Breathe Buffalo is a dashboard-style web application developed for the UB Air Monitoring project. This dashboard is a key dissemination product for the community-based participatory research (CBPR) study focused on air quality in Buffalo’s African-American community, particularly on the east side of Buffalo, New York.
The east side of Buffalo’s African-American community faces a disproportionate burden of chronic disease compared to neighboring regions. Poor air quality is a well-known contributor to adverse health outcomes, but until recently, it has been difficult to assess community-specific air pollution exposures due to a lack of local monitoring. To address this gap, we partnered with the Buffalo Center for Health Equity (BCHE), a local environmental justice organization, and launched a two-year air monitoring initiative using CBPR principles. Prototypes of study results were developed to report fine particulate matter (PM2.5) and volatile organic compound (VOC) levels. An interactive data dashboard—this very application—was iteratively co-designed and refined with input from BCHE members and study participants through virtual meetings and online surveys. This process emphasized the use of color, large font for key results, and a preference for maps over traditional plots, ensuring that the dashboard is accessible, engaging, and tailored to community needs.
The dashboard serves as a central tool to communicate findings from a ten-month community air monitoring campaign. Future dissemination efforts will also include public education events and virtual meetings to share results with the broader community.
- Real-time Air Quality Monitoring: Visualize sensor data (PM2.5, VOC, etc.) on an interactive map using Leaflet.
- Zipcode-based Measurements: View air quality metrics by zipcode, with gauges and bar charts for selected regions.
- Intuitive UI: User-friendly interface built with shinydashboard (no sidebar, clean layout).
- Spatial Data Integration: Utilizes the
sfandleafletpackages to display and facilitate the analysis of spatial datasets. - Rich Visualizations: Supports various R visualization packages including plotly, ggplot2, DT for tables.
- Location Map: Real-time air quality by region, visualized on a map.
- Air Quality Gauges: Gauges showing both the selected zipcode and city-wide air quality.
- Color Ramp: Visual color ramp indicating air quality levels (
Goodin green toHazardousin red).
The following R packages are required:
- shiny, shinydashboard, shinybusy, flexdashboard, lubridate, leaflet, DT, sf, data.table, base64enc, jsonlite, DBI, RPostgres, plotly, ggplot2, rsconnect
The project uses renv for dependency management:
# In the repo root
renv::restore()Store sensitive information (e.g., DB credentials) in .Renviron and place it in the repo root.
# From the repo root
shiny::runApp()or run app.R directly.
- Use
config.Rfor deployment settings (rsconnectto deploy to a server). - Easily run and deploy from within RStudio.
- Continuous Deployment:
When you push to therepo-mainbranch, the app is automatically deployed to shinyapps.io via GitHub Actions.
Breathe-Buffalo/
├── app.R # Main Shiny application code
├── config.R # Deployment (rsconnect) configuration script
├── renv/ # R package dependency management (renv)
│ ├── activate.R
│ └── ... # Other renv infrastructure files
├── renv.R # renv bootstrap/init script
├── data/ # Project data assets
│ └── zip/target.gpkg # Geopackage: spatial boundary/target data
├── .Renviron # (Not committed: for local secrets/env vars)
├── LICENSE # License file
└── README.md # Project documentation (this file)
- Some files, such as
.Renvironare referenced in the code but not committed to version control. - The
renv/folder may contain additional files (lock, settings, cached packages) created by the R package management system. - Additional files or folders may exist for logs, temporary files, or user-generated content if applicable.
See the LICENSE file for details.