prevailing anti cyclonic weather regimes in the colder season. 2. Materials and Methods 2.1. Study Sites and Sensors The study was conducted on five consecutive weekdays in September (14–18 Septem- ber 2020) and March (1–5 March 2021) in Mainz, the capital and largest city (approx. 220,000 habitants) of Rhineland-Palatinate in south-west Germany (50.0◦ N, 8.26◦ E, Figure 1). Located in a slightly hilly landscape along the river Rhine, Mainz is an inland town and one of the cities with the highest PM concentrations in Germany [28]. The climate is moderate with an annual average temperature of 10.7 ◦C and precipitation of 620 mm (Koeppen Cfb) [29,30]. The study route includes three urban quarters of different characteristics: Altstadt, Hartenberg, and Neustadt (Figure 1). The Altstadt quarter is the old part of the town characterized by compact low- to midrise buildings, mostly paved streets, and pedestrian zones [31]. The urban architecture of the Hartenberg quarter, on the contrary, is a dis- trict with open low- to midrise buildings, a small grove, and low motorized traffic. The Neustadt quarter is characterized by mainly five-story-high buildings and narrow streets (~10 m wide), small parks (<150 m across), and low traffic intensity in a grid-based street layout. Large multi-lane roads with high traffic intensities surround this quarter as well as the Altstadt. Atmosphere 2022, 13, 694 3 of 14 Atmosphere 2022, 13, x FOR PEER REVIEW 3 of 13 Figure 1. Study tracks in the Mainz Altstadt, Hartenberg, and Neustadt (black, orange, and blue lFinigeus,r ree1s.pSetcutidvyeltyra),c tkhseiinr jtohienMt satainrtz aAnldts etanddt ,pHoianrtte antb tehreg m, aanidn Nsteautisotna d(mt (abglaecnkt,ao),r aanngde t,haen md bolnuietolrininegs, srietseps eocft ivtheely )Z,ItMheEirNj oninettwstoarrkt aantd Menaidnzp-oPianrtcautstshtreamßea i(ndsatrakt iroend()m aangde nMtaa)i,nazn-dZitthaedemlloen (irteodri)n ogfs itthees ZoIfMthEeNZ InMetEwNornke, tawnodr mk aatpM ofa Ginezr-mPaarncyu ssshtorawßieng(d tahrek lorecdat)iaond ofM Maianinzz-Z (iotraadnegllee).( red) of the ZIMEN network, and map of Germany showing the location of Mainz (orange). The total length of the study route was ~15 km or ~3 h walking by foot. To mitigate potenTthiael tcohtaanl gleensg othf loofcathl ecostnucdenytrroautitoenws dasu~ri1n5g ksmucohr a~ l3ohngw taimlkien sgpbayn,f owoet .dTiovimdeitdig tahtee rpooutteen itniatol cthhareneg ceisrcoufllaorc tarlaccokns,c eeancthra lteiaodnisndg uthrirnogugsuhc ohnae olof nthget idmisetrsipctasn. E, waceh dtriavcidk ewdatsh 5e kromu tleoningt o(~t1h hre beyc ifrocoutl)a arntrda cskhsa,reedac thhele saadmineg sttharrotiunggh anonde eonfdtihneg dpiositnritc atst. tEhaec Mh atrinaczk trwaians s5taktmionlo (n5g0.(0~0117h°◦ bNy, f8o.o2t5)9a5n° d◦E; s hFaigreudret h1 emsaagmeenstata drtoint)g. Tahned deinvdisiniognp inoitnot daitsttrhiectM traaicnkzs tarlasion ssutaptpioonrt(e5d0 .m00u1l7tipNle , m8.e2a5s9u5remE;eFnitgsu preer1 dmaya.g Weneta cdonodt)u. cTtehde fdoiuvris imoneaisnutroemdiesntrti crtutnrsa cokns eaalscoh sturapcpko, rbteefdorme ualntidp ldeumrienags uthreem meonrtnsipnegr adnady .aWfteerncooonnd urcutsehd hfoouurrsm steaarstuinrgem ate n6t ar.umn.s, 0o7n:3e0a cah.mtr.,a 4ck p,.bme.f oarneda 0n5d:3d0u pr.imng., twheitmh tohren ienxgceapntdioanf toefr n1o4 oSneprutesmh bheoru 2r0s2s0ta wrthinegn awt e6 oan.mly., m07e:3a0suar.emd. ,in4 tph.em a.fatenrdno0o5n:3.0 Fpo.rm e.a,cwh ittrhacthke, oenxec edpetvioicne owfa1s4 uSseepdt ecmombeprri2s0e2d0 owf ah ePnMw seenosnolyr Amlepahsausreendsien OthPeCa-Nfte3r [n2o2o],n a. FEoSrPe3a2c chotnrtarcokl,leorn [e32d]e, vai cGePwS ams oudseudlec [o3m3]p, rainsedd ao mf aicProMSDse cnasrodr tAol psahvaes emnseeasOuPreCm-Nen3t[ d22a]t,aa (FEiSgPu3r2e c2oan).t rTohllee rse[3n2s]o,ras GwPeSrem mooduunlete[d3 3a]t, aanddulat bmriecartohS Dhecigahrdt (t1o.6s amve) mone athsue rfermonetn todf aat aw(eFairgaubrlee 2raa)c.kT thoe rseednusocer sinwfelureenmceosu noft etdhea tpaedrsuoltnb craerartyhinhge itghhet d(1e.v6imce) o(Fnigthuerefr o2nbt).o Tf ao wsueaprpaobrlte rthacek dtoetreecdtiuocne ionffl luoecnacl eesmofitttheersp edrusroinngca prroysitn-gprtohceedevice(Figure 2b). To support the detection of local emitters during post-processing, eversysirnugn, ewvaesryfi lrmune dwwasit fhilamceadm weritaha att caacmheedrat oattthaechraecdk t.o the rack. Atmosphere 2022, 13, x FOR PEER REVIEW 4 of 13 Atmosphere 2022, 13, 694 4 of 14 Figure 2. (a) DesFiiggunr ea2n. d(a )cDoemsigpnoanndencotms poofn ethntes omf teheasmuearseumremenentt ddeevviciece(d (imdeimnsieonnss: i1o1n.5sc:m 1×1.154 ccm× × 14 cm × 12.5 cm), and (b) pic1t2u.5recm o),fa and p(be)rpsioctnur ceaofrarypienrsgon tcharer yriangckth.e rack. 2.2. Inter-Sensor Variability 2.2. Inter-Sensor VaTrhieaAbliplhitayse nse OPC-N3 sensors are low-cost optical particle counters following a light scattering principle [34]. The detected particles are put into bins considering their estimated The Alphsaizsee[n35s]ea nOdPsuCbs-eNqu3e nstelyncsoonrvser taerdei nltoowma-scsocsotn coepnttriactiaonl sp[3a6r]t. iTchlee mcoeausunrteemresn tfollowing a light scatteringra npgreionfctihpelAel p[h3a4s]e.n sTehOeP Cd-Net3efcotrepdar tpicalerstiiscl0e.3s5 atore40 pµumt[ 2i2n].toT hbeinhasn dcyoOnPsCid- ering their estimated sizeN [335un] itasnadre ssuuitbabsleeqfoureanmtloyb ilceomnevaseurrteemde nitnrtaock ,mafafosrsd acbolen(c~e300 €), and performwell under laboratory conditions [37] considering the European ENn4t8r1asttiaonndasr d[3a6nd]. The meas- urement rangem aonfu ftahcteu rAe claplibhratsioenn[s2e6] .OHPowCe-vNer,3to ffourt hperaarstsiecssleacsc uisra c0y.3an5d taod d4re0s sµinmter -[s2en2s]o.r The handy OPC-N3 unitsv aarriaeb isliutyi,taabstlaeti ofnoarry afi emldocbaliiblera tmioneainsuanreenmvieronntm reanct ksim, ailfafrotordthaebsltued (y~a3r0ea0i s€), and per- recommended [19,20,38–41]. Such a calibration was conducted on the Hartenberg district form well undfreorm l1a8–b2o2rNaotvoermyb ecr o20n20d, i5t–i8oJannsu a[r3y720]2 1c,oands2id0–e2r3iFnebgr utahrye2 0E21u.rSoinpceetahenr eEwNere 4no81 standard and manufacturerfeer ecnacelidbevriacteisothna t[f2ea6tu].r eHa coowmpeavraebrle, tteom pfourralthreesorl uatisosne( 60% according to Crilley et al. [26] was applied. The correction formula is based on the κ-Köhler theory, with κ = 0.33 as a composition of hygroscopic particles in the ambient air and a dry particle density of 1.65 g/cm3 [39]. Ambient RH measurements were taken from the long-term station Mainz-Zitadelle of the ZIMEN network. The processed data were then analyzed using descriptive statistics, i.e., arithmetic mean, median, and standard deviation (SD). In order to validate our absolute PM2.5 measurements for PM2.5 hotspot identification, a comparison of the mean PM2.5 values of each track and run in September and March against the regular long-term station data from Mainz-Parcusstraße, which is characterized by urban traffic, and Mainz Zitadelle, which resembles the urban background, was performed. These two measurement stations are part of the ZIMEN network, which carries out measurements with Thermo Fisher SHARP 5030 instruments [43] to monitor PM2.5 and PM10 on behalf of the state. For the detection of highly polluted spots, the following steps were conducted. To counteract time-related fine particulate gradients, the data of each run were linearly de- trended. Subsequently, the measurements of the simultaneously conducted runs in each district were combined and highly polluted locations (spots with 10% highest PM2.5 values) were identified: we determined highly polluted locations, for each period and season, by overlaying the extreme data of the respective runs and looking for matches. A match was recorded if the same location within a radius of 20 m indicated a pollution hot spot (i.e., 10% highest values) in several runs. After identifying highly polluted locations, we calculated PM2.5/PM10 ratios for the September and March data to evaluate emission sources. Atmosphere 2022, 13, 694 6 of 14 3. Results and Discussion 3.1. Absolute PM2.5 Concentrations in September and March In September, the uncorrected mean PM2.5 concentrations were in line with the ZI- MEN measurements and showed a diurnal pattern in PM2.5 characterized by 50 to 220% higher concentrations in the morning compared to the afternoon runs. This pattern was recorded during the first three days of the September campaign and followed by declining Atmosphere 2022, 13, x FOR PEER REVcIoEnWc entrations toward the end of the week (Figure 4, for median PM10 concentration6 so,fs e13e Figure S4). Figure 4. Mean PM2.5 concentrations in the Altstadt, Hartenberg, Neustadt (black, blue, and orange cFoilgourrse, r4e.spMecetaivnePlyM), 2a.5ndco ZnIcMenEtNra tdioantas firnomth ethAe lMtsataindzt,-PHaarcrutesnsbtrearßge, aNnedu Mstaadintz(-bZlaitcakd,ebllleu (ed, aarnkd reodr- aanndg erecdo lcoorlso,rsr,e srepsepceticvtievlyel)y, )a dnudriZnIgM thEeN stduadtya pfreormiodtsh einM (aa)i nSzep-Pteamrcbuesrs tarnaßde (ba)n dMaMrcahin wz-iZthi tcaodrerlele- s(pdoanrkdirnegd baonxdprloetds caonldo rasr,itrhesmpeetcicti vmeelya)nsd u(yreinllgowth deosttsu)d. Tyhpee urinofdilsleidn d(ao)tsS esypmtebmobliezre atnhde h(bu)mMidairtcyh- cwoirtrhectoerdr ePsMpo2.n5 mdienagsuborexmpleonttssa, anndda trhiteh fmilleetdic dmotesa snysm(ybeolliozwe PdMot2s.5) .coTnhceenutnrfiatllieodnsd wotisthsoyumt bhoulmiziedtihtye chourmreicdtiotyn-.c orrected PM2.5 measurements, and the filled dots symbolize PM2.5 concentrations without humidity correction. This change in PM2.5 variability could be associated to changes in the weather regime: the fiTrsht itshcrheaen dgaeyisn wPeMre2 .5chvaarraiacbteilriitzyecdo buyld wbaerams sloacteia steudmtmo echr awnegaetshienr tchoenwdietaiothnesr croengsiimste-: itnhge ofifr shtitghhr edeadilayy ms waxeirme uchma raaicrt teermizepderbaytuwreasr m(TAla)te >s3u0m °Cm, emr owdeeartahteer mcoenadni tRioHn s< c5o7n%si,s atinndg loofwh imghaxdimailuymm waxinimduspmeeadirs tm3 s0o◦uCth, meaosdt deriaretectmioenasn (0R; HFig12 µg/m3 higher concentrations on average, though this seemed to be a single outlier that we could not explain. After 3 March 2021, the PM2.5 concentrations decreased due to a change in weather, upcoming north wind (max. 3.2 m/s), and a short- term shower, followed by decreasing of TA and RH. The differences between the ZIMEN measurements and those conducted by us were again small. Table 1. Meteorological conditions during the study measurement periods in September and March including mean air temperature (TA) (◦C), mean relative humidity (RH) (%), precipitation sum (mm), atmospheric pressure (hPA), wind speed (m/s), wind direction (◦), mean convective inhibition (CIN) (J/kg), and mean mixing layer height (MLH) (m). Adapted with permission from Refs. [18,44] 2021, Landesamt für Umwelt Rheinland-Pfalz. Date 14.09. 15.09. 16.09. 17.09. 18.09. 01.03. 02.03. 03.03. 04.03. 05.03. TA (◦C) 23.0 23.3 23.8 19.5 18.8 7.7 9.2 8.2 9.6 5.8 RH (%) 53.6 56.6 57.1 51.2 46 66.8 62.7 71.5 71.7 59.4 Precipitation (mm) 0.3 0 0.1 1.1 1.9 0 0 0 9.5 0 Atmospheric pressure (hPA) 1013 1009 1007 1013 1012 1021 1020 1017 1008 1014 Wind direction (◦) 56 63 147 105 100 143 122 148 225 166 Wind speed (m/s) 0.2 0.1 0.4 0.6 0.6 0.5 0.3 0.2 0.6 1.1 CIN (J/kg) 177 171 171 60 85 163 148 219 110 27 MLH (m) 170 165 115 312 249 135 163 102 226 455 In both study periods, the differences in absolute PM2.5 between our runs and the stationary data were large and could not be explained by spatial variability. Our find- ings confirm the results of, e.g., Li et al. [45] and Sousan et al. [37], who identified this underestimation of OPC-N3 sensors compared to reference instruments. A stronger un- derestimation of the humidity-corrected PM2.5 concentration could be explained by the fact that any correction lowers the values [39]. Furthermore, the missing diurnal pattern during the first 3–4 days in September could be related to higher RH in the afternoon causing larger corrections. In March, however, the uncorrected measurements still showed diurnal patterns, whereas both the humidity-corrected and ZIMEN data did not. The humidity correction looked more suitable in the spring campaign, possibly because the prevailing higher RH resulted in corrections and hence stronger underestimations. The overall substantial underestimations and incomprehensible differences between mobile and stationary data led to the conclusion that our PM sensors cannot be used to assess absolute PM2.5 concentrations. For this reason, we used relative instead of absolute PM2.5 data to evaluate highly polluted areas. Atmosphere 2022, 13, 694 8 of 14 3.2. Highly Polluted Places and Sources The identification of highly polluted sites, i.e., sites with the highest 10% of PM2.5 concentrations throughout the entire study area, was conducted using 35 of 38 runs. Two runs were excluded due to sensor malfunctions (run two on 1 March 2021 and run two on 3 March 2021) and one run was omitted because of onsetting rain causing strongly Atmosphere 2022l,o 1w3, xe rFeOdR PPEMER, RwEhVIiEcWh did not allow reasonable spatial comparisons of that run (run three on 8 of 13 4 March 2021). Our result showed that no location was consistently identified as a highly polluted area throughout theOeunr trierseumlt eshaosuwreedm theantt npoe lroiocadtio(Fni gwuarse co5n).siTsthenistlrye isduelnttwifiaeds uasn ae xhpigehcltye dpolluted as our measureamreean tthsrotouoghkopulta tchee eanlotinreg mreoaasdusreamnedntl apregrieodin (tFeirgsuercet 5io).n Tshwis irtehsuhlti gwhasv uenheixcpleected as traffic, reportedotourb me emasauinresmoeunrtcse tooofkfi pnleacpea arltoicnlge rcooandcse anntdra latirognes ininterusrebcatinonasr ewaisth[ 4h6ig].hO venhtihclee traffic, contrary, locatiornepsowrtietdh thoi bgeh mleavine lssooufrcPeM of2 f.5inceo pualrdticble icdonencetnifitreadtions ainll uthrbraene atreaacsk [s4.6W]. Ohinl ethe con- all PM2.5 hot sptortasryw, leorceartieocnosr wdeitdh ihnigth eleNveelsu sotfa PdMt d2.5i sctoruicldt dbeu ridinegnttifhiedS oenp taelml thbreerem troacrknsi.n Wg hile all runs, the afternoPoMn2.r5 uhnost asplsootsi nwcelured eredcohridgehdl yinp othlleu Nteedupstlaadcte sdiisntrtihcte dAulrtisntga dtht.eF Soerpttheme beear lmy orning morning runs inruMnas,r tchhe, ahfotetrsnpoootns rwunesr ealssool ienlycluddeetedc hteigdhilny pthoelluHteadr tpelnabceesr gin, athned Ainltsltaatdetr. Fruorn tsh, e early the highly pollumteodrnpinlagc reusnws einr eMreacrcohr,d heodt ospnoatsll wtrearcek ssoyleelyt fdoectuecsteedd iinn tthhee HHaarrtteennbbeerrgg, aanndd in later Neustadt distrircutsn.s,P tehoe phlieghwlye preoltlhutuesd epxlapcoess ewdetroe rheicgohrdpeda rotnic alell ctroanckcse nyetrt afoticounsesda itnv tahrey Hinagrtenberg places in differeanntdu Nrbeaunstsaedttt dinisgtsricdtes.p Peenodpilne gwoenret thheusd eaxyptiomseeda tno dhisgeha psoarnt.icle concentrations at vary- ing places in different urban settings depending on the daytime and season. Figure 5. LocationFisgsuhreo w5. iLnogcathtieon1s0 s%hohwiginhge sthteP 1M0% hcigohnecsetn PtMra2t.5i ocnonsc(ernetdradtiootnss) (trherdo duogths)o tuhtrothugehrouunts the runs during the September and March ca2m.5paigns. during the September and March campaigns. However, thereHwowereevelor,c tahteioren ws wereh elorceaptioendse swtrhiearne spedestrians were exposed more frequently. In > 50% of the September runs, we identified 21w deifrfeereexnpt sopsoetds smhoowreinfgre rqecuuernritnlyg. hIingh PM2.5 >50% of the Secpotnecmenbterratirounnss t,hwroeugihdoeuntt aifille tdrac2k1s d(Fiifgfeurree n6t, lsepfto ptasnsehl)o. wThien glarrgeecru nrurimnbgerh oigf hhotspots in September could be assigned to the low absolute PM2.5 concentrations and minor dif- ferences among districts (Figure 4). At low particle concentrations, local emissions have a large influence on absolute PM2.5 peaks, and the mitigated track differences further the Atmosphere 2022, 13, 694 9 of 14 PM2.5 concentrations throughout all tracks (Figure 6, left panel). The larger number of Atmosphere 2022, 13, x FOR PEER REVIEW hotspots in September could be assigned to the low absolute PM concentration 9s oaf n1d3 Atmosphere 2022, 13, x FOR PEER REVIEW 2.5 9 of 13 minor differences among districts (Figure 4). At low particle concentrations, local emissions have a large influence on absolute PM2.5 peaks, and the mitigated track differences further sthperesapdre oafd hoofthspootstps.o Itns. MInaMrcahr,c thh,etrhee rweewree raet altelaesats stesveevne nhhigighhlyly ppoolllulutetedd llooccaattiioonnss,, mmoossttllyy rreesccpoorrreddaeeddd o oof nnh otthhtseep ooovvtsee.rr aaInllll Mmmaoorrrceeh pp, oothlllleuurtteee ddw eHHraea rratteet nnlebbaeesrrtgg s tterrvaaeccnkk ..h HHigoohwwlyee vpveeorrl,,l uwwtehhdee nnlo iicnnacctrrieeoaansssii,nn mgg otthhsteel y tthhrrreeecssohhroodlleddd ttoon dd teehffieinn oeev hheiirggahhllll yym ppooorellll uupttoeelddlu aaterrdeeaa Hss attoort >e>n660b0%e%r goo ftf rttahhceek .rr uuHnnossw,, ttehhvee rnn, uuwmmhbbeeenrr ionoffc rhheooatts issnppgoo ttthss e ddteehccrlliiennseehddo lmmd aatssoss iidvveeflliyyn tteoo h ooinngllhyyl yffii vvpeeo lllooucctaaettdiioo annrsse ((aFFsii ggtouu rr>ee6 660,,% rrii ggohhf ttt pphaaenn reeull;n; FFs,ii ggtuuhrreee n SSu33m)).. bTTehhree orrfee mmhoaatii nnsiipnnoggt s SSeedppettceelmimnbebdeer rm hhaoosttssippvooettlssy wwtoee orreen liiynn faaiv NNe eelouucssattaatiddottn pps aa(Frrkkig aaunnredd 6tt,hh reei gAAhllttt sspttaanddett lpp; Feedidgeeussrtterrii aaSnn3) z.z Toonnhee,, rwwemhheearrieneaainss g tthhSeee pttwwteoom rrbeememr aahiinnoiitnnspggo MMtsaa wrrccehhr ehh oointtss ppa ooNttsse wwuseetrraeed lltoo pccaatrteekdd a nneedaa rrth aae mm Aaaljjtoosrrta hhdootu upsseiinndgge scctoorinnassntt rrzuuoccnttiieoo,nn w sshiitteer iienna s HHtaahrrett eetnnwbboee rrggem aannadidn aain ttgrra aMfffifaiccr--cllohoaa hddoeetddsp bborrtiidsd gwgeee nrnee aalorrc ttahhteee dttrr anaiienna srst taa ttmiiooannj.o. r housing construction site in Hartenberg and a traffic-loaded bridge near the train station. Figure 6. Locations with 10% highest PM2.5 concentrations in >60% of all runs in September (left pFiaFgniugerul)er ae6n .6dL. MoLcoaacrtaciothino (snriswg whitthi tph1a 0n1%0el%)h. iChgiohglehosertssP tr MePfMe2r.52 t.5co oc ponancrecknesnt r(tagratriteoieonnns)s,i npine> d>6e60s0%t%rioa onf f a zlaollnlr uer un(bnslsiu nien Sa eSnpedtp ectmyemabneb)re, rc( lo(enlfe-tft pstaprnuaecntle)iol)an na sndidteM M (amarrcachghe( n(rritigagh)h, tat pnadn ae lbl)r.. iCdCgooello o(rvrssi orerleefeft)er. r tot opaprakrsk (sg(rgereene)n, p),epdedstersiatrnia znonzeo n(belu(bel uaneda cnydancy),a cno)n, - cosntrsutrcuticotnio snitsei t(em(amgaegnetna)t,a a),nadn ad barbidrigdeg (ev(iovlieotl)e.t ). The March hotspots could clearly be attributed to anthropogenic sources: the origin of theT TpheahreMt iMcalreacsrh caht othhtosept sboprtoisdtcsgo ecu oclduolucdll deca lbreleay ralbsyes ibagetnt areitdbtr utiotbe uvdteethodic atlone tsah, nraotshp trohogeperoeng iwcenasosic ua sr hcoeiugsr:hct ehinse:t etohnreisg iotiynri oogffi n ttrhoaeff fptihcae ro tpnica tlrhetseic amletsut hlateti -tlbhareni deb grrieodacgdoe u ccrlodousblsdein bagse ts ahigsesn biegrdnidetgdoe tv;o te hhveei chelmeicsli,esass,si oatnsh tehrerwtraffic on the multi-lane road crossing the bridge; the emission ssoouur ec eawssa oasf h at hihgeihg chion nitnsettnreusnicsttyiitoyon fo f sittrea wffiacs o sne etmhei nmgulylt ri-ellaanteed r otoa db ucrilodsisningg p trhoec ebsrsiedsg aen; dth fr eemission sou rcrecseso fotfh theec ocnonstsrtuructcitoinon sistietew wasass eseememiningglylyr erlealtaetdedt otob buuilidldininggp prorocecsessessesa annddf rfe qquueenntt ccoonnssttrruuccttiioonn vveehhicles [7,47requent construction vehicicleless[ 7[7,4,47 ]].. TThhese conclusions were supported 7]. Tehseeseco cnoclusions were supported b by low apportionment of PM2.5 in PM10 at these loca- tions (Figunrcel u7s).i oWnsh iwlee trhee s huipgpho PrtMed y lboyw loapwp oaprtpioonrmtioenmt oefnPtM of2 .P5 Min2P.5M in1 at these locations (Figure 7). While the high PM /P2.5M/PM1r0a rtaiotioa tatt htheeb brirdidggeen neeaarrt thhee mm 0PaMin1 0s taatt tiohnes (e0 .l7o3c)a - intdioincast (eFdi gtuhree p 7a)r. tWiclhei lseo tuhrec eh itgoh2 .o 5PrMig2.5/1P0M10 ratio at the bridge near the m aianins tsattaiotinon( 0(.07.37) indicated the particle source to originainteatper epdroedmoimnainntalnytflryo mfroamnt harnotphorogpenoigceenmici sesmioinsssidounes 3 ) diuned tioc actoemd bthues tpioanr ticle source to originate predominantly from anthropogenic emissions todcuoem tob ucsotmiobnupsrtoiocne psrsoecseosfsvese hoifc vleesh, itchleeslo, twheer lroawtieor( r0a.6ti3o) a(0n.d63h) iagnhdv hariigahb ivliatryia(ibnitleitryq u(ianrt processes t eilre- qraunagretil(eIQ raRn)g=e 0(.I1Q3R) a) t=t 0h.e13c)o nats tohfe v ceohnicslterus,c tthioen l oswiteesr praotiinot e(0d. 6t3o) aa nmdi xhtiugrhe voaf rrieasbuilsiptye n(idnetder - dquusta artnile range (IQR) = 0.13) a ttruction sites pd particles from combus ttihoen c pornoscterussceti oointeds. n site tso paominitxetdu rteo oaf mreisxutuspree nodf erdesduuspstenandded padrutisctl aesndfr opmartcioclmesb fursotmio ncopmrobcuesstsieosn. processes. Figure 7. PM2.5/PM10—ratio boxplots of the PM2.5 hot spots in the park (green) and pedestrian zones (FciFygiaugnru erae7n .d7P. bMPlMu2.e52.)/5 /PfPoMr 1S100e—ptrreaamttiioob ebbroo xx(gpprlleooettsns o,o fcf ytthhaene P,P aMn2d2..55 bhhlooutte ss ppcooolttossr iisnn) ttahhneed pp atahrrkek ( c(ggorrneeesentnr)u)a cantnidodnp p esdeidteese ts(rtmiraianagnze oznnoteans)e s (acn(ycday nmanao ntadonrdbizl uebdelu) refo) arfdoS re( vpSiteoeplmetteb)m efrobr(eg rMr e(agernrce,hec.ny Y,a encly,loaanwn,d dabnoldtus e bincluodleioc rcasot)eloa mrnsde)a tanhn evdac otluhnees stcr. ounctsitornucstiitoen( msiateg e(nmtaa)gaenndta) maontdor mizoedtorroizaedd( rvoioalde t()vfioorleMt) afrocrh M. Yaerlclho.w Ydelolotswin ddoitcsa tiendmiceaatne vmaeluanes v. alues. The three September hotspots were particularly surprising, as there was no motor- ized trTafhfeic t phrreesee Snet patte tmheb etirm heo. tTsphoist sis w ine rceo nptarratsict utola orluyr shuyrpportihsiensgis, tahsa tth tehree hwoatssp noots m woetroer - reilzaetded tr taoff ihce parveys etnrat fafitc t hster etiemtse .a Ts hthise i sm inai cno dnrtrivasetr toof o huirg hhy PpMothe csoisn tcheantt rthatei hot2.5 ons sipno utsr bwaenr e arreealast [e4d6 ]t.o T hheea hviyg ht rvaaflfuice ss tirne ethtse apse dthees tmriaanin z odnrievse cro oufl dh nigehv ePrMth2e.5l ecsosn bcee noftr aantitohnros pino guernbiac n orairgeians, [e4m6]i.t tTehde bhyig thh ev aelxuheas uinst t hsyes pteemdess torfi atnh ez ornesetsa cuoruanldt nkeitvcehretnhse lbeslosw bein ogf afinnteh rpoaprotigcelensi c origin, emitted by the exhaust systems of the restaurant kitchens blowing fine particles Atmosphere 2022, 13, 694 10 of 14 The three September hotspots were particularly surprising, as there was no motorized traffic present at the time. This is in contrast to our hypothesis that the hotspots were Atmosphere 2022, 13, x FOR PEER REVreIEW late d to heavy traffic streets as the main driver of high PM2.5 concentrations in1u0 robfa 1n3 areas [46]. The high values in the pedestrian zones could nevertheless be of anthropogenic origin, emitted by the exhaust systems of the restaurant kitchens blowing fine particles dduurriinngg ddeeeepp--ffrryyiinngg aanndd rrooaassttiinngg oonnttoo tthhee ssttrreeeettss [[4488]].. PPaarrttiicclleess wweerree lliikkeellyy aaddddiittiioonnaallllyy eemmiitttteedd iinnt thheeo ouutdtdoooorra raeraesaso fotfh tehree rsetastuaruarnatns t(sF i(gFuigreurSe3 ,Sp3a, npealn3e)l d3u) edtuoes tmo oskminogkiancgti vacittiievs- aitsieres paos rrteepdobryteBdi rbmy iBliiremt aill.i [e4t9 a]l.. H[4i9g]h. HPMigh2. 5P/MPM/1P0Mrati orast>io0s.6 >a0t.6t haet steheloc2.5 10 sea ltoiocantsiosunsp psuoprtptohret ctohne cclounsicolunstihoant tahnatth aronpthorgoepnoicgesnouicr cseosuwrceerse wtheerem tahien memaiintt eerms,itatserdso, eass tdhoeefsa ctht eth faatctth tehsaet stihteessew seitreesi dweenrteifi idedenatsifPieMd2 a.5s hPoMtspo htsotinsptohe2.5 ts ainft etrhneo aofnterrunnoso,ni. er.u,nast,t iim.e.e, sawt thimenesr ewsthaeunra rnetss- wtaeurreahnitgs hwlyerfere hqiugehnlyte fdr.equented. TThhee ppaarrttiiccllee ssoouurrcceessi nint htheep parakrkc ocuoludldn ontobt ebaet tartitbruibtuedtetdo tcoo mcobmubstuiostniopnr opcreoscseesssaess ians tihne thoeth oetrhheor thsoptostpso. tTsh. Tehme umchuclhow loewr emr emaneaPnM P2M.5/P2.5/PMM10 r10 raatitoio= = 00..2244 aanndd ssmmaallll IIQQRR == 00..0077 ppooiinntteedd ttooa ah hoommooggeennoouussp paratritcilcelec ocmompopsoistiiotinond udruinrignSge Spetpemtebmebreart atht itshlios claotcioatnio(Fni g(Fuirgeu7r)e. T7h).e Tsehevsael uveasluweesr we eeirteh eeritrheelra treedlattoedfu tgoi tfiuvgeitdivues td[u11s]t, [i1.e1.],, iim.ep., eirmvipoeursviaorueas saarenads faonodtp faotohts- cpoantthasi ncoinngtalionoisnegt looposmaterial, or to re-suspended road dust from the multi-lane road rightnext to the park. The sep taotpia ml daitsetrriiablu, toiro ntoo rfet-hsuesPpMende/dP 2.5 M road 10 r datuiosts ffroormS etphtee mmubletri-ilnadniec raoteadd trhigathht onreixzto tnot athl ter apnasrpko. Trthoef sfipnaetipaal rdtiicslterisbfurotimont hoef tchloes PeMro2a.5d/PwMa1s0 urantliiokse lfyor( FSiegputreem8b).eRr aintidois- ra0t.i5o douf r>i0n.g5 dcoulrdinerg sceoalsdoenr sse(aausotunms (na–uwtuimntne–r)w. iInn- oteurr).c aInse ,otuhre cinascere, atshee ininPcrMeaser ainti oPwMa2.s5 lriakteiloy wadads itliioknelayll yadadffietciotendalblyy aafpferoctneodu nbyce da cporool- nou 2.5thermncaeldin cvoeorls tihoenrm(Taabl ilnev1e; rFsiigounr (eTSa2b)l.eT 1h; eFsiegucoren Sd2it)i.o Tnhsecsoen csotnradiinteiodntsh ceovnesrttriacianlemd itxhien gveorf- atiicr,awl mhiicxhinlged oft oaiarn, winhcirceha sleedi ntofi anne ipnacrrteicalsee cionn fcinene tpraatritoicnlse actognrcoeunntrdatlieovnesl .aTt hgeropurnedva lielvinegl. lTohwe wpirnedvasiplienegd lsoswu bwseinqdu esnptelyedasm spulbifiseedqudernytldye pamospitliiofinedo fdcroya rdseeppoasrittiicolne so, fw choiacrhsein ptaurrtni- icnlcerse, awsehdicthh ienP tMurn inapcrpeoarsteiodn tmhee nPtMin2.5P aMppor[5ti1o–n5m3]e.nt in PM10 [51–53]. 2.5 10 4. Conclusions Using mobile low-cost devices containing Alphasense OPC-N3 sensors, small-scale PM2.5 hotspots along a 15 km transect in an urban area were identified. Three sensors showed a high agreement among each other but severely underestimated the measured PM2.5 concentrations of the ZIMEN network, particularly after applying a widely used humidity correction [39]. Absolute PM2.5 values were not considered, but additional Atmosphere 2022, 13, 694 11 of 14 4. Conclusions Using mobile low-cost devices containing Alphasense OPC-N3 sensors, small-scale PM2.5 hotspots along a 15 km transect in an urban area were identified. Three sensors showed a high agreement among each other but severely underestimated the measured PM2.5 concentrations of the ZIMEN network, particularly after applying a widely used humidity correction [39]. Absolute PM2.5 values were not considered, but additional calibration against high-resolution reference instruments could possibly improve the data accuracy of OPC-N3 sensors. The identification of (relatively) heavily polluted locations revealed persisting PM2.5 hotspots in >60% of all runs, though the locations varied between the September and March study periods. The March hot spots were most likely triggered by local anthropogenic emissions including traffic emissions and construction work. This conclusion was sup- ported by PM2.5/PM10 ratios >0.6 indicating combustion processes as the main particle source. The September hotspots, however, were located in areas dominated by pedestrians, and the PM sources were attributed to restaurant cooking exhaust air and outdoor seating activities. Exceptionally low PM2.5/PM10 ratios of 0.24 recorded in a park pointed to particles originating from locally emitted natural dust from unpaved footpaths, bare soils, and gravel surfaces. The PM2.5/PM10 ratios also increased from September to March as additional heating due to cooler temperatures and stable weather conditions prevailed during the spring campaign. The composition of sources can be further differentiated by analyzing the chemical composition of particles, which we recommend for further studies. The work detailed here revealed the capability of low-cost sensors to identify small-scale PM2.5 hotspots and sources. While the accuracy of absolute PM2.5 concentrations was insufficient, highly resolved spatiotemporal measurements may complement the stationary data and support the identification of highly polluted areas in the urban environment. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/atmos13050694/s1, Figure S1: weather conditions during Septem- ber measurement period; Figure S2: weather conditions during March measurement period; Figure S3: Pictures of the high polluted areas. Figure S4: Mean PM10 concentrations in the Altstadt, Hartenberg, Neustadt and ZIMEN data. Author Contributions: Conceptualization, L.H. and J.E.; methodology, L.H., T.S. and J.E.; formal analysis, L.H. and T.S.; investigation, L.H.; data curation, L.H.; writing—original draft preparation, L.H.; writing—review and editing, T.S. and H.S.; visualization, L.H.; supervision, J.E. All authors have read and agreed to the published version of the manuscript. Funding: J.E. received support from the Gutenberg Research College, SustES (CZ.02.1.01/0.0/0.0/16_ 019/0000797), and ERC (AdG 882727). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Acknowledgments: We thank several students of the Johannes Gutenberg University in Mainz including Joelle Juretzek, Leonard Köster, and Jan-Erik Schmitz for supporting the mobile measure- ments. We are grateful to the environmental state office of Rhineland-Palatinate, particularly Michael Weißenmayer and Margit von Döhren from Referat 60 and Matthias Zimmer and Matthias Voigt from Referat 63, for providing data of the official meteorological measurement stations and radiometer as shown in Table 1, Figure 4, Figure 5, Figures S1 and S2. Conflicts of Interest: The authors declare no conflict of interest. Atmosphere 2022, 13, 694 12 of 14 References 1. Mirabelli, M.C.; Ebelt, S.; Damon, S.A. Air Quality Index and air quality awareness among adults in the United States. Environ. Res. 2020, 183, 109185. [CrossRef] 2. Maione, M.; Mocca, E.; Eisfeld, K.; Kazepov, Y.; Fuzzi, S. Public perception of air pollution sources across Europe. Ambio 2021, 50, 1150–1158. [CrossRef] 3. Torres-Ramos, Y.D.; Montoya-Estrada, A.; Guzman-Grenfell, A.M.; Mancilla-Ramirez, J.; Cardenas-Gonzalez, B.; Blanco-Jimenez, S.; Sepulveda-Sanchez, J.D.; Ramirez-Venegas, A.; Hicks, J.J. Urban PM2.5 induces ROS generation and RBC damage in COPD patients. Front. Biosci. 2011, 3, 808–817. [CrossRef] 4. Gualtieri, M.; Ovrevik, J.; Mollerup, S.; Asare, N.; Longhin, E.; Dahlman, H.-J.; Camatini, M.; Holme, J.A. Airborne urban particles (Milan winter-PM2.5) cause mitotic arrest and cell death: Effects on DNA, mitochondria, AhR binding and spindle organization. Mutat. Res. 2011, 713, 18–31. [CrossRef] [PubMed] 5. Lelieveld, J.; Klingmüller, K.; Pozzer, A.; Pöschl, U.; Fnais, M.; Daiber, A.; Münzel, T. Cardiovascular disease burden from ambient air pollution in Europe reassessed using novel hazard ratio functions. Eur. Heart J. 2019, 40, 1590–1596. [CrossRef] 6. Chen, L.C.; Lippmann, M. Effects of metals within ambient air particulate matter (PM) on human health. Inhal. Toxicol. 2009, 21, 1–31. [CrossRef] [PubMed] 7. Karagulian, F.; Barbiere, M.; Kotsev, A.; Spinelle, L.; Gerboles, M.; Lagler, F.; Redon, N.; Crunaire, S.; Borowiak, A. Review of the Performance of Low-Cost Sensors for Air Quality Monitoring. Atmosphere 2019, 10, 506. [CrossRef] 8. Schlesinger, R.B. The health impact of common inorganic components of fine particulate matter (PM2.5) in ambient air: A critical review. Inhal. Toxicol. 2007, 19, 811–832. [CrossRef] [PubMed] 9. Xu, G.; Jiao, L.; Zhang, B.; Zhao, S.; Yuan, M.; Gu, Y.; Liu, J.; Tang, X. Spatial and Temporal Variability of the PM2.5/PM10 Ratio in Wuhan, Central China. Aerosol Air Qual. Res. 2017, 17, 741–751. [CrossRef] 10. Parkhurst, W.J.; Tanner, R.L.; Weatherford, F.P.; Valente, R.J.; Meagher, J.F. Historic PM2.5/PM10 Concentrations in the Southeastern United States-Potential Implications of the Revised Particulate Matter Standard. J. Air Waste Manag. Assoc. 1999, 49, 1060–1067. [CrossRef] 11. Evagelopoulos, V.; Zoras, S.; Triantafyllou, A.G.; Albanis, T.A. PM10–PM2.5 Time Series and Fractal Analysis. Glob. NEST J. 2019, 8, 234–240. [CrossRef] 12. Querol, X.; Alastuey, A.; Rodriguez, S.; Plana, F.; Ruiz, C.R.; Cots, N.; Massague, G.; Puig, O. PM10 and PM2.5 source apportion- ment in the Barcelona Metropolitan area, Catalonia, Spain. Atmos. Environ. 2001, 35, 6407–6419. [CrossRef] 13. Czernecki, B.; Półrolniczak, M.; Kolendowicz, L.; Marosz, M.; Kendzierski, S.; Pilguj, N. Influence of the atmospheric conditions on PM10 concentrations in Poznań, Poland. J. Atmos. Chem. 2017, 74, 115–139. [CrossRef] 14. Graham, A.M.; Pringle, K.J.; Arnold, S.R.; Pope, R.J.; Vieno, M.; Butt, E.W.; Conibear, L.; Stirling, E.L.; McQuaid, J.B. Impact of weather types on UK ambient particulate matter concentrations. Atmos. Environ. X 2020, 5, 100061. [CrossRef] 15. Wagner, P.; Schäfer, K. Influence of mixing layer height on air pollutant concentrations in an urban street canyon. Urban Clim. 2017, 22, 64–79. [CrossRef] 16. Tang, G.; Zhang, J.; Zhu, X.; Song, T.; Münkel, C.; Hu, B.; Schäfer, K.; Liu, Z.; Zhang, J.; Wang, L.; et al. Mixing layer height and its implications for air pollution over Beijing, China. Atmos. Chem. Phys. 2016, 16, 2459–2475. [CrossRef] 17. Speranza, A.; Caggiano, R.; Margiotta, S.; Summa, V.; Trippetta, S. A clustering approach based on triangular diagram to study the seasonal variability of simultaneous measurements of PM10, PM2.5 and PM1 mass concentration ratios. Arab. J. Geosci. 2016, 9, 1–8. [CrossRef] 18. ZIMEN.-Luft-Überwachung in Rheinland-Pfalz-: Zentrales Immissionsmessnetz-ZIMEN-. Available online: https://luft.rlp.de/ de/startseite/ (accessed on 2 March 2022). 19. WMO. Low-Cost Sensors for the Measurement of Atmospheric Composition: Overview of Topic and Future Applications. 2018. Available online: https://www.ccacoalition.org/en/resources/low-cost-sensors-measurement-atmospheric-composition- overview-topic-and-future (accessed on 21 January 2022). 20. Alfano, B.; Barretta, L.; Del Giudice, A.; de Vito, S.; Di Francia, G.; Esposito, E.; Formisano, F.; Massera, E.; Miglietta, M.L.; Polichetti, T. A Review of Low-Cost Particulate Matter Sensors from the Developers’ Perspectives. Sensors 2020, 20, 6819. [CrossRef] 21. Al-Ali, A.R.; Zualkernan, I.; Aloul, F. A Mobile GPRS-Sensors Array for Air Pollution Monitoring. IEEE Sens. J. 2010, 10, 1666–1671. [CrossRef] 22. Alphasense. Technical Information Release December 2018: Alphasense Particle Counter OPC-N Range Product Update. Available online: http://www.alphasense.com/WEB1213/wp-content/uploads/2019/02/OPC-N3-information-update-Dec- 18.pdf (accessed on 1 April 2021). 23. Morawska, L.; Thai, P.K.; Liu, X.; Asumadu-Sakyi, A.; Ayoko, G.; Bartonova, A.; Bedini, A.; Chai, F.; Christensen, B.; Dunbabin, M.; et al. Applications of low-cost sensing technologies for air quality monitoring and exposure assessment: How far have they gone? Environ. Int. 2018, 116, 286–299. [CrossRef] 24. Sousan, S.; Koehler, K.; Hallett, L.; Peters, T.M. Evaluation of the Alphasense Optical Particle Counter (OPC-N2) and the Grimm Portable Aerosol Spectrometer (PAS-1.108). Aerosol Sci. Technol. 2016, 50, 1352–1365. [CrossRef] [PubMed] Atmosphere 2022, 13, 694 13 of 14 25. Brattich, E.; Bracci, A.; Zappi, A.; Morozzi, P.; Di Sabatino, S.; Porcù, F.; Di Nicola, F.; Tositti, L. How to Get the Best from Low-Cost Particulate Matter Sensors: Guidelines and Practical Recommendations. Sensors 2020, 20, 3073. [CrossRef] [PubMed] 26. Crilley, L.R.; Shaw, M.; Pound, R.; Kramer, L.J.; Price, R.; Young, S.; Lewis, A.C.; Pope, F.D. Evaluation of a low-cost optical particle counter (Alphasense OPC-N2) for ambient air monitoring. Atmos. Meas. Tech. 2018, 11, 709–720. [CrossRef] 27. Di Antonio, A.; Popoola, O.A.M.; Ouyang, B.; Saffell, J.; Jones, R.L. Developing a Relative Humidity Correction for Low-Cost Sensors Measuring Ambient Particulate Matter. Sensors 2018, 18, 2790. [CrossRef] 28. ZIMEN. Jahresbericht 2020: Zentrales Immissionsmessnetz-ZIMEN-. Available online: https://luft.rlp.de/fileadmin/luft/ ZIMEN/Jahresberichte/ZIMEN-Jahresbericht_2020.pdf (accessed on 20 January 2022). 29. Deutscher Wetterdienst. Niederschlag: Vieljährige Mittelwerte 1981–2010. Available online: https://www.dwd.de/DE/ leistungen/klimadatendeutschland/mittelwerte/nieder_8110_fest_html.html?view=nasPublication&nn=16102 (accessed on 10 May 2021). 30. Deutscher Wetterdienst. Temperatur: Vieljährige Mittelwerte 1981–2010. Available online: https://www.dwd.de/DE/leistungen/ klimadatendeutschland/mittelwerte/temp_8110_fest_html.html%3Fview%3DnasPublication (accessed on 10 May 2021). 31. Stewart, I.D.; Oke, T.R. Local Climate Zones for Urban Temperature Studies. Bull. Am. Meteorol. Soc. 2012, 93, 1879–1900. [CrossRef] 32. Espressif. ESP32. Available online: https://www.espressif.com/en/products/socs/esp32 (accessed on 5 February 2021). 33. Simcom. Sim28. Available online: https://www.simcom.com/product/SIM28.html (accessed on 5 February 2021). 34. Mie, G. Beiträge zur Optik trüber Medien, speziell kolloidaler Metallösungen. Ann. Phys. 1908, 330, 377–445. [CrossRef] 35. Bohren, C.F.; Huffman, D.R. Absorption and Scattering of Light by Small Particles; Wiley: Hoboken, NJ, USA, 1998; ISBN 9780471293408. 36. Walser, A.; Sauer, D.; Spanu, A.; Gasteiger, J.; Weinzierl, B. On the parametrization of optical particle counter response including instrument-induced broadening of size spectra and a self-consistent evaluation of calibration measurements. Atmos. Meas. Tech. 2017, 10, 4341–4361. [CrossRef] 37. Sousan, S.; Regmi, S.; Park, Y.M. Laboratory Evaluation of Low-Cost Optical Particle Counters for Environmental and Occupa- tional Exposures. Sensors 2021, 21, 4146. [CrossRef] 38. Gysel, M.; Crosier, J.; Topping, D.O.; Whitehead, J.D.; Bower, K.N.; Cubison, M.J.; Williams, P.I.; Flynn, M.J.; McFiggans, G.B.; Coe, H. Closure study between chemical composition and hygroscopic growth of aerosol particles during TORCH2. Atmos. Chem. Phys. 2007, 7, 6131–6144. [CrossRef] 39. Crilley, L.R.; Singh, A.; Kramer, L.J.; Shaw, M.D.; Alam, M.S.; Apte, J.S.; Bloss, W.J.; Hildebrandt Ruiz, L.; Fu, P.; Fu, W.; et al. Effect of aerosol composition on the performance of low-cost optical particle counter correction factors. Atmos. Meas. Tech. 2020, 13, 1181–1193. [CrossRef] 40. Hagler, G.S.W.; Williams, R.; Papapostolou, V.; Polidori, A. Air Quality Sensors and Data Adjustment Algorithms: When Is It No Longer a Measurement? Environ. Sci. Technol. 2018, 52, 5530–5531. [CrossRef] [PubMed] 41. Chatzidiakou, L.; Krause, A.; Popoola, O.A.M.; Di Antonio, A.; Kellaway, M.; Han, Y.; Squires, F.A.; Wang, T.; Zhang, H.; Wang, Q.; et al. Characterising low-cost sensors in highly portable platforms to quantify personal exposure in diverse environments. Atmos. Meas. Tech. 2019, 12, 4643–4657. [CrossRef] 42. Shepard, D. A two-dimensional interpolation function for irregularly-spaced data. In Proceedings of the 1968 23rd ACM national conference, New York, NY, USA, 27–29 August 1968; Blue, R.B., Rosenberg, A.M., Eds.; ACM Press: New York, NY, USA, 1968; pp. 517–524. 43. Thermo Fisher Science. Model 5030i SHARP: Instruction Manual. Available online: https://tools.thermofisher.com/content/sfs/ manuals/EPM-manual-Model%205030i%20SHARP.pdf (accessed on 31 January 2022). 44. Umweltmeteorologie RLP. Available online: https://luft.rlp.de/de/umweltmeteorologie/ (accessed on 1 March 2022). 45. Li, J.; Mattewal, S.K.; Patel, S.; Biswas, P. Evaluation of Nine Low-cost-sensor-based Particulate Matter Monitors. Aerosol Air Qual. Res. 2020, 20, 254–270. [CrossRef] 46. Kumar, P.; Rivas, I.; Sachdeva, L. Exposure of in-pram babies to airborne particles during morning drop-in and afternoon pick-up of school children. Environ. Pollut. 2017, 224, 407–420. [CrossRef] [PubMed] 47. Yue, D.L.; Hu, M.; Wu, Z.J.; Guo, S.; Wen, M.T.; Nowak, A.; Wehner, B.; Wiedensohler, A.; Takegawa, N.; Kondo, Y.; et al. Variation of particle number size distributions and chemical compositions at the urban and downwind regional sites in the Pearl River Delta during summertime pollution episodes. Atmos. Chem. Phys. 2010, 10, 9431–9439. [CrossRef] 48. Kang, K.; Kim, H.; Kim, D.D.; Lee, Y.G.; Kim, T. Characteristics of cooking-generated PM10 and PM2.5 in residential buildings with different cooking and ventilation types. Sci. Total Environ. 2019, 668, 56–66. [CrossRef] 49. Birmili, W.; Rehn, J.; Vogel, A.; Boehlke, C.; Weber, K.; Rasch, F. Micro-scale variability of urban particle number and mass concentrations in Leipzig, Germany. METZ 2013, 22, 155–165. [CrossRef] 50. Paas, B.; Schneider, C. A comparison of model performance between ENVI-met and Austal2000 for particulate matter. Atmos. Environ. 2016, 145, 392–404. [CrossRef] 51. Huang, R.-J.; Zhang, Y.; Bozzetti, C.; Ho, K.-F.; Cao, J.-J.; Han, Y.; Daellenbach, K.R.; Slowik, J.G.; Platt, S.M.; Canonaco, F.; et al. High secondary aerosol contribution to particulate pollution during haze events in China. Nature 2014, 514, 218–222. [CrossRef] Atmosphere 2022, 13, 694 14 of 14 52. Wallace, J.; Kanaroglou, P. The effect of temperature inversions on ground-level nitrogen dioxide (NO2) and fine particulate matter (PM2.5) using temperature profiles from the Atmospheric Infrared Sounder (AIRS). Sci. Total Environ. 2009, 407, 5085–5095. [CrossRef] 53. Blanco-Becerra, L.C.; Gáfaro-Rojas, A.I.; Rojas-Roa, N.Y. Influence of precipitation scavenging on the PM2.5PM10 ratio at the Kennedy locality of Bogotá, Colombia. Rev. Fac. Ing. Antioq. 2015, 76, 58–65. [CrossRef]