Maps, datasets, code and web applications from the laboratory’s research on urban forests, wetlands, grasslands, wildfire and rehabilitated land, and on hyperspectral image analysis.
01
Google Earth Engine app · PlanetScope 3 m · 2020 and 2025
Land use and land cover maps of the Toronto and Region Conservation Authority (TRCA) jurisdiction, produced from 3 m PlanetScope imagery with deep learning. The app shows the 2020 and 2025 maps side by side, together with the training samples behind them.
02
Brampton, Ontario · Sentinel-2 and WorldView
Land use / land cover and wetland subclass maps of Brampton, Ontario, classified from Sentinel-2 and WorldView-2/3 imagery, shared together with the image composites, reference wetland data, accuracy assessments and temporal stability tests.
03
Drone remote sensing · southern Ontario · 2023
Maps of fractional green and dead vegetation cover for three rehabilitated surface-mining sites in southern Ontario, estimated from monthly drone imagery collected from May to August 2023, with the field plot measurements used to build them.
04
Hyperspectral imagery · 301 channels · 0.3 m
Three 301-channel hyperspectral images of suburban, urban and forest landscapes in southern Ontario, collected and pre-processed by the RSSEM Lab, with manually digitized reference polygons for classification and accuracy assessment.
05
Landsat 30 m · aerial photographs · tree rings · Region of Peel
Open data and code from Dr. Mitchell Bonney’s research with Prof. Yuhong He: yearly 30 m tree canopy cover maps for 1972–2020, residential canopy and inequality, 2018 municipal canopy maps, tree density since 1944 and tree-ring records for the Region of Peel, Ontario.
06
Central Yukon · Landsat and ERA5-Land · 1986–2020
Open data, a trained XGBoost model and Python code from a driver analysis of wildfire severity (dNBR) for 70 large fires in the central Yukon, Canada, combining Landsat burn severity with climate, topographic and vegetation variables.
07
Airborne imaging spectroscopy · Koffler Scientific Reserve · 2016 and 2017
Airborne hyperspectral images, leaf and canopy trait measurements, topographic data and analysis code for a mixed grassland in Ontario, used to study how topography shapes plant functional traits and species’ responses to the 2016 drought.
For data requests and research collaboration, please contact Prof. Yuhong He.
Email the labEach dataset is shared under the licence given on its product page and in its repository: Creative Commons Attribution 4.0 (CC BY 4.0) or CC0 1.0. Code is released under MIT, GPL-3.0 or CC0 1.0. Please cite the associated publication when using them.
