How can I screen and clean hourly PM2.5 from urban monitoring stations before entering the data to BenMAP-CE?


(Bayat) #1

Hello,
Thank you for your supports.
I think if we enter incorrect data to BenMAP-CE, we can’t expect correct results.
I have hourly measurements of PM2.5 from 10 monitoring stations for 5 years.
Before entering the data to BenMAP-CE, my first challenge is data cleaning. By comparing and looking to temporal plots of the data, it is obvious that there are some outliers. Each year has 8760 records; So, visual checks are not sufficient. Is there any spatio-temporal data screening for air pollution data? Consider that some of the data are correct and should not be deleted by ordinary cleaning methods.
I have found below report. But it is not obvious. I want to understand the steps and then write a VBA macro in excel for data cleaning. I’ve had some idea like using LOOCV method; but I have not found any reference for it.

“A Tool for the spatio-temporal screening of AirBase Datasets for abnormal values”
link

Thank you,
Reza Bayat
r.bayat@ut.ac.ir