The Citizen Sense Airsift Dustbox Data Analysis Toolkit, the Airsift PM2.5 Data Analysis Toolkit and the Airsift Frackbox Data Analysis Toolkit make it possible to analyze and download citizen-generated air-quality data points collected in northeastern Pennsylvania, an area that is heavily populated with unconventional natural-gas infrastructure. You can use the Airsift toolkits to explore PM2.5 data and Frackbox data, create plots and identify air pollution problems in northeastern Pennsylvania. Watch the how-to videos below to learn how to use the Airsift PM2.5 Data Analysis Toolkit or the Airsift Frackbox Data Analysis Toolkit, and to create data stories that can be used to identify local pollution sources. You can also access the open source code for Airsift on our Airsift github repository.
Plot and download data
- Select a site from the drop-down menu. For the line graph, you can select a second site if you would like to compare locations. The sites are named by township in Pennsylvania in order to blur exact monitoring locations. At each site, participants used a Citizen Sense Kit, including a Speck device to monitor particulate matter (PM2.5). The Frackbox data was generated from 3 discrete monitoring locations.
- Select a graph or plot from the left-hand menu. More information on the different graph and plot data analysis options is available below.
- Choose a weather condition to analyze.
- Define the time period. You can select the entire monitoring period, or a specific set of days.
- Next, select a mean of 24 hours, 1 hour or 1 minute. The 24-hours mean can be useful for identifying sustained elevated levels of pollutants, and the 1-minute mean can be useful for identifying shorter episodes and spikes in pollutants.
- Click the “Enter” button to plot the dataset.
- Click on the “Download plot” button to save your graph.
- Click on the “Download csv” button to download the data you have analyzed as comma-separated values.
The line graph function generates a time-series plot of 24-hour, 1-hour and 1-minute mean concentrations. The line graph can be useful to generate an overview of a site, and to investigate if there are many occasions when the 24-hour mean of PM2.5 exceeds the WHO guideline of 25 µg m-3. It can also be used to look at changes in levels of NOx and VOCs (with the Frackbox data).
If your graph contains a very high or very low reading that makes the graph difficult to read because the rest of the readings appear very small on the plot, use the date function to exclude this data or to zoom in on a particular time period. This will mean you can see the readings in more detail. Some of the Frackbox readings may show a very low reading -500 or + 500 at the start of the monitoring period, this indicates the time during which the sensor was heating up.
The line graph also allows for comparison between sites. To compare data gathered at any 2 sites, select a first and second site, as well as a time period. The line graphs can be overlaid (single variation), or displayed side by side (multiple variation). If the graph shows an error, there is not enough data to overlay the data, and you should choose the multiple variation option. Use a comparison of 2 sites to investigate if the graph shows regional or local sources of pollution, since regional sources can be identified as broad “humps,” and local sources can be identified as short “spikes” typically affecting one site and not the other.
The scatter plot allows for easy comparison between variables. In order to compare pollutants (with the Frackbox), or pollutants in relation to specific weather conditions for a specific site, select these variables from the drop-down menus. For example, the scatter plot can be used to look at the relationship between wind speed and PM2.5 to identify if elevated levels are present at low winds, thereby indicating possible local source(s). For a more detailed example, see the data stories for specific monitoring locations.
The historical weather data is taken from the Weather Underground location closest to the site selected, and is used to generate the scatter plot.
The polar plot shows concentrations of pollutants in relation to wind speed and wind direction. The polar plots can provide information on source identification, depending upon the wind speed and wind direction. The polar plot will indicate a source of pollution, but unlike the wind rose it will not indicate how frequently it occurs.
Increasing wind speeds can lead to lower concentrations of pollutants. However, in some circumstances wind speeds may increase pollutant concentrations, particularly through suspension of particulate matter.
The polar plot function includes the option to plot all PM2.5 Speck data for a specific site, as well as to plot PM2.5 Speck data above 15 µg m-3 so that higher concentrations can be analyzed (in relation to the baseline for regional or background pollution sources).
To use the polar plot, first select a site, then select the time period to be analyzed. Historical wind speed and wind direction data from the Weather Underground location closest to this site is used to generate the polar plot.
The timeplot shows when “spikes” in pollution occur by grouping concentrations by hour, month and day of the week. Sources of pollution related to commuter or transit traffic typically show peaks in concentrations coincidental with peaks in traffic flow, i.e., morning and evening rush hour with notably lower levels at night and on Sundays. These charts can be used to match patterns in the occurrence of spikes with working patterns of particulate-generating activities in the area.
To use the time plot, first select a site, then select the time period to be analyzed.
The wind rose plot shows wind direction and wind speed, with “paddles” indicating percentage of time when wind is from a particular angle and speed.
The “quantiles” for PM2.5 indicate the concentration of particulate matter within particular wind speeds and angles. To use the wind rose plot, first select a site, then select the time period to be analyzed. Historic wind speed and wind direction data from the Weather Underground location closest to this site are used to generate the wind rose plot.
The calendar plot shows the daily mean concentration of a specific pollutant of the selected site in a calendar format. Select a site, and dependent upon whether there is sufficient data for that site, a calendar with daily mean concentrations will be generated.
It is possible to assemble the above plots and graphs together to understand pollutant levels at particular monitoring locations and times, and in relation to possible sources. View the data stories, developed by Citizen Sense, which show the potential of citizen monitoring and citizen data analysis. The data stories were developed in collaboration with atmospheric scientist Dr Benjamin Barratt and digital designer Raphael Faeh.