Calculate PM2.5 D24 Hour Mean


(Emily Saunders) #1

How do I calculate the D24 hour for PM2.5 concentrations when creating the input file for BenMAP? Thanks.


(Neal) #2

Hi Emily,

Thanks for your question. Can you tell me more about the air quality data you’re using? For example, are you using air quality monitoring or modeling data? Are you starting with hourly concentrations?

Thanks,
Neal


(Emily Saunders) #3

Hi Neal,

Thanks for your reply, the data I am using it modeling data from a GEOS-Chem model and the model produces output every 15 minutes. Will I be able to use the output for every hour?


(Neal) #4

Hi Emily,

Yes, you’ll need to average those 15-minute predictions to 1-hour values. For each grid cell, you should have a comma-delimited string of 8760 hourly values. BenMAP-CE will use these to calculate the D24HourMean metric.

You may also want to consult the BenMAP-CE input file specifications workbook, which you can find here.


(Emily Saunders) #5

Hi Neal,

Thank you for the reply, but I have a different question about creating a new data set. I recently completed the exercise using the Nigeria data and it was very informative and helpful. I want to know where can I find the information I need for the create the health impact function data set, pollutant data set, incidence data set, population data set, and valuation method data set.

Emily


(Neal) #6

Hi Emily,

Can you tell me a little more about the questions you would like to answer with the tool? This will help me guide you to the appropriate source of input data.

Best wishes,
Neal Fann


(Emily Saunders) #7

Hi Neal,

The question I would like to answer with the tools is how does PM2.5 exposure impact human health in India during biomass burning events and how does the population influence the health impact estimates. Also, I have a question about creating a new shapefile and the BenMAP input file. I am using ArcGIS to create the shapefile using the model output from a global model (NASA’s GEOS-5 model). The model runs a simulation over the entire globe and the output is produced over the entire globe, so I have to use a tool in ArcGIS to clip the shapefile over the area I want to study. The clipped shapefile has 35 columns and 16 rows and the original output file has 1440 columns and 721 rows. I wanted to know when creating the input file, do I have to use the PM2.5 values for the clipped shapefile column and row even though the output file I am getting the information from includes PM2.5 values for all 1440 columns and 721 rows. Thanks for your help in advance.

Emily


(Neal) #8

Hi Emily,

I just posted to the BenMAP-CE website a step-by-step guide for creating a new BenMAP-CE input grid from a shapefile. Take a look at this link and let me know if it answers your question.

Best wishes,
Neal Fann


(Emily Saunders) #9

Hi Neal,

I do not these steps will work for me. I have been working with Esri to learn how to create the shapefile for BenMAP. The steps I am utilizing to create the shape file are the following:

  1. First I input the netcdf output from GEOS-5 in the ‘Make NetCDF Raster Layer’ tool to create the raster layer since the GEOS-5 model output is a 25km resolution over the entire globe I have to clip the layer over the area I am studying
  2. Then I create a shapefile over the area of study and then clip the raster using the ‘Clip (Data mangement) Tool’, while using that tool I use the original netcdf raster as the ‘input raster’ and the ‘output extent’ is the new shape file I created. After running the tool it clips the raster, which includes the X & Y minimum and maximum and column and row of the clipped area of study.
  3. After that I run the Raster to Polygon tool to create the final shapefile I will use in BenMAP and that tool uses the clipped raster to create the shape file and last I input the column and row information using Libre Office.

Do these steps make sense to you and back to my original question since the clipped shapefile has 35 columns and 16 rows and the original output file has 1440 columns and 721 rows. I wanted to know when creating the input file, do I have to use the PM2.5 values for the clipped shapefile column and row even though the output file I am getting the information from includes PM2.5 values for all 1440 columns and 721 rows. Thanks for your help in advance.

Emily


(Neal) #10

Hi Emily,

Yes, those steps make sense to me. I think if I understand your question, then you only need to supply PM2.5 data for the clipped shapefile–that is, the shapefile containing the subset of grid cells.


(Emily Saunders) #11

Hi Neal,

Thanks for the reply, The clipped shaped does contain the subset of grid cells, but the .dbf file does not include the column and row values for the clipped area. Can I just add the values manually using Libre Office and then upload it into BenMAP.

Emily


(Neal) #12

Hi Emily–yes, you can definitely give this a try. Alternatively, you can import your shapefile directly to BenMAP-CE. The program will assign it a column/row index automatically.


(Emily Saunders) #13

Hi Neal,

Back to my original question about calculating the D24hour mean for PM2.5 using the GEOS-5 model output. I re-ran the model to produce output by the hour instead of every 15 minutes. Now that I have values for every hour, how do I calculate the D24HourMean for multiple days. Do I take the average for all 24 time steps for each day and then take the average or max for the selected days to produce the PM2.5 value that will be placed into the ‘Values’ column in the BenMAP input file. After calculating the D24 hour mean do I multiply the values by 100, so the final value is in significant figures (58.907) not scientific notification (5.8907x10^-01). Also, which units do I have to use for PM2.5. Thanks.

Emily