Enhancing forest disturbance monitoring with ALS data integration

Jens Wiesehahn

Northwest German Forest Research Institute

Background

• to provide context

• forests are facing a number of challenges

  • increased demands on forest (recreation, biodiversity, timber)
  • reduced staff (more area per forester)
  • climate change

Climate Change

• severe drought stress

• 2018 and following years showed severe droughts in Central Europe

Forest Disturbances

• severe drought stress

amplified massive bark beetle outbreaks in Germany

conditions amplified bark beetle outbreaks

• Harz 2017: Forestry world still fine
• Harz 2022

Forest Disturbances

• severe drought stress

amplified massive bark beetle outbreaks in Germany

• lead to large-scale forest disturbances in recent years
• lead to a strong need for more information to manage disturbances

Disturbance Monitoring

• severe drought stress

amplified massive bark beetle outbreaks in Germany

caused a need for information

shows disturbances we mapped in our project

• more than 500.000 ha (5 %) forest loss (2018-2020) (Thonfeld 2022)
• 255 mio. m³ affected timber (2018-2022) (BMEL)

Disturbance Monitoring

terrestrial (sample-based)

e.g. National Forest Inventory, Forest Condition Survey

established systems
detailed data (many attributes)
no information on local level
low repition rates
costly

robust statistics on national level

remote (satellite-based)

e.g. Waldmonitor DE, Forestwatch-DE, FNEWS

sparse additional information
wall-to-wall maps
high repition rates
low-cost

disturbance maps on national level

Airborne Laser Scanning

barely used in (German) forestry until now

detailed data
low to medium repition rates
costly ($)
wall-to-wall maps
on medium to large scale

enhancing existing systems

• systems currently applied

• what can we do with ALS in disturbance monitoring?

  • detailed data on local level
  • covering large areas
  • but not frequently available

Pre-Disturbance Data

differentiate between als data collectes before and after disturbance

Canopy Height Model

• standard product

Retrospective analysis

• delineate basic stand characteristics

valuable information

e.g.

• stand structure of affected vs unaffected stands

Canopy Gaps

Retrospective analysis

• unstocked forest areas prior to disturbance

easy delineation

differentiate disturbance from already unstocked areas

these might be false positives

Individual Trees

Retrospective analysis

• disturbance characteristics on the tree level
  • size
  • number
  • composition
  • …

intersect disturbances with individual trees

Species Group

Retrospective analysis

• species group based on
  • Spatial point metrics
  • Intensity (under certain conditions)

Understory Vegetation

Retrospective analysis

• vertical stand structure

Current Management

• potential regrowth
• prioritize afforestation

• use cases befor might work with aerial images
• lidar data shines at providing further information below the canopy

Terrain

Retrospective analysis

• site conditions

Current Management

• site conditions
• infrastructure network
• trafficability

high resolution topographic data provied valuable informatio - site conditions (wetness, solar radiation) - which stands were affected - future management - plan harvesting operations - existing forest roads / skid tracks - unaccessible areas -

Example

Sample site

• 4.4 ha
• disturbance between 2018 and 2021
• by windthrow / bark beetle
• detected with Sentinel-2

Example - Volume

Retrospective analysis

• 853 trees
  • 104 deciduous
  • 749 evergreen
• 196 trees/ha
• 27.28 m avg. height
• 0.9 m³ avg. volume
• 851 m³ volume

Example - Trafficability

Current Management

• 254 m forest road (58 m/ha)
• 768 m skid tracks (176 m/ha)
• terrain obstacles

Example - Site index

Retrospective analysis / Current Management

• wetness
• solar exposure

Post-Disturbance Data

Post-Disturbance

Retrospective analysis

• disturbance evaluation

Current Management

• existing regrowth
• potential seeding trees

Processed

• operation evaluation
  • skid trails (FSC-conform?)
  • wood piles

Post-Disturbance

Retrospective analysis

• disturbance evaluation
• operation evaluation
  • skid trails
  • wood piles

Current Management

• existing regrowth
• potential seeding trees

Unprocessed

• log detection
• operation planning
  • skid trails

• based on neighbourhood metrics of points
• could be done with aerial imagery as well for stand replacing disturbances

Post-Disturbance

Retrospective analysis

• disturbance evaluation
• operation evaluation
  • skid trails
  • wood piles
• log detection

Current Management

• existing regrowth
• potential seeding trees
• operation planning
  • skid trails
  • trafficability

• but can also be done below canopy for non-stand-replacing disturbances

Multitemporal Data

• rarely available (in desired period)
• pre- / post-disturbance information

plus

• direct change detection

Conclusion

ALS data provides valuable information for

1. Retrospective disturbance analysis
2. Ongoing disturbance management

(combination with satellite-based systems for timely disturbance detection seems benefitial)

wiesehahn.jens@gmail.com

JensWiesehahn

Slides available under:

https://wiesehahn.github.io/presentation-silvilaser23