An Introduction to Lidar & Forestry

May 2013

Lidar technology

Derivatives from point clouds Applied to forestry

Publish & Share

Futures

Introduction to Lidar & Forestry

Lidar – Light Detection And Ranging

Different Types of Lidar • Atmospheric Lidar

• Bathymetric: Senses up to 50 meters deep BUT requires clear water!

• Terrestrial: Tripod mounted or mobile

• Airborne Laser Scanner (ALS)

Lidar Light Detection And Ranging

• Similar in concept to Radar & Sonar

• Measures distances (through laser pulses) that strike and reflect off of features on the surface of the earth

• Converts scanning angle and distance-from-sensor information into georeferenced data points

• Current sensors can collect hundreds of thousands of positions each second (‘point clouds’)

energy or echoes of light

ALS - Lidar Systems

Airborne Lidar System Components

• Scanning laser emitter-receiver unit

• Differentially-corrected GPS in the plane

• Inertial measurement unit (IMU)

• Computer to control the system and store data

• Imagery often collected simultaneously Color and/or Color Infrared Imagery

IMU pitch

roll yaw

X

y z GPS

Lidar – Characteristics

Lidar Data Characteristics

• A “Return” is a portion of the Lidar pulse that is reflected back to the sensor

• Most laser systems can record several returns or “multiple returns” for each pulse

- Multiple returns occur when the laser beam is only partially blocked

- The remaining laser energy continues downward until it is reflected back by the next feature

• Up to 4 returns per pulse, but typically only receive 2-3 returns

Lidar Data Characteristics

• Lidar point classification

1st Return (Canopy Ht)

Intermediate Return

Intermediate Return

Last Return (Ground)

1st returns (69%)

Lidar Data Characteristics

2nd returns (26%)

Lidar Data Characteristics

3rd returns (4%)

Lidar Data Characteristics

4th returns (<0.1%)

Lidar Data Characteristics

All returns (100%)

Lidar Data Characteristics

All returns (100%)

Lidar Data Characteristics

1st returns (69%) 2nd returns (26%) 3rd returns (4%) 4th returns (<0.1%)

Lidar Data Characteristics

Lidar Data Resolution (PPM) • The only consistent measure of Lidar data resolution (when

considering the full point cloud) is the number of pulses per unit area or Pulse Density (pulses/m2).

• Return density can vary depending on the target being scanned.

Lidar Data Characteristics

• Lidar Data Resolution (PPM) is dependent on:

• Laser scan rate (more pulses @ same speed à higher resolution)

• Flying height and speed (higher or faster à lower resolution)

• Side-lap coverage (e.g. 50% side-lap will result in 2x resolution)

1–Meter 1–Meter

Lidar – Pulse Density & Products

• Low Pulse Density (≤ 1 Pulse/m2) - Product: Moderate Resolution Topographic

Products (≥ 2 meter Grid)

• Moderate Pulse Density (1-3 Pulses/m2) - Products: Stand Level Vegetation Metrics (e.g.

canopy height, canopy cover) and High Resolution Topographic Products

• High Pulse Density (≥ 3 Pulses/m2) - Products: Forest Structure

$$

$$$

$$$$

• Slope (flat vs. steep) • Ground cover (pavement vs. trees) • Acquisition altitude • Quality of instruments • Capabilities of provider • Best case: +/- 5cm vertical

Lidar Data Characteristics Spatial accuracies depend on:

Lidar – Point Data • True 3D Data!

True 3D Data!

Lidar | Point Data

Lidar –Point Data Return data colored by height

LIDAR Intensity Image

© Copyright 2003. Optech Incorporated. All rights reserved.

Lidar – Point Clouds to Derivatives

9 Tiles: N.W. Montana, U.S. • Collected 2009 • PPM: +/- 4 • 32 million points • Approximately 1,600 acres/650ha • Almost 1GB

NAIP Aerial Photography • Collected 2009 • 1 meter per pixel

Data Thanks Go To • Montana State Library & NRIS for

making the dataset available and the acquiring organizations including: Montana DNRC, Lake County, the Flathead Basin Commission, and the City of Whitefish.

Our Sample Data Set

Lidar Point Clouds to Derivatives

Digital Elevation Model (DEM) • Bare Earth (DTM) • Contours

Can also be viewed in 3D

Lidar| Point Cloud to Derivatives

Digital Surface Model DSM

Lidar Point Clouds to Derivatives

•Digital Surface Model (DSM)

Lidar Point Clouds to Derivatives

• Canopy Height Model (CHM)

• After generating the 1st return (DSM) and bare earth (DEM) rasters, use the Minus geoprocessing tool to determine the difference

• The difference represents the canopy height.

Lidar Point Clouds to Derivatives

• Canopy Density Model (CDM)

• The ratio of all returns to ground returns as seen from the air

• Applications include the estimation of biomass, forest extent and condition, and biodiversity

• Slope

• Aspect

• Hydrography

• Stand Delineation

• Roads (historic, new design)

• Cultural features (i.e. buildings)

Additional Derived Layers

Lidar Applications for Forestry

• Topographical analysis • Roads: Design and operations • 3D viewshed analysis • Quantify forest stand characteristics

- Biomass estimates - Fire fuel modeling

• Landslide analysis • Change over time

Roads: Evaluation & Design

• Creating Profile Transects: - Assess construction complexities - Quickly identify areas of concern due to slope

• Calculating Road Slope - Create road vector (2D or 3D) - Use existing 2D road vectors

• Mapping Side Slope Risk - Identify locations where upslope (rock, mud, snow

slides) or downhill hazards may exist

Roads: Creating Profile Transects

• Use ArcMap to digitize proposed routes and quickly assess any issues relating to slope

• Easily done with existing 2D road vectors

Roads: Calculating Road Slope • Display based on maximum road segment slope

Viewshed Analysis

• The Goal: View the proposed cutblocks from multiple observation positions to determine potential visual impacts

Roads: Viewshed Analysis Remove the vegetation within the proposed cutblock(s)

• Clip the surface values inside the cutblock polygon(s) from the DEM

• Replace the cells in the DSM with the values removed from the DEM. This modified DSM represents the resulting “surface” after completing harvest activities (assumes all trees within the cutblock are removed)

Viewshed Analysis

• Using the modified DSM, run the 3D Analyst tool Viewshed with a selected observer position.

• The Output is a raster map

• Repeat from other observer positions

Viewshed| Analysis

3D Visualization | Inspect the point cloud

3D Visualization | Profile View & Measure

3D Visualization | Measure & Inspect Individual Trees

Publish & Share

• Take your Lidar data into the field. Go mobile.

• Easy to do with derivative data sets. Leave your point clouds at home.

• Publish with ArcGIS Server

• Share with ArcGIS On-Line

Lidar – Procurement Considerations

• What will the Lidar be used for? - Pulse density - Spatial accuracy - Laser sensor scan angle - Point cloud classification - Flight line overlap (i.e. 50%)

• Deliverables?

• Combine with high resolution imagery? Fused?

Futures

• Existing Lidar Technology - Faster pulse rates - Multiple sensors in one box

• New Lidar Technologies

• Better Feature Extraction Software - More automation - Improved results

• Change Detection (PC to PC)

Lidar For Forestry - Summary

• Airborne LIDAR is a tool that provides Foresters with: - Highly accurate topographical data - Forest stand information including canopy

height and density - 3D visualization and measurements of the

ground and land cover - A source of accurate GIS information, even in remote areas - A quantifiable and repeatable data source

• One flight = Many GIS data layers (derivatives)

Acknowledgements

• U.S.D.A. Forest Service Pacific Northwest Research Station - Bob McGaughey and Steve Reutebuch

• U.S.D.A. Forest Service Remote Sensing Applications Center (RSAC) http://fsweb.rsac.fs.fed.us - Brent Mitchell

• Optech Incorporated