You need to determine the trajectory of these elements in the air.
Much easier to use a low volume system, collected on Teflon membrane filters and do XRF. The key question is whether you are looking at local or distant sources. For local sources, you can combine the measured concentrations and local wind data to use tools like CPF and NRP to look at the directions of the major sources. See Atmospheric Environment 38 (2004) 4667–4673 for details.
Try having a look at these two papers and follow the references whenever you need more details:
Reis et al. Science of the Total Environment 407 (2009) 5726–5734
Almeida et al. Science of the Total Environment 521–522 (2015) 411–420
I think these will give you an overall picture of various possibilities, related to : (1) direct sampling and measurement using fast, multi-elemental nuclear analytical techniques; (2) direct measurement using biomonitoring (either passive or active using transplants); (3) tracking air mass trajectories for predicting elements long range origin.
You need to use a High volume sampler, to collect the sample, followed by acid digestion of filter and content before measurement with Flame emission photometry for Ca and Flame atomic Absorption spectrophotometry for Lead. You can also use X-RAY Fluorescence straight away.
As has been suggested, running a high volume sampler and using either nuclear techniques (Ion Beam Analysis) or acid digestion approaches will get you the information about the chemical (ionic or elemental) composition of the aerosols. The other problem is attributing these pollutants to local or remote sources. Under stable, or near stable atmospheric conditions, the nocturnal measurement footprint is quite limited (some 10s of km in radius), during the day, however, when the atmospheric boundary layer is deep and well mixed, aerosol contributions can come from much further afield. If you had a high-volume sampler operating for two different parts of the day (say from 2-8am and 10am-4pm), it might then be possible to more easily attribute your observations to local or distant sources. For the samples collected during the middle of the day you could then run back trajectories to determine the most likely fetch regions. If you would like any information about ways to categorise nocturnal stability conditions for such an approach, I'd be happy to discuss that separately. Cheers, Scott.
