Vegetation encroachment is a term used to describe the growth of trees, shrubs, or other vegetation near or around utility lines. This can pose a risk to the infrastructure, as vegetation growth can interfere with powerlines, communication lines, or other utility lines. Vegetation encroachment on or near utility lines is a serious concern for power utilities, as it can cause power outages, damage infrastructure, and potentially dangerous situations for utility workers and the public.
Utility companies often use different methods to manage vegetation encroachment on utility lines. However, with tools from ArcGIS, you can analyze lidar point clouds with tools that find powerlines and towers, low, medium, and high vegetation, model powerline sway, visually identify and perform classification, and understand the spatial relationship between utility lines, towers and poles, and the nearby vegetation.
A key piece of technology includes machine learning and Esri’s pretrained deep learning models. ArcGIS pretrained deep learning models eliminate the need for huge volumes of training data, massive compute resources, and extensive artificial intelligence (AI) knowledge. The following workflow will use two models to find trees and powerlines within lidar point cloud data.
Workflows
The process of determining if vegetation is encroaching upon utility lines typically starts with lidar data. ArcGIS offers a range of tools that enable the classification of lidar data and facilitates the identification of various features such as ground, buildings, trees, and powerlines. Once the lidar has been correctly classified, you can assess if the powerlines are too close to the vegetation.
To classify powerlines and vegetation, you must install deep learning libraries for ArcGIS, and download the pretrained deep learning models for classifying powerlines and trees.
Classify lidar
First, bring the point cloud data into ArcGIS Pro as a LAS dataset layer. Often lidar will be delivered from a surveyor with some existing classification; most commonly, basic classification may include noise, ground, and buildings.
To verify if the lidar has any classification, you open the dataset properties from the Catalog pane, select the Statistics pane and review the point count per classification code. Alternatively you can symbolize the LAS dataset points by the LAS classification code and visually check if the lidar is classified correctly.
If the data has not been previously classified or the classification is incorrect you can use the lidar Classification tools in 3D Analyst to classify noise, ground and buildings. Next, use the Classify Point Cloud Using Trained Model geoprocessing tool to classify trees and powerlines.
Assess vegetation encroachment
There are several workflows in ArcGIS to understand vegetation encroachment. You can use the Extract Power Lines From Point Cloud tool to extract 3D powerline features from the lidar. These 3D line features can be used as input for modeling a clearance zone using the Generate Clearance Surface geoprocessing tool.
If you need to understand vegetation encroachment at scale (city-wide), it is recommended that you use a raster-based approach that creates rasters for the powerlines features and vegetation areas. This approach not only helps in identifying encroachment sites but also in prioritizing them based on whether the vegetation encroachment is above the wire, below the wire, or intertwined with the wires. This saves costs and optimizes the process of vegetation management for utility companies.
Considerations
The quality of the results of the encroachment analysis is dependent on the quality of your lidar. Factors such as lidar point spacing and lidar classification can influence your results.
The recommended VRAM for running training and inferencing deep learning tools in ArcGIS Pro is 8 GB. If you are only performing inferencing (detection or classification with a pretrained model), 4 GB is the minimum required VRAM, but 8 GB is recommended.
Required software
You'll need ArcGIS Pro with ArcGIS ArcGIS Enterprise to manage, classify, and analyze point clouds for vegetation encroachment. To classify powerlines and vegetation, you must install deep learning libraries for ArcGIS. You may also want to use ArcGIS Online or ArcGIS Enterprise to share point clouds and analysis results as 3D scene services.
Explore the following resources to learn more about understanding vegetation encroachment in ArcGIS.
ArcGIS help documentation
Reference material for ArcGIS products:
- What is lidar data?
- LAS datasets
- How to store lidar data
- Use lidar in ArcGIS Pro
- Convert point cloud data
- Share 3D point cloud data across the platform
- Deep learning in ArcGIS
- Classification toolset in 3D Analyst
- Deep learning classification toolset in 3D Analyst
- Deep learning frequently asked questions
- Extract Power Lines From Point Cloud geoprocessing tool
- Generate Clearance Surface geoprocessing tool
ArcGIS blogs, stories, and technical papers
Supplemental guidance about concepts, software functionality, and workflows:
Videos
Esri-produced videos that clarify and demonstrate concepts, software functionality, and workflows:
Tutorials
Guided, hands-on lessons based on real-world problems:
Esri training
Authoritative learning resources focusing on key ArcGIS skills:
Developer resources
Resources and support for automating and customizing workflows:
Esri community
Online places for the Esri community to connect, collaborate, and share experiences:
- Ask the 3D community questions.