Data Summary

Allegheny County -Tree Canopy Change, 2015-2020

2024 - Allegheny County


Metadata | Download | Preview | KMZ | GeoJSON | Add to ArcMap: Image or Feature | Add to ArcGIS Pro

API
REST:

WMS:


ADDITIONAL RESOURCES
n/a


ABSTRACT
The purpose of this project was to conduct an assessment of tree canopy change that occurred between 2015 and 2020 utiliing LiDAR data and a previously prepared canopy dataset. Comprehensive canopy change statistics are provided for various geographies down to the parcel-level scale. Tree canopy was extracted from LiDAR data collected in the fall of 2019 and spring of 2020 in ArcMap. A digital surface model (DSM) was created by interpolating the maximum values of the first returns of each laser pulse across a 3-foot grid surface (raster). A speckled output was created because some pulses can entirely or partially pass-through tree canopy before detecting a return, so maximum focal statistics in a 3 by 3 rectangular grid window were applied to the DSM to create a smooth surface. Another raster representing the elevations of solid surfaces which LiDAR does not penetrate - usually ground and buildings, but occasionally dense evergreens as well, was created by interpolating the minimum values of the last returns (which are also the first return in instances of single return). Mean focal statistics in a 3 by 3 cell window were applied to this raster. The last return raster was subtracted from the first return raster, creating a canopy height model (CHM) – a representation of the heights of objects with complex return structures above the ground. In addition to trees, this includes built structures such as power lines, poles, transmission towers, gantries, etc. The edges of buildings also appeared in the CHM as a result of different cell assignment and focal statistics types applied to the first and last return rasters. The heights of dense evergreens were underestimated due to the inability of LiDAR to penetrate to the ground for a proper base for height. A constant raster of CHM cells with a height greater than 15 feet was created. Holes less than 500 square feet were filled to eliminate dubious small gaps while preserving discernable canopy gaps. This raster was then shrunk by 2 cells and expanded back by 2 cells. This process eliminated narrow or small features such as building edges, power lines, and poles. This raster was then converted into a vector polygon format for editing.