Smart Solutions for a Resilient Urban Forest Presentation.pdf

Explore how cutting-edge technologies including GIS, remote sensing, LiDAR, AI, ML and data analytics, can be harnessed to design, manage, and maintain resilient urban forests. By integrating these tools, cities can better plan for climate change, enhance biodiversity, and improve the overall health of urban green spaces.

1. Smart Solutions for Resilient
Urban Forests:
Leveraging Technology for a
Sustainable Urban Forest
Josh Behounek
Davey Resource Group

2. Right
Decision,
at the
Right Time,
on the
Right Tree

4. Feedback Loops

6. • Warming winters &
summer heatwaves
• Zone 8a to 9a
• Heavy floodingwith
intermittent drought
• More hurricanes
Louisiana
Davey.com/climate

8. Software & Hardware Constraints

9. 1900
• mastectomy
1967
• lumpectomy
1980’s
• Chemotherapy
Current
• Targeted
therapy

10. Current Process

11. Unnecessary Data Collection with Standard Fields

12. Reactive

13. Subjective

14. Feedback Loop -Buffalo, NY Inventory Update
Difference 2014 2001
2,635 127,080 124,445 Sites
-53,546” 817,627” 871,173” Total DBH
-1” 6” 7” Average
DBH
-34 247 281 # Species
2,039 2,707 668 # Removals
-4,142 44,619 48,761 # Planting
Sites

15. Feedback Loop -Condition Change Assessment
2001
Inventory
2014
Inventory
5,698
Poor
3 Dead
145 Poor
11 Fair 0 Good
293 Plant
5,246 New
38,199
Fair
298 Dead
3,445 Poor 26,830 Fair
962 Good
1,365 Plant 5,299 New
25,632
Good
259 Dead 717 Poor 8,952 Fair
9,783 Good 1,878 Plant 4,043 New

16. Individual Trees

17. Static Pull Testing

18. Static Pull Testing

19. Static Pull Testing

25. Analysis & Recommendations

26. Analysis & Recommendations

27. Analysis &
Recommendations

28. Sonic Tomography

29. Technologies Used in Tree Evaluations:
Ground Penetrating Radar
October 5, 2020
Ground Penetrating Radar

30. Tree Sensors & 
Internet of Things •Smart Landscape 
•Watering efficacy
•Clients can view data generated by 
sensors in their landscape
•Measure:
•Soil moisture –2 depths
•Soil temperature –2 depths
•Ambient temperature –for Nature 
Clock
•Tilt
•Notifications, useful links, etc
Photo @banksy

33. Urban Forests

34. Smart Tree
Inventories

36. Tree
Tree
Tree

37. Step 2: Create a 4D Digital Tree Twin of each tree
4D DIGITAL TW IN
Multispectra l Satellite Images
Panoram ic Images
Point Cloud

38. Step 3: We analyze each tree and extract information
Clearance Issues
Live Crow n Ration
% Dieback
Ecological Benefits
Digital Tree Twin
Change Over Time
Cohort Analysis
Size (DBH, Height, etc)
Species
Leaf Area Index
Leaning Angle

39. Step 4: Define outliers Absolute Outliers (cohorts) ●Dead trees
●Too much lean
●Leaf Area compared to Size
●Canopy Width vs Tree Height Relative Outliers (filtering) ●Trees > x”
●Trees in certain neighborhoods
●Certain species of trees
●Trees >40% dieback

40. Step 4: Define outliers Absolute Outliers (cohorts) ●Dead trees
●Too much lean
●Leaf Area compared to Size
●Canopy Width vs Tree Height Relative Outliers (filtering) ●Trees > x”
●Trees in certain neighborhoods
●Certain species of trees
●Trees >40% dieback

41. In Field 25-100%
Remotely 0-20%
Step 5: Davey Arborists assess Outliers

42. Outlier Assessments

44. Base Information –Species, Size, etc.

45. Health & Vitality

50. Trunk Stability Index (beta)

51. Trunk Stability Index

52. Machine Learning
Advantages
●
Objective
●
Repeatable
●
Efficient
●
Precise

54. Removing Implicit Bias

55. The Future Is Now!
Smart Solutions for Resilient
Urban Forests
Josh Behounek
[email protected]
573-673-7530