Interpolation method + Background concentration


I have a couple of questions that would like to discuss:

  1. Interpolation method: we use Voronoi Neighbor Averaging for interpolate 19 stations in Bangkok Metropolitan Region. Input file was daily average (so, around 365 value for a year). Would BenMAP interpolate using daily value (24h-average value)?
    Currently, we input time series of almost 10 years data. When we analyze daily data before entering BenMAP, the values are consistency over years. However, after interpolation, the values of grid among years are large different. Any issues we should closely look?

  2. At some point, during BenMAP training for roll-back analysis, 5.8 was used as background value. May you recommend us a reference for that? or would you recommend a background value for Asian countries?

Thank you for your reply in advance.



Hi Thao,

Thanks for your questions. Responses below:

1a. If you are importing daily PM values, you will need to supply the program with a full set of 365 values exactly. So, if you are missing values for any day of the year, I recommend filling in those missing values (perhaps using the average of the distribution to impute the missing values).
1b. The unusual distribution of PM values may be due to the issue above. After filling in the missing values, re-interpolate the data and see whether the spatial distribution looks more reasonable.
2. 5.8 ug/m3 corresponds to the lowest measured level (LML) of the Krewski et al. (2009) long-term PM mortality study conducted in the U.S. I believe it also corresponds to the lowest concentration at which the Integrated Exposure Response function quantifies effects. If you choose to use this LML, you could cite Krewski et al. 2009 and/or Burnett et al. 2014.

I hope this helps!


Hi Neal,

Thank you very much for your rely. Very helpful answer!
Thanks for important note that we should provide full set of values.