Skip to content

Slides and workshop materials for the 2026 Air Sensors International Conference

License

Notifications You must be signed in to change notification settings

MazamaScience/ASIC-2026

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Workshop on R packages for Air Quality Data

Last updated February 04, 2026


This workshop will introduce participants to two open source R packages designed specifically for working with air quality data from regulatory monitors and low-cost sensors. The AirMonitor and AirSensor2 packages have been developed with funding from, and have been used by the US EPA, the US Forest Service and the South Coast Air Quality Management District. These R packages are part of a suite of R packages that provide core functionality for environmental monitoring at fixed locations. The full suite of packages represent a decade of continuous development with a focus on compact data formats, robust data analysis, compelling data visualization and a simple, easy-to-learn coding style.

Goals

Attendees will become familiar with the AirMonitor and AirSensor2 R packages and will be able to quickly download, process and visualize data from AirNow and OpenAQ as well as sensor manufacturers PurpleAir and Clarity. Various analysis and data visualization functions will be introduced but the focus will be on data access and the creation of uniform dataframes of hourly measurements appropriate for use in research studies and reports.

Audience

The R packages presented are designed for individuals who sometimes need to work independently (without the support of IT staff) and provide functionality for data ingest and manipulation. The target audience includes anyone who works with Air Quality data from regulatory monitors and low cost sensors and who is a regular user of R/RStudio. Attendees should have a basic understanding of R data types and common functions. Familiarity with the dplyr package will be especially helpful.

R Packages for Environmental Time Series

Over the last decade, Mazama Science created multiple open source R packages focused on environmental monitoring time series. These are used operationally in data processing, analysis and visualization systems at the US Forest Service AirFire Team, in the EPA AirNow Fire & Smoke map and elsewhere.

Jonathan Callahan currently maintains the following R packages:

Each package has a dedicated Slack channel for announcements, support and to help build communities of practice around these shared tools. You may request an invitation to join from [email protected].

Workshop Preparation

Attendees are strongly encouraged to install the required packages and data files before the workshop begins.

As with any hands-on workshop, advanced preparation by individuals will allow us to quickly start working with data. Please ensure that the following software and data files have been installed on your laptop in advance of the workshop:

R and RStudio

This workshop is designed for those who are already familiar with R and RStudio and use them on a regular basis.

Ensure that you are running R version 4.4 or higher.

Ensure that you have installed RStudio Desktop version 2025.09 or higher.

TODO: install appropriate packages with pak

CRAN packages

Most (but not all) of the packages we will be using are available on CRAN. The easiest way to install these is to install the AirMonitor (>= 0.4.5) package which will recursively install all dependency packages.

At the RStudio Console prompt type:

install.packages("AirMonitor")
...

Non-CRAN packages

A few of the packages have not yet made it to CRAN. For these, you will need to install them directly from GitHub. Tools in the devtools pacakge make this very easy.

At the RStudio Console prompt type:

install.packages("devtools")
...
devtools::install_github("mazamascience/AirMonitorPlots")
...
devtools::install_github("mazamascience/AirSensor2")
...

Spatial data

The MazamaSpatialUtils package is used to enhance spatial metadata when ingesting sensor data. Simplified spatial datasets for countries and timezones are installed with the package but other, larger datasets must be installed manually in a dedicated directory. Work with the AirSensor2 package requires installation of dastasets with state and county boundaries.

The default location for spatial data used in this workshop is underneath your home directory in ~/Data/Spatial/. The following chunk of code will create this directory and populate it with the required datasets.

At the RStudio Console prompt type:

dir.create("~/Data/Spatial", recursive = TRUE)
MazamaSpatialUtils::setSpatialDataDir("~/Data/Spatial")
MazamaSpatialUtils::installSpatialData("NaturalEarthAdm1")
MazamaSpatialUtils::installSpatialData("USCensusCounties")

Workshop repository

Finally, you should install all of the documentation, data and scripts for this workshop by cloning this repository.

From RStudio

You can create a new project from a GitHub repository by using the RStudio menubar:

  1. File > New Project...
  2. Version Control > Git
  3. URL: https://github.com/MazamaScience/ASIC-2026.git

Choose a directory into which to clone the repository, check "Open in new session" and then press "Create Project".

From a terminal

If you prefer, you can use a separate terminal window or the RStudio Terminal window to issue a shell command to clone the GitHub repository.

At the terminal prompt, type:

git clone [email protected]:mazamascience/ASIC-2026.git
...

If you don't have a git account or can't log in, you may need to use the https location:

git clone https://github.com/MazamaScience/ASIC-2026.git
...

Congratulations!

You have finished installing all required software for this workshop.

About

Slides and workshop materials for the 2026 Air Sensors International Conference

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages