3. Airborne LiDAR requires purchase, but offers a number of advantages; Airborne LiDAR requires purchase, but offers a number of advantages; Spatial resolution typically <=1m; Spatial resolution typically <=1m; Identification of features beneath the canopy; Identification of features beneath the canopy; Potentially a supplement to forest inventory. Potentially a supplement to forest inventory.

5. Ground Canopy No understoryUnderstory Stand boundary Example data: Dillwyn, VA (2008)

6. RW19 (Appomattox, VA) - a region wide study examining the effects of fertilization and thinning in mid-rotation stands. Dillwyn (Appomattox, VA) RW18 (Brunswick, VA) - a region wide study designed with the objective of understanding optimal rates and frequencies of nutrient additions for rapid growth in young stands. Henderson (Vance, NC) - It was established in 1982 with the objective of monitoring the effects of soil management practices on soil structure, organic matter and nutrient contents, and pine growth. SETRES (Scottland, NC) – The aim of the study was to quantify the effects of nutrient and water availability on above and below ground productivity and growth efficiency in loblolly pine VA NC

8. Tree canopy Understorey Ground

9. In some cases the estimate can be improved with detecting curves between max and min (zero-crossing) Height to the living crown

10. Lorey’s Average height(m) Average HTLC(m) Henderson (Vance, NC) RMSE1.161.24 RW18 (Brunswick, VA) RMSE0.511.00 SETRES (Scottland, NC) RMSE1.021.17 RW19 (Appomattox, VA) RMSE0.691.02 Dillwyn (Appomattox, VA) RMSE0.551.08

11. Example comparison plot was extracted from 50m to the west of original. (Left) vegetation control; and (right) no-control. absence presence

13. Stem removal only + SPD Herbicide Stem removal only + CD Herbicide Whole tree removal + CD Herbicide Whole tree removal + SPD Herbicide Example data: Henderson, NC (2008) Non-herbicide

15. Crown top location And Horizontal area (m 2 ) Stem no. per plot RMSE: 5-10 Note: Omission may be suppressed trees. Example data: Henderson, NC (2008)

16. Extract height bins for ITC object Produce vertical profile Height (m) returns Canopy Height to canopy Map individual height to living crown

17. Height (m) returns Estimate crown volume Canopy Top height (m) Bottom height (m) Horiz. Area (m 2 ) Volume (m 3 ) [Cylinder] % no. returns in living canopy 24.717.59.669.760.8% Sample metrics per crown: Map crown geometric volume

18. Height(m) No. returns 1x1m

19. Ground Canopy No understoryUnderstory Stand boundary Example data: Dillwyn, VA (2008)

20. Canopy Height modelUnderstory Height model Example data: Dillwyn, VA (2008)

21. Canopy Height model Understory Height model Example data: Henderson, NC (2008)

22. Understory Height modelUnderstory density class (by proportion of returns) Example data: Henderson, NC (2008)

23. LAI is defined as the leaf area per unit of ground surface area – it ranges from 0 (bare ground) to over 10 (dense forests).

24. LAI indirectly sampled using Li-Cor LAI 2200 under and above the forest canopy; at dawn and dusk; two 15m transects were conducted at each field plot location; horizontal measurements were taken every 1m; vertical measurements were taken at: 0 m; 1.5 m; 2.5 m. 0m 1.5m 2.5m

25. Duke forest Parkers tract Parker Tract – 8 sites: High levels of understory; Similar overall structure between plots. Duke Forest – 20 sites: Variety of stand ages; Variety of stem spatial arraignments; Varying levels of understory.

26. Metrics extracted: Proportion of returns vs. total(%), Max, min, mean, STD, variance, CV, skewness and kurtosis for : Proportion of returns vs. total(%), Max, min, mean, STD, variance, CV, skewness and kurtosis for : ALL, vegetation (>0.2m), ground (<0.2m), canopy and understory layers (variable - stratified by layers); LiDAR return number. Percentiles (5 th …95 th ) for All and Veg. Percentiles (5 th …95 th ) for All and Veg. Cumulative fractional cover (5 th …95 th percentiles) Cumulative fractional cover (5 th …95 th percentiles) Canopy and understory layer(s)* top, middle and bottom heights; (*some merged) Canopy and understory layer(s)* top, middle and bottom heights; (*some merged) Canopy openness Canopy openness LiDAR penetration index and ratios of vertical profile elements (stratified by All, Canopy and Understory returns); LiDAR penetration index and ratios of vertical profile elements (stratified by All, Canopy and Understory returns); Canopy density slices: +/-5m of canopy mode: mean, STD, variance and CV. Canopy density slices: +/-5m of canopy mode: mean, STD, variance and CV.

27. Modified equation of Leaf area index proxy (Mosdorf et al, 2006) stratified by VP layer: e.g.

28. yModel R-squared Metrics used LAI (0 m)0.745FR_UND_LAI + VEG kur + CRW skw LAI (1.5 m)0.827FR_UND_LAI + VEG kur + no_layers LAI (2.5 m)0.759FR_UND_LAI + UND med + no_layers Preliminary results:

29. Modified equation of Leaf area index proxy (Mosdorf et al, 2006) stratified by VP layer: Model Summary RR SquareAdjusted R Square Std. Error of the Estimate.785.616.601.627 LAI (1.5m) From 2013 data the relationship between LAI (1.5m) and ULAI proxy appears linear. And accounts for ~60% of the variation.

30. ITC analysis can estimate stem number; ITC analysis can estimate stem number; And potentially provide more individual tree level metrics; LAI can be estimated statistically – where input metrics are linked to canopy layers. LAI can be estimated statistically – where input metrics are linked to canopy layers.