Atmospheric Environment 41 (2007) 6063–6073 Characterisation of bio-aerosols during dust storm period in N–NW India Sudesh Yadav a, , M.S. Chauhan b , Anupam Sharma b a School of Environmental Sciences, Jawaharlal Nehru University, New Delhi 110067, India b Birbal Sahni Institute of Palaeobotany, 53 University Road, Lucknow 226007, India Received 21 May 2006; received in revised form 15 February 2007; accepted 9 March 2007 Abstract Bio-investigations for pollen and spores were performed on dry free-fall dust and PM 10 aerosol samples, collected from three different locations separated by a distance of 600 km, situated in dust storm hit region of N–NW India. Presence of pollen of trees namely Prosopis (Prosopis juliflora and Prosopis cinearia), Acacia, Syzygium, Pinus, Cedrus, Holoptelea and shrubs namely Ziziphus, Ricinus, Ephedra and members of Fabaceae, Oleaceae families was recorded but with varying proportions in the samples of different locations. Poaceae, Chenopodiaceae/Amaranthaceae, Caryophyllaceae, Brassicaceae and Cyperaceae (sedges) were some of the herb pollen identified in the samples. Among the fungal spores Nigrospora was seen in almost all samples. Nigrospora is a well known allergen and causes health problems. The concentration of trees and shrubs increases in the windward direction just as the climate changes from hot arid to semiarid. The higher frequency of grasses (Poaceae) or herbs could either be a result of the presence of these herbs in the sampling area and hence the higher production of pollen/spores or due to the resuspension from the exposed surface by the high- intensity winds. But we cannot ascertain the exact process at this stage. The overall similarity in the pollen and spore assemblage in our dust samples indicates a common connection or source(s) to the dust in this region. Presence of the pollen of the species of Himalayan origin in our entire samples strongly point towards a Himalayan connection, could be direct or indirect, to the bioaerosols and hence dust in N–NW India. In order to understand the transport path and processes involved therein, present study needs further extension with more number of samples and with reference to meteorological parameters. r 2007 Elsevier Ltd. All rights reserved. Keywords: Palynology; Pollen; Spores; Dust/Aerosols; Thar Desert 1. Introduction Biosphere is a major source of primary aerospora and the cellular particles, which constitute a large fraction of the atmospheric aerosols (Jaenicke, 2005). Particles of such origin are often called as bio-aerosols among which pollen and spores are the important constituents. Recently, the study of bio- aerosols is getting impetus because of their adverse effects on human health (Taylor, 2002; Griffin et al., 2003; Wu et al., 2004) as well as their strong role in the climate change as they are hygroscopic in nature and can act as cloud condensation nuclei (Jaenicke, 2005). The effects of dust storm activities on the quantity and quality of aerosols (physico-chemical ARTICLE IN PRESS www.elsevier.com/locate/atmosenv 1352-2310/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2007.03.050 Corresponding author. Tel.: +91 9891625426. E-mail address: [email protected] (S. Yadav).
Transcript
ARTICLE IN PRESS
1352-2310/$ - se
doi:10.1016/j.at
�CorrespondE-mail addr
Atmospheric Environment 41 (2007) 6063–6073
www.elsevier.com/locate/atmosenv
Characterisation of bio-aerosols during dust stormperiod in N–NW India
Sudesh Yadava,�, M.S. Chauhanb, Anupam Sharmab
aSchool of Environmental Sciences, Jawaharlal Nehru University, New Delhi 110067, IndiabBirbal Sahni Institute of Palaeobotany, 53 University Road, Lucknow 226007, India
Received 21 May 2006; received in revised form 15 February 2007; accepted 9 March 2007
Abstract
Bio-investigations for pollen and spores were performed on dry free-fall dust and PM10 aerosol samples, collected from
three different locations separated by a distance of 600 km, situated in dust storm hit region of N–NW India. Presence of
pollen of trees namely Prosopis (Prosopis juliflora and Prosopis cinearia), Acacia, Syzygium, Pinus, Cedrus, Holoptelea and
shrubs namely Ziziphus, Ricinus, Ephedra and members of Fabaceae, Oleaceae families was recorded but with varying
proportions in the samples of different locations. Poaceae, Chenopodiaceae/Amaranthaceae, Caryophyllaceae,
Brassicaceae and Cyperaceae (sedges) were some of the herb pollen identified in the samples. Among the fungal spores
Nigrospora was seen in almost all samples. Nigrospora is a well known allergen and causes health problems. The
concentration of trees and shrubs increases in the windward direction just as the climate changes from hot arid to semiarid.
The higher frequency of grasses (Poaceae) or herbs could either be a result of the presence of these herbs in the sampling
area and hence the higher production of pollen/spores or due to the resuspension from the exposed surface by the high-
intensity winds. But we cannot ascertain the exact process at this stage. The overall similarity in the pollen and spore
assemblage in our dust samples indicates a common connection or source(s) to the dust in this region. Presence of the
pollen of the species of Himalayan origin in our entire samples strongly point towards a Himalayan connection, could be
direct or indirect, to the bioaerosols and hence dust in N–NW India. In order to understand the transport path and
processes involved therein, present study needs further extension with more number of samples and with reference to
meteorological parameters.
r 2007 Elsevier Ltd. All rights reserved.
Keywords: Palynology; Pollen; Spores; Dust/Aerosols; Thar Desert
1. Introduction
Biosphere is a major source of primary aerosporaand the cellular particles, which constitute a largefraction of the atmospheric aerosols (Jaenicke,2005). Particles of such origin are often called as
e front matter r 2007 Elsevier Ltd. All rights reserved
bio-aerosols among which pollen and spores are theimportant constituents. Recently, the study of bio-aerosols is getting impetus because of their adverseeffects on human health (Taylor, 2002; Griffin et al.,2003; Wu et al., 2004) as well as their strong role inthe climate change as they are hygroscopic in natureand can act as cloud condensation nuclei (Jaenicke,2005). The effects of dust storm activities on thequantity and quality of aerosols (physico-chemical
ARTICLE IN PRESSS. Yadav et al. / Atmospheric Environment 41 (2007) 6063–60736064
or geochemical) in atmosphere are well documentedor being studied all over the world with differentpurposes varying from climate change (Broecker,2000; Prospero et al., 2002) to environmentalpollution (Kim et al., 2004; Yadav and Rajamani,2006) to desertification processes (Reynolds et al.,2001; Yadav and Rajamani, 2004). But we lackinformation about the effects of dust storm onbioaerosols especially pollen and spores and theireffects on life system. A few studies have shown thatintercontinental sandstorm may transmit a fairlyhigh amount of viable microbes and thereby causinghealth problems (Taylor, 2002; Yeo and Kim, 2002;Wu et al., 2004).
In India, whole N–NW region experiences severedust storm activities through strong S–SW windslocally called as ‘‘aandhi’’ during summers.The aerobiological studies in India have beenexecuted from some of the cities such as Lucknow(Khandelwal, 1988), Bangalore (Agashe andAbraham, 1988), Vishakhapatanam (Subba Reddiand Ramakrishna, 1978), Aurangabad (Tilak andRamachandra Rao, 1987) and Bhopal (Tripathi andOomachan, 1981) to understand the various biolo-gical entities particularly pollen and spores sus-pended in the air during different seasons. Thesereports have shown that pollen/spores belonging todifferent plant groups are found in variablenumbers, depending upon the regional vegetationcomposition and whether plants are in full blossomor in partial flowering condition. But there is noinformation available on characterization of bio-aerosols in the dust storm hit region of N–NWIndia, which could otherwise have improved ourunderstanding about bioaerosols and their sources.The geochemical studies on aerosols in this regionhave suggested a common immediate source of dusti.e. the Great Indian Thar Desert and certainlithotectonic units of Himalayas as an ultimatesource (Yadav and Rajamani, 2004). These findingshave led one of the author here to study thequalitative and quantitative nature of biologicalentities in this dust storm hit region in order tounderstand first, the nature of bio-aerosols andsecondly the interconnection between bio-aerosolsand dust sources.
The pollen and spore analysis were undertaken onthe three air-catches or free-fall dust samplescollected during summer season from Bikaner,situated in the upwind direction in the Thar Desertand Delhi in down wind direction through Jhunj-hunu located in-between these two sites. All of our
sampling locations fall on same air back trajectory.The fourth air-catches of PM10 aerosols (sizeo10 mm; inhalable fraction) from Jhunjhunu wasalso studied. We have observed a similarity betweenpollen/spores identified in samples of differentlocations. All our samples have pollen of Himala-yan gymnosperms (Pinus and Cedrus) although insmall percentage pointing towards a Himalayanconnection to the dust, which is coherent withprevious geochemical understanding on dust in thisregion (Yadav and Rajamani, 2004). The presenceof allergens in dust samples particularly, PM10 maybe responsible for certain health problems in thisregion. The results are quite interesting andmotivating and require more detailed study in thisdirection with more number of samples in terms ofaerosol size and sampling stations coupled withclimatic and meteorological parameters.
2. Study area
The aerosol sampling was done at three differentlocations viz. Bikaner (28.011N and 73.221E locatedin the Thar Desert), Jhunjhunu (28.061N and75.251E) and Delhi, the national capital, (28.381Nand 77.121E) covering a total distance of about600 km long along the common S–SW windtrajectory (Fig. 1; air back trajectories can be seenin Yadav and Rajamani, 2006). Selection of thesampling points was based on the meteorologicalparameters (wind speed and direction) and was suchthat it covers maximum dust storm hit regions. Theclimate is hot arid in desert region and changes tosemiarid on moving eastward to Delhi. S–SW windsare the dominant wind pattern during summerperiod and transport tremendous amount of dust inwindward direction and deposit in the neighboringstates of the Thar (for more sampling area detailssee Yadav and Rajamani, 2004, 2006).
3. Sampling and analytical methods
The aerosol samples were collected simulta-neously at Bikaner, Jhunjhunu and Delhi stationsusing different techniques. Free-fall dust sampleswere collected for a few days between 15th and 20thMay 2005 during summer season while PM10
aerosols were collected for 2 days on 16th and17th May. The sampling duration was based on ourprevious experiences in the field, so that we couldget sufficient quantity of samples for bioaerosolanalysis. Dust samples falling freely due to gravity
ARTICLE IN PRESS
Fig. 1. Study area map showing sampling locations and the wind direction.
S. Yadav et al. / Atmospheric Environment 41 (2007) 6063–6073 6065
were collected at all the stations at a height of 20mabove the ground level by using marble dustcollectors (MDCs) international standard methodfor free-fall dust collection, using plastic trays of6 cm depth, filled with two layers of glass marbles(Reheis and Kihl, 1995). Free-fall samples werecollected on dry only basis during no rain period.The second type of aerosols, PM10 (o10 mm in size)was collected at the same height above the groundlevel using PM10 high volume air samplers, fittedwith EPM 2000 glass microfiber filter papers. Theinstrument was operated at an average flow rate of12 lmin�1 after Lin et al. (1999). PM10 aerosolscollected on glass microfiber filter papers were indry powder form and, therefore, were easily scrapedoff the filters (Yadav and Rajamani, 2004, 2006). Atotal number of three free-fall air-catches one eachfrom Bikaner, Delhi and Jhunjhunu and the onlyPM10 sample from Jhunjhunu was selected for thepresent bio-investigations. The selection of airsamples for bio-analysis was based on the quantityof the sample available.
Samples were treated with glacial acetic acid inorder to dehydrate them. Thereafter, the conven-
tional method of acetolysis (Erdtman, 1943) wasfollowed, using acetolysing mixture (9:1 ratio ofacetic anhydride and conc. sulphuric acid) toremove the cellulose content of pollen/spores.Finally, the samples were prepared in 50% glycerinsolution for microscopic examination. The identifi-cation of pollen/spores was carried out by compar-ing the pollen spectra of our samples with referencepollen slides available with the Herbarium of BirbalSahni Institue of Palaeobotany (BSIP), Lucknow,India (Plates 1 and 2). The recovered palynomorphshave been grouped as trees, shrubs, herbs and fungiand are arranged in the same sequence in the pollenspectra (Fig. 2).
4. Pollen composition of the air-catches
The preliminary bio-investigation of four differ-ent air-catches collected from three different loca-tions in the dust storm hit regions of India hasrevealed a diverse composition of pollen and spores.The four different categories, trees, shrubs, herbsand fungal spores were identified and studied foreach sample. Pollen of most of the trees and a few
ARTICLE IN PRESS
Plate 1. Pollen recovered in the air-catches (Magnification � 200) 1–3. P. cineraria; 4 and 5. P. juliflora; 6. Acacia Type I; 7. Acacia Type
II; 8. Syzygium; 9. Holoptelea; 10. Ephedra; 11 and 12. Cheno/Am Type I; 13. Cheno/Am Type II; 14.Artemisia; 15. Borrasus;
16.Unidentified Type I; 17. Cyperaceae Type I; 18. Cyperaceae Type II; 19. Cedrus; 20. Pinus; 21. Alnus; 22 and 23. Typha; 24. Capparis.
S. Yadav et al. / Atmospheric Environment 41 (2007) 6063–60736066
ARTICLE IN PRESS
Plate 2. Pollen/spores recovered in the air-catches (Magnification � 200) 25 and 26. Brassica spp.; 27 and 28. Tribulus; 29. Tubuliflorae
Type I; 30. Tubuliflorae Type II; 31. C. sativa; 32. Ziziphus; 33. Polyad (Mimosaceae); 34. Nigrospora; 35 and 36. Diplodia; 37 and 43.
S. Yadav et al. / Atmospheric Environment 41 (2007) 6063–6073 6067
ARTICLE IN PRESS
Fig. 2. Pollen spectra of air-catches of different sampling sites in N–NW India.
S. Yadav et al. / Atmospheric Environment 41 (2007) 6063–60736068
shrubs and herbs retrieved in the aerosol have beenidentified up to generic level only, as there is muchoverlapping in characters at specific level. Further,the pollen of plants belonging to Poaceae, Cyper-aceae, Caryophyllaceae, Brassicaceae, Oleaceae andFabaceae, are very identical and their genericsegregation is difficult in disperse conditions. Hence,such pollen are designated by the name of theirrespective families here in the text. Similarly, thepollen of Chenopodiaceae and Amaranthaceaecould not be differentiated upto family level dueto their close resemblance among themselves excepta few taxa and hence, put together under a commongroup, Cheno/Am. This is also to mention that, allthe plants belonging to Poaceae, Cyperaceae,Caryophyllaceae, and Brassicaceae are alwaysherbaceous and, therefore, kept under herbs cate-gory. Whereas, pollen identified as Oleaceae andFabaceae are included in shrubs since they showclose resemblance with the shrubby plants of thesefamilies occurring in the study area. The biologicalspecies present under each category are representedin pollen spectra (Fig. 2) for all our samples. Thephotographs of all the biological entities identifiedand studied in all our samples are given in Plates 1and 2 for ready reference.
Pollen grains exhibit a wide range of variabilityconcerning their shape, size and sculpturing of theexine layer. The pollen of anemophilous plantsbelonging to Poaceae (grasses), Chenopodiaceae/Amaranthaceae, Caryophyllaceae and Cyperaceaefamilies are smooth, rounded to elongated, in thesize range of 15–60 mm and are produced in hugequantity to ensure pollination. However, the pollen
of conifers (Pinus and Cedrus) are exceptionally verylarge (100–200 mm) and winged. In contrast, theentomophilous plants such as most of trees andshrubby members of Fabaceae, Oleaceae, Brassica-ceae, etc. often have sticky and usually large pollenwith a wide size-range varying from 10 to 80 mm andare produced in low amount. They are also providedwith complex sculpturing, which assists them inadhering to insects.
The normal distance for dispersal of pollen aftertheir release from anther is 50–100 km (Faegri andIversen, 1989). Hence, it is self evident that thegreatest quantities of pollen get deposited longbefore this limit. However, a small fraction of pollenstill remains in the air for more than 1 day, and aresubsequently transported to long distances i.e.41000 km by wind turbulence. Again the transportand dispersal pollen depend on the factors viz., size,shape and weight as well as meteorological para-meters such as wind speed and direction, tempera-ture, atmospheric turbulence and precipitation.
The relative proportions of pollen of differentplant species on number counting basis i.e. numberpercentage distribution in all our samples are shownin pie diagrams (Fig. 3). The mass concentration ofsamples i.e. dust load in atmosphere show asystematic decrease in downwind direction fromwest to east (Bikaner—3.9 gm�2 day�1; Jhunjhu-nu—2.1 gm�2 day�1; Delhi—0.48 gm�2 day�1). ThePM10 load was 800 mgm�3 of air at Jhunjhunu site.We observe the variations in the number percentagedistribution of pollen of different plant groups asfunction of sampling site and the particle size (Free-Fall—coarse; PM10—fine). Although the mass
ARTICLE IN PRESS
Fig. 3. Pie diagram showing percentage distribution of pollens of major plant groups recovered in the air-catches. Although grasses,
Chenopodiaceae/Amaranthaceae, Caryophyllaceae, Brassicaceae and Cyperaceae constitute herbs only but present in significant amount
therefore represented separately in the pie diagrams.
S. Yadav et al. / Atmospheric Environment 41 (2007) 6063–6073 6069
concentration could have provided a good insighton quantitative nature of bioaerosols we could notget the mass concentration distribution pattern dueto our experimental limitation in this preliminarystudy on pollen in air. The pollen assemblagesobtained in the air-catches are dealt separately foreach site and size of aerosols here below:
4.1. Delhi free-fall air-catch
The pollen composition of Delhi Free-fall air-catch has brought out the dominance of non-arboreals (herbs) as compared to arboreals (trees
and shrubs). Among the arboreals Prosopis 7.45%,Acacia Type-I and Type-II 3.47% and drifted pollenof Pinus 2.48% contribute moderately to theaerospora, whereas Capparis and Borrasus 0.99%each, Meliaceae and Cedrus 0.49% each are scantlyrepresented. Fabaceae 1.49% and Oleaceae 0.49%with low frequencies, represent the shrubby ele-ments in the aerospora. In the non-arboreals,Poaceae (grasses) 30.3% is the dominant constituentof air-catch. Besides, Chenopodiaceae/Amarantha-ceae 9.94%, Cerealia, Brassica spp. and Cyperaceae(sedges) 6.46% each, Cannabis sativa, Caryophylla-ceae and Typha (an aquatic element) 5.46% each
ARTICLE IN PRESSS. Yadav et al. / Atmospheric Environment 41 (2007) 6063–60736070
and Tubuliflorae 3.47% are also recovered in goodrequencies. Liguliflorae and Tribulus 0.99% eachare meagerly present. Among the fungal spores,Nigrospora 0.99% and other fungal remains 0.49%are extremely sporadic.
4.2. Bikaner free-fall air-catch
The Free-fall air-catch studied from Bikaner hasalso revealed the dominance of non-arboreals just asDelhi samples; however, the arboreals here exhibitlow diversity and frequencies in the total aerospora.The trees are few and among them Moringa 4.67%is better represented in contrast to Acacia 1.78%,Prosopis and Syzygium 0.88% each. The pollen ofHimalayan plants, Cedrus 1.16% and Pinus 1.58%are also encountered, though in low frequencies.Ziziphus 1.16% and Fabaceae 1.58% are therepresentative of shrubby vegetation in the ambi-ence of air catch.
Among the non-arboreals, Poaceae 27.4% fol-lowed by Caryophyllaceae 14.6%, Chenopodiaceae/Amaranthaceae and Cyperaceae 11.68% each arethe prominent ingredients of aerospora. Tubuliflor-ae (Asteraceae) 7%, Typha 2.92% and Artemisia
2.33% are recorded in moderate to low frequencies,whereas C. sativa, Xanthium and Cerealia 1.16%each are sporadic. Nigrospora 3.5% the only fungalspore, is present in moderate frequency.
4.3. Jhunjhunu air-catches
The pollen study from Jhunjhunu is based on twoair-catches, Free-fall and PM10 (o10 mm size,inhalable fraction). The former is marked by therelatively higher frequencies of arboreals, whereasthe latter has quantitatively better assemblage.
4.3.1. Free-fall air-catch
This air-catch exhibits a poor assemblage of treepollen. However, Prosopis 15.8% is characterizedby its maximum frequency. Moringa 2.72% isencountered moderately. Pinus and Acacia 1.17%each and Tsuga 0.77% are recorded in moderate tolow frequencies. Fabaceae 1.16% is the onlyrepresentative of shrubby vegetation. Poaceae14.39% among the non-arboreals has much reducedfrequency as compared to other sites. Chenopodia-ceae/Amaranthaceae 14.39%, Caryophyllaceae10%, Brassica spp. 5.9% and Cyperaceae 8.9%are important ingredients of non-arboreal complex.Cerealia, Artemisia, Solanum and Typha 2.3% each,
despite of low frequencies are better represented incontrast to Tribulus, Tubuliflorae and Ranuncula-ceae (not shown in spectra Fig. 2) which are presentin low levels i.e. 1% each. Barring Nigrospora 3.5%,Tetraploa and Diplodia 1.16% each are among thefirst time recorded fungal spores. Other fungalremains add to 2.5%.
4.3.2. PM10 air-catch
This sample exhibits low pollen content of treeswith the highest frequencies of 3.65% of Prosopis
pollen. Others such as Holoptelea 1.57%, Acacia
and Pinus 1.05% each, Alnus and Cedrus 1.02%each are lowly present. Fabaceae 3.15% andEphedra 1.57% are better represented among allother shrubby elements viz., Ricinus, Oldenlandia
and Oleaceae counting to 0.53% each. Poaceae37.8% attains the maximum frequency in inhalableaerosols of Jhunjhunu compared to all othersamples although the size differences do exist indifferent samples. In addition, Chenopodiaceae/Amaranthaceae 14.7%, Cyperaceae 6.83%,Caryophyllaceae 6.31%, Brassica spp. 4.73%and Artemisia 4.20% also form a good fraction ofthe air-borne pollen in the aerospora. Otherssuch as Tubuliflorae and Typha 2% each,Cerealia 1.57% and Solanum 1.05% show scantypresence. Nigrospora 2.57% represents the fungalremains.
5. Discussion
The present investigation on air-catches from thedust storm hit regions has brought out a much-diversified composition of aerospora. The observedvariations in the number distribution of pollen(Fig. 3) are not very systematic rather they arespecies and site specific. It could be due tothe presence of the species (s) in and around thesampling site or due to the wind-assisted transport.The S–SW winds show a significant decrease in thespeed along eastward direction i.e. Delhi region andhence in their carrying capacity. This wind regimecould have resulted in differential pickup, transpor-tation and dispersal of pollen of various plantgroups at local level i.e. at individual sampling siteas well as throughout the whole sampling domainfrom west to east. The difference in particle size(Free-fall—coarse fraction; PM10—finer fraction) isalso responsible for the numerical variation inpollen assemblage as pollen of different plantgroups have different size. The presence of
ARTICLE IN PRESSS. Yadav et al. / Atmospheric Environment 41 (2007) 6063–6073 6071
pollen of trees e.g. Cedrus and Pinus, which arecomparatively bigger in size, in PM10 fraction musthave come due to the physical breakdown of thesepollen during their transportational process bywinds. In general the overall assemblage is char-acterized by the much higher frequencies of non-arboreal (herbaceous elements) in contrast toarboreals (trees and shrubs). The excessive higherfrequencies of Poaceae (grasses) followed byChenopodiaceae/Amaranthaceae, Caryophyllaceae,Brassicaceae and Cyperaceae (sedges) in the aero-spora envisage that the members of these familiesgrow abundantly around the provenance of air-catches. As the particle size decreases the abundanceof these species increases in PM10 sample comparedto other free-fall samples. The high pollen produc-tion of these taxa during summers can be a directreason for their high pollen influx in the atmo-sphere. But these species remain in full floweringblossom in February month, at least 2–3 monthsbefore dust storm period. Therefore, we suggest thatresuspension of the available pollen/spores by theheavy intensity winds/dust storms from the exposedsurface in and around the sampling region could bemore appropriate reason for their higher occurrenceas also suggested by Jaenicke (2005). However, wecannot ascertain the exact reason at this stage. Inaddition, the meager presence of Artemisia, Sola-
num, Tribulus, Xanthium and other herbaceouselements, depict the sparse growth of these plantsaround the provenance of their discharge in the airas well as they might have not been in flowering atthe time of gleaning of the air-catches. Pollen of C.
sativa and Cerealia (cf. Zea mays), allergenicity ofwhich is well known, are present consistently in allof our samples but in variable numbers. Thefrequent retrieval of these in the aerospora of Delhi,in particular, infers the profuse occurrence of theseplants in the wasteland and around the cultivatedarea in the outskirt of the city.
The low and poor presence of arboreal pollen inthe aerospora could be attributed to the scantyoccurrence of trees and shrubs in the vicinity of theinvestigated sites. In Bikaner samples pollen ofAcacia, Syzygium, Ziziphus and Holoptelea aremarked by their low frequencies, except forMoringa, which contributes moderately to theaerospora. However, the atmosphere of Dehli andJhunjhunu is charged frequently with a goodnumbers of pollen of Prosopis (Prosopis juliflora
and Prosopis cineraria). This could be attributed tothe recent plantation P. juliflora as an avenue tree in
Delhi and common presence of P. cinerea in thenatural vegetation in Jhunjhunu. In general, pollenof trees and shrubs increase in Jhunjhunu and Delhisamples. This is due to the increased populationdensity of trees and shrubs in this region as theclimate changes from hot arid to semi-arid andhumid on moving down from Bikaner in the west toDelhi in the east and hence become favorable forvegetation. The encounter of pollen of conifers andtemperate broad-leaved elements such as Pinus,Cedrus, Tsuga and Alnus in the aerospora in all thelocalities implies their transportation by winds fromHimalayan region, where these taxa grow luxur-iantly in the natural vegetation. Furthermore, thewinged nature of conifer pollen and high buoyancyof all such pollen also favor their long distancetransportation by winds from their natural prove-nance to the present area of investigation but theroute map of transport is not clear at this stage. Oneprobable process is through the riverine transport ofsediments from Himalayas to the Thar Desert andthen to N–NW India by S–SW wind-assistedtransport similar to that reported by Yadav andRajamani (2004) based on geochemical studies. Butthe possibility of direct transport through easterliesin the upper troposphere from Tibetan plateau towestern India can also not be ruled out at this stage.At the same time we also envisage other way ofpollen transport of Himalayan origin tonorthern plains of India through N–NW winds inwinter. This last route may not be equally probableas the wind speed is very low and if at all ithappens, it cannot transport to Jhunjhunu site asthe wind pattern does not support this. But thepresence of the pollen of Himalayan origin in ourentire samples strongly point towards a Himalayanconnection to the bioaerosols and hence dust inN�NW India.
The fungal spores are few in number andfrequencies and among them Nigrospora is some-what better represented in contrast to the extremelylow presence of Tetraploa, Diplodia and otherremains. This poor fungal assemblage infers theprevalence of dry weather at the time of air-catchcollection, which does not favour the proliferationof fungi. The allerginicity of some of the pollengrains and fungal spores has been well determinedin India (Vishnu-Mittre and Khandelwal, 1973;Khandelwal, 1991). Among the most commonaero-allergens pollen-Holoptelea integrifolia, Proso-
pis, Amaranthus, Z. mays, Ageratum conyzoides
and fungal spores-Nigrospora, Alternaria, Mucor,
ARTICLE IN PRESSS. Yadav et al. / Atmospheric Environment 41 (2007) 6063–60736072
Rhizopus, Curvularia, Penicillium and Aspergillus
are well known (Yeo and Kim, 2002; Wu et al.,2004). Some of them viz., Holoptelea, Prosopis, C.
sativa, Cerealia (cf. Z. mays), Chenopodiaceae/Amaranthaceae (cf. Amaranthus spinosus) andNigrospora have also been recorded in the aerosporaof Delhi and Rajasthan in variable frequencies. Theknown sources of Nigrospora include decaying plantmaterial and soils (St-Germain and Summerbell,1996). Griffin et al. (2003) and Kwaasi et al. (1998)also observed increased concentration of Nigrospora
in dust samples particularly and in PM10 samplesduring a dust storm period in Saudi Arabia andAfrica respectively.
6. Conclusion
The present study on characterization of pollenand spores in the dust samples and PM10 collectedfrom dust storm hit region of India has revealed thepresence of trees, shrubs, herbs and few fungalspores. The proportions of all these species variedsignificantly from Bikaner to Delhi through Jhunj-hunu site. The population of trees and shrubsincreases from west side (Bikaner) samples to eastside samples just as the climate changes from arid tosemi-arid. The herb species concentration variedfrom species to species. We do not know the reasonfor such variations as well as for the existence of thepollen of grasses (Poaceae) during the summerperiod. The higher frequency of grasses (Poaceae) orherbs could either be a result of the presence of theseherbs in the sampling area and hence the higherproduction of pollen/spores or due to the resuspen-sion from the exposed surface by the high intensitywinds. But we cannot ascertain the exact process atthis stage. The presence of Nigrospora, a highallergen fungal species is noticed in all the samples.There could be potential health risk in the regiondue to the presence of the allergenic species. Thenoticeable similarity in the pollen and sporecharacteristics of the dust samples collected fromthree different locations separated by 600 kmsuggest that the dust have some common sourceand is also biologically related to Himalayas, maybe directly or indirectly. The route map of transportis not clear at this stage. A comprehensiveaerobiological investigation is required to generatedatabase pertaining to periodic presence andabundance of pollen/spores in the atmosphere, totheir sources and to the effects of dust storm. This
will also enable us to ascertain the precise identity ofpollen/ spores responsible for allergic diseases andalso to develop suitable preventive measures againsttheir harmful affects on health.
Acknowledgments
S. Yadav is thankful to DST, GOI, New Delhi,for financial assistance as Fast Track project andProf. V. Rajamani for work at NFGR, JNU.Chauhan and Sharma are thankful to DirectorBSIP, Lucknow, for his help in carrying out thisresearch. SY acknowledges the help of AnkitTandon in MS preparation. Authors duly acknowl-edge the two anonymous referees and editor fortheir help in improving the manuscript quality.
References
Agashe, S.N., Abraham, J.N., 1988. Pollen calendar of
Bangalore city-part I. Indian Journal of Aerobiology 1 (1),
