Canopy Analytics

An Interactive Resource for Tree Discovery in the Portland Metro Area

Jackson Voelkel, Geospatial Research Analyst   |  

SUPR Lab   |   Portland State University

Portland Metro Area Trees

Green, Green, Green

Important questions:

  • How do we measure these trees?
  • What trees are important (what is 'important', and to who)?
  • How do we identify trees most in need of preservation?
... without climbing each one!
~180' up a 270' Douglas Fir

Data

LiDAR

  • Airborne Laser Scanning/Mapping
  • ~12 points/m² on flat surface
  • ~21cm point spacing
  • ~63,600,000,000 (63.6 Billion) points in most recent flight

LiDAR Products

  • Digital Elevation Model (DEM)
  • Digital Surface Model (DSM)
  • Digital Height Model (DHM)
  • Canopy Density * Height (CDM or "Biomass")
Incoming solar radiation (insolation) measured with and without Canopy
Toggle trees on/off [click and hold]

Exploring the Urban Forest

46ft

The Tallest Tree in Portland

252ft     |     Street View

N Williams

137ft     |     Street View

Powell Butte

201ft  [126]     |     Street View

Oswego Lake

200ft     |     Street View

Portland State University

Source: Portland State University Archives (1962)

Portland State University

Source: Portland State University Archives (1975)

Portland State University

Source: Portland State University Archives (1978)

Portland State University

126ft     |     Street View

Eastmoreland Sequoias

156ft     |     Street View

Arnold Creek

212ft     |    4 of 11 Private 200ft+ trees    |     Street View

Forest Park / Linnton

242ft     |     Google Earth

The Grove

200-242ft  [118]     |     Google Earth

What about every tree?

Steps:

  • Identify every tree crown.
Toggle tree crowns on/off [click and hold]

Steps:

  • Identify every tree crown.
  • Describe each tree crown.

Machine Learning-derived evergreen/deciduous classification (~90% accuracy). Collaboration with Metro. Open Data.

Informing Statistical Models with Tree Data

Air quality - NO2 Modeling

Rao, M., George, L. A., Rosenstiel, T. N., Shandas, V., & Dinno, A. (2014). Assessing the relationship among urban trees, nitrogen dioxide, and respiratory health. Environmental Pollution, 194, 96-104.
Estimation of NO2 reduction due to targeted tree plantings in Washington County.
Shandas, V., Voelkel, J., Rao, M., & George, L. (2016). Integrating High-Resolution Datasets to Target Mitigation Efforts for Improving Air Quality and Public Health in Urban Neighborhoods. International Journal of Environmental Research and Public Health, 13(8), 790.

Urban Heat Islands

Multi-temporal traverse-based modeling

Predictive Heat Surfaces Estimated with Machine Learning Alogrithms

Final UHI Models
Voelkel, J., V. Shandas, and B Haggerty (2016). Developing High-Resolution Descriptions of Urban Heat Islands: A public health imperative. Preventing Chronic Disease.

Assessing the Role of Canopy in Temperature Reduction

In the City of Portland we see trees alone reducting the temperature in localized areas by up to 15°F!

Current Progress of UHI Research

Initial processing of Portland Metropolitan Area UHI model. Finalized version expected by November 2016.

Current Progress of UHI Research

Initial processing of Portland Metropolitan Area UHI model. Finalized version expected by November 2016.

The Canopy Analytics Tool

We know where all of the trees are!

https://climatecope.research.pdx.edu/canopyanalytics/

Moving Forward

What do we hope to accomplish?

  • Increase public knowledge of the impact of trees.
  • Educate policy makers to the connections trees have to the environment
  • Nominate for Herritage Tree status that play major roles in:
    • Reduction in the Urban Heat Island Effect
    • Reduction in air pollutants

Future research

  • More modeling!
    • Tree species descriptions
    • Explore the relationships between tree type and Air Quality / UHI
  • New LiDAR: change detection
    • permitted removals
    • ... non-permitted removals
  • Move beyond the Portland Metropolitan Area

Eugene, Oregon UHI Research

Eugene Weekly coverage of research.

Questions?

Thank you!