radarusesradiowaves.Combiningparticle counting withbioaerosol measure-
mentsmayallowdetectionof rapidvariations instantaneously and indicate
furthermicrobialmeasurements.No absolute relationship, however,can be
established between bioaerosol and particle counts in variable working
environments (Paratet al., 1999).
37.5.7 Conclusionsregardingsamplers
In Table37.1 microbiological air samplers, collectionmethods and a few
comments on the methodsare presented. In addition to the samplersin Table
37.1,thereis other commercialapparatus on the market; scientificinformation,
however, is oftennot available.In the foodindustry, the settleplates and
impactor- or centrifugal-type hand-held samplers are commonly used air-
samplingmethodsin routinemonitoring. The Andersen impactorgivesthe most
reliableresults,but it is less practicalfor routineuse. The filtration methods
maynot be optimalfor counting vegetativecellsdue to the stressit placeson
cellsthroughdehydrationduring sampling.A glass impingeris inexpensive and
simpleto operate, but viabilityloss may occurdue to the amount of shearforces
involvedin collection (Kang& Frank,1989b,c).The impingersare likelyto
perform betterthan the filtration methodsfor sampling airborne bacterial
microbes (Liet al., 1999).Impingementis suitablefor yeastbioaerosols (Lin&
Li, 1999a)but fungal spores are more reliably collected withfiltrationthan
impingement.Lowerlevelsof totalbacterialrecoverywereobservedin filter
samplers than impinger samplers. This was primarily due to the higher
biological stressoccurring duringthe samplingprocess of filtration. Sampling
flowrate did not significantlyaffect bacterialrecovery.Sampling timerelated
to dehydration effectdid play a rolein bacterial recoveries,especially for
sensitive strains. The nucleopore filtration method was recommended for
mouldsand thermophilicorganismsratherthanthe AGI-30,but for yeastsand
mesophilic bacteriathe AGI-30 methodwas morefavourable(Thorneet al.,
1994).
Eachof the proposed monitoringmethodshas limitationsthat the user should
be awareof. Monitoring effectivenessis dependent to a greatextenton the
monitoringmethodsusedand the natureof the aerosolpresent (Kang& Frank,
1989a). Foodplantaerosols havenot beenstudiedsufficientlyto accurately
generalize aboutparticle-sizedistribution. Particle density is influencedby RH
and particlecomposition.It can be statedthat an effectiveseparationvolume
variesmore due to the massof the particlethanto the particlesize.Bioaerosol
resultsmeasured withdifferentsamplertypesare strongly influencedby the
samplingenvironment and the results are sometimes contradictory. The choice
of instrument is clearlydependent on the experimental parameters and the
secondarymeans of identification (Neefet al., 1995).Airbornecounts recorded
fromdifferentsamplingtechniquesare not directlycomparablebut wouldbe
expected to showsimilartrendsin levelvariation(Holahet al., 1995).
Improving air sampling 631