itree-support-tools

This repository provides a workflow for preparing municipal tree data for i-Tree Eco analysis and extrapolating the results to full the study area extent, using lidar-segmented tree crowns and auxiliary GIS datasets.

---

Code is provided for the following tasks:

1. i-Tree Eco Data Preparation: preparing an input dataset for i-Tree Eco analysis by supplementing existing municipal tree inventories with crown geometry from the ALS data and auxiliary spatial datasets following the workflow by Cimburova and Barton (2020).

2. i-Tree Eco Extrapolation: extrapolating the outputs from i-Tree Eco analysis to all trees in the study area following the workflow by Cimburova and Barton (2020).

The repository is applied on the Norwegian municipalities: Bærum, Bodø, Kristiansand and Oslo.

---

Project Installation and Configuration

The code is build in an ArcGIS Pro 3.1.0. conda environment with the spatial analyst license enabled.

1. Clone the repository.

2. Open Project structure to view the structure of the project.

3. Set up your Python Environment:

   a. Create a new conda environment using the environment.yml file or clone the arcgispro-py3 environment from your ArcGIS Pro installation and install the required packages listed in the requirements.txt file.

       cd path/to/project/folder
       conda env create -f environment.yml
       conda activate project-name

   b. (Optional) Install linters using pipx

       # install linters using pipx
       make install-global
       # test linters
       make codestyle

   note: As pre-commit unfortunately often gives acces-denied errors on Windows OS, I would recommend to run make codestyle command before you commit your changes. This command runs black, isort and ruff on all files using the configuration specified in the pyproject.toml file.

   c. Install as a local package

       pip install .
       pip install -e . # for development

   • installs project packages in development mode
   • creates a folder package-name.egg-info

   d. Configure your project.

   • Copy template.env to $user/.env and fill in the variables. ENSURE THAT YOU DO NOT COMMIT .ENV TO THE REPOSITORY
   • check that your data is located in the correct folders, look at the Project structure and the Catalog for more details.

   e. Define your municipality in the parameters file.

   d. Run config.py in the conda env to test your project config.

---

Workflow | i-Tree Eco Data Preparation

Detailed description of the workflow is provided in the project note (in prep).

1. Prepare Data entry point: prepare_data.py tasks: (i) load the lidar-segmented tree crown polygons from the ALS data per neighbourhood (ii) load the in situ tree stems from the municipal tree inventory (iii) clean the in situ tree stems - manual municipality-specific cleaning tasks (see REF) - automatic cleaning tasks: - set standard field design - translate tree species - ensure that each tree stem contains: stem_id, dbh, height, crown_diameter (iv) group tree stem points by neighbourhood

2. Join the in situ tree stems with the lidar-segmented tree crowns entry point: join_data.py tasks: (i) classify the geometrical relationship (ii) split lidar-segmented tree crowns that overlap with multiple tree stems (iii) model the crown geometry of tree stems that do not overlap with lidar-segmented trees (iv) quality control wether each crown polygon is assigned to a single tree stem (v) join the in situ tree stems with the lidar-segmented tree crowns

   Geometrical Relations:

   • Case 1: one polygon contains one point (1:1), simple join.
   • Case 2: one polygon contains more than one point (1:n), split crown with voronoi tesselation.
   • Case 3: a point is not overlapped by any polygon (0:1), model tree crown using oslo formula.
   • Case 4: a polygon does not contain any point (1:0), not used to train i-tree eco/dataset for extrapolation.

3. Compute tree attributes and auxillary attributes

   entry point: compute_attributes.py tasks: (i) compute tree crown attributes (all trees in thes study area) - overlay attributes (pollution zone, neighbourhood code) - crown_id (based on neighbourhood code and objectid) - tree height, crown area (ii) compute tree stem attributes (in-situ trees) - overlay attributes (e.g. pollution zone, neighbourhood code, land use) - tree attributes (e.g. dbh, height, crown diameter) - join crown attributes (e.g. crown_id, crown area, crown volume, crown shape) - building related attributes (e.g. building distance, building direction) - crown condition (e.g. crown light exposure)

   IMPORTANT NOTES: do not run building related attr. and crown condition attr. within pipeline. Run them separatly and cosely check the results.

   STEP 3 NEEDS CLEANING, BUILDING CROWN CONDITION SUPER SLOW (e.g. +12h runtime)

---

Workflow | i-Tree Eco Extrapolation

Detailed description of the workflow is provided in the project note.

---

References

• Cimburova, Z., & Barton, D. N. (2020). The potential of geospatial analysis and Bayesian networks to enable i-Tree Eco assessment of existing tree inventories. Urban Forestry & Urban Greening, 55, 126801. https://doi.org/10.1016/j.ufug.2020.126801

Acknowledgments

This repository is part of the project:

TREKRONER Prosjektet | Trærs betydning for klimatilpasning, karbonbinding, økosystemtjenester og biologisk mangfold.

This repository uses code adapted fromt the repository i-Tree-Eco by Cimburova, Z. 2022, this repository is licensed under the GNU General Public License (GPL).