Advanced search
Publication Cover
Canadian Journal of Remote Sensing
Journal canadien de télédétection
Volume 41, 2015 - Issue 3
Submit an article Journal homepage
1,501
Views
22
CrossRef citations to date
0
Altmetric
Original Articles

Estimating Forest Site Productivity Using Airborne Laser Scanning Data and Landsat Time Series

Piotr TompalskiFaculty of Forestry, University of British Columbia, 2424 Main Mall, Vancouver, V6T 1Z4, BC,, CanadaCorrespondence[email protected]
,
Nicholas C. CoopsFaculty of Forestry, University of British Columbia, 2424 Main Mall, Vancouver, V6T 1Z4, BC,, Canada
,
Joanne C. WhiteCanadian Forest Service (Pacific Forestry Centre), Natural Resources Canada, 506 West Burnside Road, Victoria, V8Z 1M5, BC,, Canada
,
Michael A. WulderCanadian Forest Service (Pacific Forestry Centre), Natural Resources Canada, 506 West Burnside Road, Victoria, V8Z 1M5, BC,, Canada
&
Paul D. PickellFaculty of Forestry, University of British Columbia, 2424 Main Mall, Vancouver, V6T 1Z4, BC,, Canada
Pages 232-245 | Received 18 Nov 2014, Accepted 12 May 2015, Published online: 11 Sep 2015

Figures & data

Study area location. The hatched areas indicate where ALS was acquired in 2012.
Study area location. The hatched areas indicate where ALS was acquired in 2012.

Table 1 List of Landsat scenes used in the study

Table 2 ALS data characteristics

Figure 2 A representative subset of the study areas on which a method of finding sample units (100 × 100 m) in the detected stand-replacing disturbances is demonstrated. A: Disturbance areas (stands) detected with Landsat time series; B: Stands are generalized; C: The borders of the stands are refined using CHM; D: 100 × 100 m sample units are placed inside the stands.
Figure 2 A representative subset of the study areas on which a method of finding sample units (100 × 100 m) in the detected stand-replacing disturbances is demonstrated. A: Disturbance areas (stands) detected with Landsat time series; B: Stands are generalized; C: The borders of the stands are refined using CHM; D: 100 × 100 m sample units are placed inside the stands.
Figure 3 Distribution of time since disturbance (TSD) by area, for the sample units (n = 2006).
Figure 3 Distribution of time since disturbance (TSD) by area, for the sample units (n = 2006).
Figure 4 Histogram of dominant height calculated for the sample units (n = 2006).
Figure 4 Histogram of dominant height calculated for the sample units (n = 2006).
Figure 5 Guide curve fitted in the height-age measurements (95% confidence intervals presented as dashed lines).
Figure 5 Guide curve fitted in the height-age measurements (95% confidence intervals presented as dashed lines).

Table 3 Result of the model parameter estimates

Figure 6 Derived site productivity classes plotted together with the height-age data used to construct them (dashed lines indicate 95% confidence intervals).
Figure 6 Derived site productivity classes plotted together with the height-age data used to construct them (dashed lines indicate 95% confidence intervals).
Figure 7 Site productivity map derived with the developed model. Three example areas (A, B, and C) shown in larger scale to demonstrate the spatial variability of the site productivity classes.
Figure 7 Site productivity map derived with the developed model. Three example areas (A, B, and C) shown in larger scale to demonstrate the spatial variability of the site productivity classes.

Table 4 The results of the comparisons of site productivity estimates

Figure 8 Scatterplots presenting the comparison of the productivity estimates in three sets. SI – site index calculated with equations used in British Columbia and input data from forest inventory (SIINV) or from chronosequence of dominant height values (SICS); SP – site productivity calculated with the developed model and input data from forest inventory (SPINV) or from chronosequence of dominant height values (SPCS); SIINV32 – site index calculated with inventory data for the reduced based age (32 years). In all sets n = 120.
Figure 8 Scatterplots presenting the comparison of the productivity estimates in three sets. SI – site index calculated with equations used in British Columbia and input data from forest inventory (SIINV) or from chronosequence of dominant height values (SICS); SP – site productivity calculated with the developed model and input data from forest inventory (SPINV) or from chronosequence of dominant height values (SPCS); SIINV32 – site index calculated with inventory data for the reduced based age (32 years). In all sets n = 120.
Figure 9 Distribution of the productivity class difference values (model-derived class – reference site class).
Figure 9 Distribution of the productivity class difference values (model-derived class – reference site class).

Related Research Data

Forests
Source: MDPI AG
Digital Aerial Photogrammetry for Updating Area-Based Forest Inventories: A Review of Opportunities, Challenges, and Future Directions
Source: Springer Science and Business Media LLC
Modelling Site Index in Forest Stands Using Airborne Hyperspectral Imagery and Bi-Temporal Laser Scanner Data
Source: Multidisciplinary Digital Publishing Institute
Estimating Tree Volume Distributions in Subtropical Forests Using Airborne LiDAR Data
Source: Multidisciplinary Digital Publishing Institute

Linking provided by 

Related research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.