After a year of field sampling, I have collected a total of 104 samples of PM2.5. However, due to various reasons, the number of samples given for sampling came out to be 40. I have performed ICP-OES for heavy metals and GC-MS for PAH analysis. Is there any source apportionment technique which can be used satisfactorily for the given number of samples?
AAS or ICP you can use for heavy metal analysis.
ICP OES is best technique here u can analyse more elements at a time.
The number of samples needed for the source apportionment really depends on the number of variables you have measured and the complexity and similarity of the source signals
You can try Principal component analysis (PCA) and Cluster analysis (CA) to identify possible sources of heavy metals.
There are different suggestions about the lowest number of samples you need to be able to apply PMF. Usually, what people suggest is that you require at least 50, but it is suggested to have 100. Since the number of variables in your case is relatively low (14), you can also give PMF a shot, but I would also suggest PCA. In any case, what you need to consider is if the variables that you have available (metals and PAHs) can provide enough information to effectively identify PM sources. You can find more info and suggestion about source apportionment (including the lowest number of required samples) here: https://ec.europa.eu/jrc/en/publication/european-guide-air-pollution-source-apportionment-receptor-models
You can start with Principal Component Analysis (PCA), and can also try Positive Matrix Factorization (PMF).
By any chance, If you also have data or planning to conduct carbon fractions and ionic analysis, then Chemical Mass Balance (CMB) is a widely used technique for source apportionment.
All the best...
- Badges
- Science topic
- Similar topics
- Geoscience
- Atmospheric Sciences
- Atmosphere