Kitchener’s urban forest is made up of individuals and groups of trees within an urban area. They can be understood as dynamic green infrastructure that provides cities and municipalities with environmental, economic and social benefits. Urban forests are forests for people. The information in this gallery is being shared with the community to help with discussion as the City starts to develop an urban forest strategy and sustainable urban forest program. To learn more about this project please refer to the Background Document – Developing a Sustainable Urban Forest Program.
.png?v=1741428548399)
Do you have a tree that is special to you?
Do you have a special tree in your life that you adore? We'd love to hear all about why trees hold a special place in your heart. Take a moment to share your personal connection with trees, and especially, share a story about the wonderful trees right here in Kitchener. Your unique perspective adds to the rich tapestry of our urban tree stories, and we can't wait to celebrate the importance of trees through your experiences! Add your favourite tree here
Open data layers useful for Tree modelling
- Leaves-off aerial RGB imagery (10-15 cm) from 2012 to 2021.
- An updated tree inventory (points) including tree species of city-owned trees (not including woodlots and trees on private lands) and metadata
Mapping urban trees across North America with the auto arborist dataset
Google Research has introduced the Auto Arborist Dataset, a groundbreaking urban tree classification dataset comprising over 2.6 million trees from 23 North American cities. This dataset, notably larger than previous efforts, leverages public tree census data, Street View, and overhead imagery to create a diverse and comprehensive resource for training machine learning models. The project addresses the critical need for urban forest monitoring, emphasizing the potential for machine learning to democratize this process and make it more accessible to under-resourced cities. The dataset, which includes Kitchener, enables the evaluation of model performance across diverse geographic regions, shedding light on the challenges of domain generalization in urban tree classification.
Find out more at Google Research.
Developing a cost-effective approach to mapping tree canopy
The Spatial Analysis Laboratory at the University of Vermont introduced a practical method to map trees in cities using computer vision techniques (segmentation). Trees in cities are crucial for things like managing storm water, cleaning the air, and providing habitat. However, many cities lack good tree maps. This method uses existing data like LiDAR and imagery to create cost-effective and detailed tree maps. It's flexible and can handle different data types. The paper showcases examples from the United States and Canada, illustrating that this approach can create detailed maps for large areas. Notably, data from Kitchener was used to train and test this method, making it applicable to diverse urban landscapes.
Find out more at Remote Sensing.
