Trees, Forests and People xxx (xxxx) xxx

Contents lists available at ScienceDirect

Trees, Forests and People

journal homepage: www.sciencedirect.com/journal/trees-forests-and-people

Forest canopy height mapping using ICESat-2 data to aid forest

management in a Canadian Arctic community: A case study of Kluane First
Nation, Yukon, Canada
Ravi Darwin Sankar a,* , Finella Pescott b , Kate Hicks c,d , Maribeth S. Murray e
a
Department of Biological and Environmental Sciences, Longwood University, 201 High Street, Farmville, VA, USA
b
Fortitude Consulting, Haines Junction, Yukon Territory, Canada
c
Kluane First Nation, Yukon Territory, Canada
d
Two Worlds Consulting, Victoria, British Columbia, Canada
e
Department of Anthropology and Archaeology, University of Calgary, 2500 University Dr. NW, Calgary, Alberta, Canada

A R T I C L E I N F O A B S T R A C T

Keywords: As an essential indicator of a forest’s growth capacity and rate, the increased accuracy, ease of access, processing
Canopy height and visualization of canopy height information can facilitate a targeted range of strategies for sustainable forest
Remotely sensed data management, especially among citizen scientists and community members. Here, forest canopy height is esti­
Forests
mated for several land parcel segments in a subarctic locale using ground-based measurements as well as photon
and elevation data obtained from NASA’s Ice, Cloud, and land Elevation satellite (ICESat-2). ICESat-2 offers a
comprehensive view of vegetation structure and provides a unique opportunity to quantify forest canopy height
changes, productivity and distribution in remote locations where it is often arduous and cost prohibitive to
acquire ground data. Average canopy heights returned from ICESat-2 data compared with field measurements of
above-ground biomass yielded a R2 of 0.53, and root mean square error of 1.45 m, amplifying the use and po­
tential value of this dataset and novel platform for multiple user groups interested in forestry mapping and
ongoing monitoring of forest canopy height with increasing frequency to facilitate community-led decision
making. This study demonstrates the utility and benefit of integrating remotely sensed data and field-based
survey measurements to generate complementary information related to forest structure and diversity in this
region.

Introduction Givnish et al., 2014).

In addition to providing information on forest ecosystem health,
As natural carbon sinks, forests play an integral role in influencing canopy height is closely correlated with numerous important environ­
climate processes and mitigating the effects of climate variability mental processes such as biomass allocation, water supply, and species
(Bonan, 2008; Lin et al., 2020). Increasing land use and land cover diversity (Macarthur and Macarthur, 1961; Moles et al., 2009; Tao et al.,
changes fueled by population growth, agriculture and exurban devel­ 2016; Zhang et al., 2016; Campbell, 2021; Enterkine et al., 2022; Shufan
opment are acting as catalysts to forest degradation and deforestation and Chun, 2022). Quantitative information on canopy height at an
resulting in amplified concentrations of atmospheric greenhouse gas improved spatial and temporal scale can provide insight into the amal­
emissions (McKinley et al., 2011). Further, the increasing frequency, gamation of factors that contribute to forest well-being and vegetation
spatial extent, and intensity of natural disturbances such as insect in­ productivity (Thomas et al., 2008; Boutsoukis et al., 2019). Additionally,
festations, soil fertility alterations, thawing permafrost, changes in the canopy height is an essential parameter for a variety of applications in
duration of growth season and weather-related phenomena (e.g., intense the fields of natural resource management, timber harvesting, risk
storms, flooding, wildfires) are working in concert to increase the mapping, urban planning, prescribed fire burns and flood modeling
vulnerability of these terrestrial ecosystems over time (Bratkovich,1994; (Hunter et al., 2013; Ahmad et al., 2017; Sintayehu et al., 2020).
Kashian et al., 2006; Smithwick et al., 2007; McKinley et al., 2011; Standard methods for obtaining estimates of forest canopy height

* Corresponding author at: Department of Biological and Environmental Sciences, Longwood University, 201 High Street, Farmville, Virginia, 23901, United States.
E-mail address: sankarrd@longwood.edu (R.D. Sankar).

https://doi.org/10.1016/j.tfp.2024.100760

Available online 17 December 2024

2666-7193/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).

Please cite this article as: Ravi Darwin Sankar et al., Trees, Forests and People, https://doi.org/10.1016/j.tfp.2024.100760
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

have generally involved the acquisition of data through costly, time dry and declining moisture levels (Cook et al., 2019) related to signifi­
consuming and labor-intensive field measurements (Korhohen et al., cant reductions in spring snow cover extent (Notarnicola, 2022)
2006; Boutsoukis et al., 2019). Contemporary advances in active remote contributed to a record shattering 2023 wildfire season in which more
sensing techniques, however, have facilitated the frequent collection of than 25 million acres of primarily boreal forest were burnt (Jones et al.,
large-scale canopy elevation data that enable quick analyses, visuali­ 2024) (Fig. 1). In addition to other factors (e.g. soil quality, species di­
zations, and planning efforts geared towards the sustainable manage­ versity and topography), quantitative estimates of canopy height can
ment of forests (Lefsky et al., 2005; Zhang et al., 2016; Khalsa et al., further our understanding of overall forest vulnerability and regenera­
2022). Previous studies (Hudak et al., 2002; Parker et al., 2004; tive capacity following a major natural disturbance, accentuating its
Andersen et al., 2005; Zhang et al., 2016; Garcia et al., 2018; Xu et al., importance as a tool for environmental management and mitigation
2019; Potapov et al., 2021; Enterkine et al., 2022) reveal the benefits of (Pierce et al., 2012; Campbell et al., 2019).
using spaceborne light detection and ranging data (LiDAR) and Landsat The launch and implementation of NASA’s Ice, Cloud, and land
data to measure and map forest canopy heights. More recently, multi­ Elevation Satellites (ICESat-1 and 2 in 2003 and 2018 respectively)
spectral imagery and Synthetic Aperture Radar (SAR) data have been aided the collection of precise elevation measurements using space-
employed for forest height estimation at higher spatial resolutions based LiDAR technology to assist in monitoring environmental varia­
(Lefsky et al., 2005; Simard et al., 2011; Urbazaev et al., 2018; Fagua tions over several surface types, including analyses of global vegetation
et al., 2019). structure. (Neuenschwander and Magruder, 2016; Markus et al., 2017;
Recent changes to Canada’s dynamic landscapes and ecosystems in Popescu et al., 2018; Narine et al., 2019; Parrish et al., 2019; Liu et al.,
response to the effects of climate variability and insect-related outbreaks 2020; Sun, Qi and Huang, 2020). In addition to providing exceptional
are particularly apparent in the interior reaches of Canada’s northern spatial coverage of Canada’s boreal forests, there are no orbital or data
boreal forests (Box et al., 2019; Campbell et al., 2019; Reid et al., 2022). acquisition constraints associated with ICESat-2 (Mulverhill et al., 2022;
In northwest Canada, anthropogenic-related disturbances in addition to Sothe et al., 2022). A photon-counting laser altimeter (Advanced
climate-induced alterations have been key drivers in increasing the areal Topographic Laser Altimeter System—ATLAS) is the only apparatus that
extent of natural boreal forest loss (Bolton et al., 2013; Zhang et al., is ascribed to the satellite and measures the return time of individual
2015). Increasing atmospheric temperatures coupled with abnormally photons (i.e., traveling from ATLAS to Earth, and returning to ATLAS) to

Fig. 1. Graphs of the annual number of forest fires and hectares burned in Canada over the period 2000 to 2023.

2
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

evaluate surface height along track (Smith et al., 2019; Campbell 2021; 2 mission in order to underscore and highlight the potential of this
Enterkine et al., 2022). The ATLAS instrument has a coverage area dataset and web-based discovery workspace for detailed scale (tens of
diameter of about 12 m and collects measurements at 70-centimeter meters) vegetation mapping, by analyzing the canopy heights for an
(cm) intervals along the satellite’s ground path. ICESat-2 repeats its array of land parcels belonging to Kluane First Nation (KFN). KFN is one
tracks every 91 days, allowing end-users to observe and monitor of eleven Self-Governing First Nations in the Yukon. In 2022, KFN
elevation changes at various times during a calendar year. Moreover, an developed a Forest Resources Management Plan to ensure their ability to
ATL08 land and vegetation product facilitates canopy height estimation exercise sovereignty over their natural resources on selected parcels of
at along-track segments of 100 m × 12 m (cross-track) (Liu et al., 2020) self-governing settlement lands in order to identify the availability of
and complements the suite of surface-specific data products that can be timber to meet the increased demands for firewood and for local con­
specifically employed to enumerate biomass and vegetation structure at struction. Land parcel selection for this study was based on feedback
multiple spatial and temporal scales (Neuenschwander and Magruder, provided by members of the KFN Forest Resources Management Com­
2016; Popescu et al., 2018; Narine et al., 2019; Duncanson et al., 2020; mittee in collaboration with community members who emphasized that
Nandy et al., 2021; Enterkine et al., 2022). parcels with existing road access, site productivity and salvage potential
Historical forest vegetation inventory data on canopy height in the be evaluated for estimation of canopy height. Our specific objective of
Yukon was acquired from aerial imagery taken in the early 1980s. The this study includes providing KFN with updated, comprehensive and
archival data on tree height did not capture major disturbances that quantifiable information at a local scale (tens of kilometers) on canopy
occurred in the region (e.g., wildfires and a spruce beetle infestation), height for nine land parcels within a dynamic setting, leading to
however, it was useful for a number of purposes including providing enhanced planning and management of forestry resources.
information on the location of productive sites in the boreal forest, and
in identifying the amalgamation of species distribution. In this research, Study area
we employed the use of a novel cyberinfrastructure platform to view and
quantify along-track geolocated photon event data from NASA’s ICESat- KFN’s administrative centre is based in Burwash Landing, which is

Fig. 2. Location Map showing the nine Kluane First Nation Category A and B Settlement land parcels that were analyzed.

3
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

located approximately 230 km to the west of Whitehorse (Yukon’s allocate the forest resources under its purview. KFN also owns the rights
capital city), and is the nestled between the northwest shores of Lhù’ààn to a further 259 km2 of Category B Settlement surface lands. Among the
Mânʼ (Kluane Lake) and the adjacent Alaska Highway (Fig. 2). The Category A Settlement land parcels are R-1A, R-3A, R-13A, R-45A, and
traditional territory of the Lù’àn Män Ku Dän (Kluane Lake People) R-47 A (Fig. 2, Table 1). Parcels R-1A and R-47A are located to the west
stretches from Lhù’ààn Mânʼ to the Ruby and Nisling mountain ranges of Burwash Landing and include the Burwash Uplands, Copper Joe
(elevation ranging from 800 to 2100 m) in the northeast and the St. Elias Creek and the Duke River. R-1A is the largest surface land plot and has
mountains (several peaks above 4200 m) to the southwest (Scudder, an area of 440.98 km2 while R-47A encompasses approximately 29 km2.
1997). Topographic variations of these mountain ranges on either side of R-45 A is nestled to the east of R-47 A along the westerly shore of
Lhù’ààn Mânʼ coupled with differences in the quaternary geology of the Lhù’ààn Mânʼ. The closest extent of Parcel R-45 A (7.08 km2) is situated
region characterize the area (Smith et al., 2004; Dearborn and Dandy, approximately 2 km from Burwash Landing along the southwestern
2018). Lhù’ààn Mânʼ is encircled by the higher peaks of the St. Elias trending shoreline of Lhù’ààn Mânʼ. R-3A includes Thorsen Bay and
Mountains on its western end and the Ruby Range to the east (Clague fronts the southeastern shore of Lhù’ààn Mânʼ. R-3A comprises a land
et al., 2006). The surficial geology of the region is characterized by the area of 58.19 km2 is characterized by areas of rolling terrain accentuated
presence of coarse glaciofluvial deposits of Pleistocene origin (Muller, by the presence of fairly flat benches and terraces. R-13A is one of the
1967; Laxton et al., 1996; Sanborn and Jull, 2010), and the land surface smallest land segments (2.11 km2) and is located downstream of the
is dominated by the presence of poor to moderately well-drained eutric Talbot Arm, north of Gladstone Creek along the northeastern shore of
brunisols and cryosols (Brabets et al., 2000; Smith et al., 2011). While Lhù’ààn Mânʼ in close vicinity to Jacquot Island. Land parcels R-2B (8.76
the area is classified as a zone of discontinuous permafrost km2) and R-29B (5.01 km2) can be found along the southwestern
(Heginbottom, 1995; Hugenholtz and Lewkowicz, 2002; Krebs et al., shoreline of Lhù’ààn Mânʼ obliquely opposite to parcel R-3A. R-2B is
2014), several zones in the surrounding valleys of Lhù’ààn Mânʼ are located on the northern side of the Alaska Highway between Dutch
underlain by incessant permafrost (Harris, 1987). In close vicinity to Harbour and Mines Creek while R 29-B is bordered by Congdon Creek
Burwash Landing, Ellwood and Nixon (1983) recorded a permafrost and Dutch Harbour, and is a region with nearly level to gentle topo­
thickness of 18 m beneath an active layer of approximately 1.4 m graphic relief. R-18B is situated on the eastern shore of the lake,
(YEWG, 2004). A semi-arid climate characterizes the region (Laxton approximately 6 km away from Destruction Bay on Lhù’ààn Mânʼ’s
et al., 1996) with cold winters and generally warm summers of short western extremity. R-18 B has a land area of 7.26 km2. Sandspit and
duration. Annual mean temperatures in the area average − 18 ◦ C in Doghead Points characterize lake adjoining segments of parcel R-5B
winter months (December to February) and 11 ◦ C during the summer which comprises a land area of 26.85 km2, and is moderately inclined.
period (June to August) (Environment Canada, 2022). Meteorologic
data from a weather station located approximately 770 m above sea Methods
level, located at the Kluane Lake Research Station which is operated by
the Arctic Institute of North America, University of Calgary returned an Canopy height (hcanopy) estimation from ICESat-2 reflected photons
average mean annual air temperature of − 1.4 ◦ C over the period 2012 to
2020, indicating warming has taken place in the past decade when The Advanced Topographic Laser Altimeter System is a photon
compared to an average yearly temperature of − 2.7 ◦ C noted at the same calculating laser divided into six beams—three weak and three strong
station over the period 1972–1980 (Hugenholtz, and Lewkowicz, 2002). pairs (Fig. 3). Beam pairs are spaced 3.3 km apart, with approximately
As of the last census (Statistics Canada, 2021), the current population 90 m between strong and weak signal beams (Neuenschwander and
of Burwash Landing is 64 persons but approximately 90 persons Pitts, 2019). The strong beams provide increased frequency of reciprocal
currently live in the area. Many citizens are employed by the KFN laser photons per laser pulse, amplified reliability and deliver data on
Government while some people are engaged in tourism, mining, logging, objects of lesser reflectance (e.g., sea-ice, vegetation) (Neuenschwander
and subsistence activities (e.g., big game hunting) for their livelihood and Magruder, 2016; Markus et al., 2017; Arendt et al., 2019; Sun et al.,
(Peetom, 2018). Lhù’ààn Mânʼ and the surrounding area holds signifi­ 2020). ICESat-2 time and position data are acquired from singular laser
cant ecological, cultural, social, and economic importance to the com­
munities within whose traditional lands the lake is located, and to the
territory more broadly. Lhù’ààn Mânʼ (61◦ 16′45.64″N, 138◦ 45′03.83″ W) Table 1
is the largest body of water in Yukon with an estimated surface area of Land area and ICESat-2 coverage date for each land parcel analyzed. A reference
400 to 410 square kilometers (km2) and maximum depths ranging be­ co-ordinate position for each land parcel is also displayed.
tween 80 and 90 m (Clague et al., 2006; Crookshanks and Gilbert, 2008; KFN Land ICESat- ICESat- ICESat-2 Latitude Longitude
Barker et al., 2014). The lake benefits from the seasonal release of land area 2 2 Date 3
meltwater from the Kaskawulsh Glacier and from a combination of other parcel (km2) Date 1 Date 2 dd/mm/
dd/ dd/ yy
natural water supplies including Silver, Christmas, Cultus, Bock, Talbot,
mm/yy mm/yy
and Gladstone Creeks (Sawada and Johnson, 2000; Brahney et al.,
2010). In 2016, climate-induced retreat of the Kaskawulsh Glacier R-1A 440.98 8-Mar- ​ ​ 61.30 − 138.88
19
resulted in a directional change of meltwater away from a northerly flow R-2B 8.76 2-Aug- ​ ​ 61.19 − 138.67
towards Lhù’ààn Mânʼ to an eastward movement to the Kaskawulsh 19
River (Shugar et al., 2017). Surrounding Lhù’ààn Mânʼ, a combination of R-3A 58.19 31-Jan- ​ ​ 61.19 − 138.48
coniferous and deciduous trees complements this Boreal Cordillera 20
R-5B 26.85 8-Mar- 02-Mar- 61.45 − 138.82
ecozone (Sanborn and Jull, 2010) with white spruce (Picea glauca),
​
19 22
balsam poplar (Populus balsamifera) and trembling aspen (Populus R-13A 2.11 3-Jan- ​ ​ 61.37 − 138.71
tremuloides) representing most of the tree species in the area (Turkington 20
et al., 2002; Krebs et al., 2009; Campbell et al., 2019). Other tree species, R-18B 7.26 3-Jan- 30-Apr- ​ 61.26 − 138.71
such as birch, willow and alder, are also abundant in this locale. As part 20 20
R-29B 5.01 4-Oct- 29- 61.14 − 138.56
of the Land Claim and Self Government Agreement (Kluane First Nation
​
19 June-22
Final Agreement, 2003; Wilson, 2019) that became effective in early R-45A 7.08 3-May- 01-Aug- ​ 61.33 − 138.96
2004, KFN retained complete possession of 647 km2 of subterranean and 20 22
surface land also referred to as Category A Settlement Lands. The Self R-47A 29 3-Oct- 6-May- 01-Feb- 61.37 − 139.09
19 20 22
Government Agreement gave KFN the ability to distribute, oversee and

4
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

Fig. 3. Diagram shows the 3 pairs of strong and weak beams from ICESat-2 ground track acquired over the R5B and R1A land parcel segments. The distance between
each pair of tracks is approximately 3.3 km apart in the cross-track direction.

pulses obtained along itemized reference ground segments (Khalsa et al., ATL03 geolocated photons product. ICESat-2 ATL03 and ATL08 data for
2022). The land and vegetation height data set, ATL08, associated with the land parcels under investigation (parcels R-1A, R-3A, R-13A, R-45A,
ICESat-2 data contains along-track measurements of elevation above the R-47A, R-2B, R-5B, R-18B, and R-29B were acquired over the period
World Geodetic System ellipsoid and provides topographic and canopy March 08, 2019 to August 01, 2022, using only strong power beams
surface information at fine spatial scales (Neuenschwander and Pitts, (Table 1). Neuenschwander et al. (2020) suggested a tendency for can­
2019). ATL08 data for canopy height estimation are reported as an opy height underestimation and increased error when using data ac­
aggregate of collected photons over a 100 m (along-track profile) × 12 m quired from weak beams. As a result, ICESat-2 derived tree height
(across-track profile) segment (Markus et al., 2017; Khalsa et al., 2022; estimates were based using data obtained only from strong beams (i.e.,
Li et al., 2020). In addition to acquisition time, each photon downlinked beams gt1r, gt2r and gt3r) (Fig. 3). To facilitate data accuracy, an effort
from ICESat-2 as part of the ATL03 data product returns information on was made to obtain beam coverage during the early spring to late
geographic coordinates as well as absolute elevation (Dandabathula summer season (March to August) as previous research in boreal forest
et al., 2021; Shufan and Chun, 2022). According to Neuenschwander regions suggests the amplification of errors during wintry conditions
and Pitts (2019), canopy heights for individual photons are calculated (Neuenschwander et al., 2020; Liu et al., 2021; Zhu et al., 2023). For this
by eliminating the ground surface height for the respective photon’s study, the h_canopy variable, representative of the 98 % height of canopy
coordinate information. Several factors however, (e.g., geolocation un­ elevation relative to ground data (Fernandez-Diaz et al., 2022; Sothe
certainties and instrument precision) can influence the accuracy of et al., 2022) was employed. Visualization and access to canopy height
elevation measurements derived from ATLAS products and surface elevation data were obtained using an altimetry-specific
(Neuenschwander et al., 2020; Neumann et al., 2020), and as a result, a graphical user interface developed by scientists at the National Snow
range of geophysical corrections (e.g., atmospheric effects and tidal and Ice Data Center, the San Diego Supercomputer Center at the Uni­
corrections) must be applied to the ATL03 data product to provide ac­ versity of California San Diego, and the Scripps Institution of Ocean­
curate elevation-based information (Wang et al., 2019; Dandabathula ography (Khalsa et al., 2019). The discovery platform employs imagery
et al., 2021). from the Environmental Systems Research Institute and Moderate Res­
For this analysis, ATL08 data were utilized in combination with olution Imaging Spectroradiometer to establish base layers for any

5
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

location on the planet in varying projections and enables access, where d is the horizontal distance between observer and tree, θ is the
retrieval, analysis, and display of elevation profile data (Khalsa et al., angle corresponding to the base to the top of the tree and Y is the dis­
2022). tance from the observer’s eye to the ground.
To facilitate an evaluation of the difference between observed field
Canopy height estimation from ground reference data heights (Hfield) and estimated canopy heights (HICESat-2), standard sta­
tistical assessment measures such as root mean square (RMSE) and R-
Field assessments of individual tree heights along parcels R-1A, R- squared were employed. Canopy tree density (stems per hectare (sph))
3A, R-47A, R-5B and R-29B (Table 3) were conducted over a two-month across the parcels identified for ground-based data collection ranged
period (June to August) in 2019 using a standard clinometer. Of the five from 2500 to 5000 sph (R-47A), approximately 6000 sph for R-3A, and
land parcels mentioned above, 46 ground-based measurements of white an average of 5900 sph for parcels R 2-B and R-29B (Campbell et al.,
spruce tree height, the dominant vegetation of the study area, were 2019).
taken at R-47 A while eight readings were conducted at R-3A, and five
measurements were obtained at R-1A. Two terrestrial observations of Results
tree heights were taken at land parcels R-5B and R-29B. The emphasis on
an increased sampling density of clinometer-based tree height data This study is unique compared to previous large-scale assessments of
collection at R-47 A as opposed to the other land parcels was due to the canopy height across Canada’s boreal forests which were conducted at
proposed strategic value and economic potential of the trees in this broader geographic scales. Here, we attempted to quantify forest canopy
parcel for future sawlog development. The next highest count of height on a finer scale of tens of meters to support focused community-
clinometer-derived assessments of tree height were obtained along the led forest management initiatives. Average canopy heights—calculated
southern portion of land parcel R-3A in response to the known timber by differencing the canopy photon height from the estimated terrain
potential of this area coupled with an expressed desire for KFN members surface—and mean surface elevation for all land parcels are summarized
to possibly construct recreational and trapline cabins in the near future. in Figs. 4 and 9/Table 2. The returned estimates of canopy height
The clinometer method calculates tree height through an evaluation of (h_canopy) from ICESat-2 photon data represents the mean value of
the measured angle (θ) that coincides with the change in angle from the elevation measurements within a 13 m diameter footprint. In the event
base to apex of a tree, when the instrument is pointed at the tree being where there were multiple trees within the 13 m satellite coverage area,
measured (Saliu et al., 2020). The height differential between the ob­ the enumerated canopy height is the average of all the trees that the
server’s eye above ground level and the horizontal distance between photon hits.
observer and tree are also recorded using a standard tape measure to Forest canopy height obtained from ICESat-2 ATL08 product across
satisfy the parameters required for trigonometrical correlation of tree all parcels ranged from 6.59 to 16.52 m with an average value of 11 m,
height according to the equation: measured at elevations ranging from 802.96 to 1180.75 m. Comparison
of ICESat-2 tree height with ground-based measurements yielded an R2
Absolute Tree Height = (d tanθ) + Y of 0.53 and root mean square error (RMSE) of 1.45 m (Fig. 5, Table 3).

Fig. 4. Graphs of average surface elevation (ellipsoidal heights) and average canopy heights for Category A and B Kluane First Nation Lands.

6
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

Table 2 heights) and canopy heights relationship for the parcels returned an R2
Summary of average canopy heights and ellipsoidal surface elevations for all value of 0.281 suggesting an asymmetrical relationship between these
KFN land parcels analyzed. The date of ICESat-2 strong beam coverage over each two variables (Fig. 6).
parcel and a reference co-ordinate position for each land parcel is also displayed.
KFN land Average canopy Average ellipsoidal Date of coverage Kluane First Nation settlement lands
parcel heights (m) surface elevation (m) (yyyy/mm/dd)

R-1A 10.20 854.5 2019-03-08 The highest average tree heights of all Category A Settlement Lands
R-2B 10.70 830.78 2019-08-02 were obtained in parcel R-13A (Figs. 4, 6, 7; Table 2). Canopy heights in
R-3A 10.57 824.16 2020-01-31 this locale returned an average value of 14.39 m while the mean surface
R-5B 7.04 1139.87 2022-03-02
R-5B 6.59 1180.75 2019-03-08
ellipsoidal elevation was on the order of 814.89 m. The increased
R-13A 14.39 814.89 2020-01-03 availability of orbital coverage for parcel R-47A facilitated an analysis of
R-18B 9.16 819.54 2020-04-30 canopy height for this land tract over a three-year period (2019–2022).
R-18B 10.77 802.96 2020-01-03 Canopy heights here yielded similar values over the time periods
R-29B 15.37 810.58 2022-06-29
examined with average elevations of 12.32 m recorded on October 03,
R-29B 16.52 820.13 2019-10-04
R-45A 9.78 822.42 2020-05-03 2019, 12.06 m (May 06, 2020), and a lower mean canopy height of
R-45A 8.73 814.62 2022-08-01 10.79 m, from the most recent beam coverage (February 01, 2022). A
R-47A 10.79 850.31 2022-02-01 similar small decrease in average canopy height was noted over land
R-47A 12.06 823.74 2020-05-06 parcel R-45A (9.78 m in 2020 to 8.73 m in 2022) for which repeat
R-47A 12.32 827.5 2019-10-03
photon coverage over a two-year period within an identical ICESat-2
strong beam geographical footprint was available.
Our results are in very good agreement with the analysis conducted by The highest average canopy height for all Category A and B Settle­
Nandy et al. (2021) that returned an RMSE value of 1.1 m, representing ment lands was located along parcel R-29B and measured 16.52 m. A
a measurement differential for ICESat-2 canopy heights and field mea­ closer examination of Parcel R-29B data (Fig. 8) also revealed one of the
surements. Our findings of average boreal forest canopy height are also
consistent with the results of several other studies (Lefsky, 2010; Simard Table 3
et al., 2011; Matasci et al., 2018; Sothe et al., 2022; Lang et al., 2023). Average heights from field measurements and ICESat-2 profiles along with
Matasci et al. (2018) reported lidar-derived values of canopy elevation height differences.
in the Canadian boreal cordillera within a range of approximately 9 to Land Average canopy height Average canopy Height difference
17 m over the period 1984 to 2016 which closely aligns to our findings. parcel from field measurements height from Ice-Sat- (HICESat-2 -Hfield)
Results from this study are consistent with recent large-scale mapping no. (Hfield) (m) 2 (HICESat-2) (m) (m)

estimates of predicted canopy height (~8 to 19 m) for the Canadian R1A 12.1 10.2 − 1.9
boreal cordillera ecozone using ICESat-2 data (Sothe et al., 2022). R3A 12 10.6 − 1.4
R47A 11.8 11.7 − 0.1
Findings from this assessment also reveal significant geographic varia­
R2B 9.7 10.7 1
tion in the patterns of forest canopy heights along the nine R29B 13.9 15.9 2
pre-determined land parcel segments. The surface elevation (ellipsoid

Fig. 5. Scatterplot between estimated canopy height derived from ICESat-2 data and field-measured values.

7
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

Fig. 6. Scatterplot of surface elevation and average canopy heights correlation for the KFN parcels analyzed. The black line represents the weighted smoothing
average bounded by the 95 % confidence interval.

Fig. 7. Distribution of canopy height values versus latitude for all Category A land parcel segments.

highest measured canopies (24.43 m) within a 14 m-diameter located at noted for land segments R-18B and R-29B where repeat ICESat-2 beam
a latitude of 61.1445 and a longitude of − 138.5604. R-29B returned the coverage data was available.
two highest canopy height averages (16.52 m in 2019 and 15.37 m in
2022) amongst all Category A and B land segments. R-18B was analyzed Discussion
twice owing to repeat ICESat-2 beam coverage over the location on
January 3, 2020, and April 30, 2020. Mean values of canopy heights We did not attempt to evaluate the predictor variables of forest
over this parcel measured 10.77 m for the ground track coverage on canopy height but instead aimed to showcase the practical applicability
January 3, 2020, and 9.16 m just a few months later. The observed trend of ICESat-2 data and the use of the Open Altimetry platform to quantify
of decreases in canopy height over time across both elevation and lati­ and analyze tree height at a finer resolution in a subarctic boreal forest
tude for some Category A land parcels (R-47A, R-45A) was similarly locale. Future studies that map forest vertical structure in this region

8
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

Fig. 8. Distribution of canopy height values versus latitude for all Category B land parcel segments.

should integrate Earth observation data from supplementary sources (e. obtained from multiple proxies could provide further insight into the
g., Geoscience Laser Altimeter System (GLAS), MODIS reflectance data, role of climate variability, anthropogenic disturbances, and other
Unmanned Aerial Systems, and Landsat) to enumerate canopy height biogeographic variables (e.g., evapotranspiration, precipitation) in
and test hypotheses related to the determinants of tree height and shaping tree height in this area (Moles et al., 2009; Klein et al., 2015;
expand on these results obtained from ICESat-2 data. Importantly, in­ Zhu et al., 2015), user error and issues related to data consistency must
formation from community-led field survey campaigns will provide also be taken into account when comparing datasets (Molinier et al.,
complementary data for validating spaceborne estimates of tree height 2016; Roman et al., 2017; Enterkine et al., 2022). According to Bragg
obtained from ICESat-2 and other sources of remotely-sensed data. (2014), variability in terrain steepness modifies the angles at which
It is important to note that ICESat-2 data is limited by its inability to ground-based measurements of tree height are acquired which can
distinguish between varying tree species, and records measurements of significantly compromise the reliability of the data. While many of the
vertical structure irregardless of the vegetation type (Mulverhill et el., land parcels evaluated in this study are characterized by rolling hills and
2022). Field based data collection of canopy height in this study was gently sloping terrain (Clague, 1981; James, 2017), it is imperative for
primarily aimed at acquiring the elevation of commercially viable white field observers to document and account for uneven terrain during
spruce. As a result, significant variability between observed measure­ future excursions in order to minimize uncertainty in tree height
ments and ICESat-2 returned values of canopy height at parcels R-1A, measurements.
R-47 A, R-3A, R-5B and R-29B were resolved in favor of the observed Incorporating predictor variables and climatic indices to elucidate
measurements (Kluane First Nation Forest Management Plan, 2022). and forecast the spatiotemporal patterns of tree heights should involve
Our values of estimated canopy heights obtained using ICESat-2 data robust statistical analyses, interpretative models and can in collabora­
agree with field measurements up to an accuracy of 1.9 m (Table 3). As tion with Traditional Knowledge holders provide additional insight into
much as possible, measures were taken to reduce errors related to the myriad of factors that influence canopy height (Lindeman et al.,
overlap probability (i.e., footprint locations of ICESat-2 compared to the 1980; Lu and Zhang, 2011; Lisein et al., 2013; Zhang et al., 2016). For
precise location of ground survey subplots), and the data employed in example, our analysis revealed that parcels R-47A, R-45A, R-18B and
this study reflect the closest satellite coverage for the land parcels R-29B experienced small reductions (~10 % decrease on average across
analyzed. Future field-based data collection efforts on KFN Settlement these four land segments) in canopy height over the period under
Lands may benefit from the strategic coordination with accessible consideration. (Table 2, Figs. 4 and 6). While these plots are in close
ICESat-2 reference ground tracks from this study to confirm the accuracy vicinity to Lhù’ààn Mânʼ, they also appear to benefit from a consistent
between these two data sets. We acknowledge this may be challenging supply of warming-induced glacial meltwater from mountainous
for forestry personnel, community members and citizen scientists given streams adjacent to Lhù’ààn Mânʼ (Chesnakova et al., 2020; Cruikshank,
the topography of the region coupled with difficult road access (Fig. 9), 2020; Main et al., 2023). Several studies (Kozlowski, 1984; Lambers
and the inability to obtain an unobstructed view of individual trees in an et al., 2008; Miyamoto et al., 2010; Tao et al., 2016) have found a trend
area known for dense stands and species mixing. of decreasing canopy height with increasing water availability as
While the plurality and availability of canopy height measurements nutrient leaching in soil is amplified when an abundance of water

9
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

Fig. 9. Spatial distribution of forest canopy heights for Kluane First Nation owned settlement lands derived from ICESat-2 data. Roadway access (paved and un­
paved) to adjacent land parcels is shown.

exceeds biological demand. As also reported by Blanke and Cooke values of canopy height over time along several parcels, results also
(2004), stomatal closure in response to excess water availability nega­ indicate that trees growing in the valley region of Lhù’ààn Mânʼ (e.g.,
tively affects photosynthesis which coupled with other physiological R-13 A) returned increased values of canopy height, possibly due to the
processes can result in forest canopy height decreases. Similar results increased accumulation and availability of nutrients. Future studies in
that disclose a negative relationship between recent increases in atmo­ this region should evaluate the site-specific meltwater contributions
spheric temperatures and tree growth are noted in numerous other from the network of surrounding glaciers to adjacent mountainous water
studies. Reid et al. (2022) found that white spruce stands in the Yukon reservoirs in close vicinity to the parcels analyzed, and by extension
displayed consistent growth declines independent of location and water-use efficiency dynamics (Hawkins et al., 2003; Moles et al., 2009;
elevation while Hogg and Wein (2005) as well as Miyamoto et al. (2010) Klein et al., 2015; Brienen et al., 2017) in affecting forest canopy height.
and Griesbauer and Green (2012) detailed the combined influences of Detailed analysis of individual parcels also reveals an observed
increasing temperatures, drought conditions and anthropogenic distur­ variation in canopy heights near segments of increased elevation. Ac­
bances in this region as catalysts to decreasing canopy height and cording to Jucker et al. (2018), topographic gradients are a critical
reduced tree growth. While we observed a small decrease in the mean driver of forest structure in mountainous terrain. Other authors also

10
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

concluded that tree heights are influenced by altitude and slope position References
(Hirata, 2004; Chadwick and Asner, 2016; Torita and Masaka, 2020;
Rahman et al., 2022). Our findings confirm this relationship as land Ahmad, F., Goparaju, L., Qayum, A., 2017. Natural resource mapping using landsat and
lidar towards identifying digital elevation, digital surface and canopy height models.
parcel R-5B, the highest average surface elevation of all plots analyzed at Int. J. Environ. Sci. Nat. Res. 2, 555580.
1180.75 m exhibited the lowest average canopy height (6.59 m) of all Andersen, H.E., McGaughey, R.J., Reutebuch, S.E., 2005. Estimating forest canopy fuel
parcels examined (Figs. 4, 6; Table 2). While these correlations could be parameters using LIDAR data. Remote Sens. Environ. 94 (4), 441–449.
Arendt, A., Smith, B., Shean, D., Steiker, A., Petty, A., Perez, F., Henderson, S., Paolo, F.,
attributed to the varying topography, future work should also expand on Nilsson, J., Becker, M., and Adusumilli, S. (2019). ICESAT-2HackWeek/ICESat2_
the role that thawing permafrost and the depth to permafrost layer plays hackweek_tutorials (Version 1.0), Zenodo.
in forest development through nutrient availability (Chang et al., 2015; Barker, O., Millar, N., Foos, A., 2014. Lake Trout and Lake Whitefish Population
Assessment Kluane Lake 2013. Yukon Fish and Wildlife Branch.
Tchebakova et al., 2016). Blanke, M.M., Cooke, D.T., 2004. Effects of flooding and drought on stomatal activity,
transpiration, photosynthesis, water potential and water channel activity in
Conclusion strawberry stolons and leaves. Plant Growth Regul. 42, 153–160.
Bolton, D.K., Coops, N.C., Wulder, M.A., 2013. Measuring forest structure along
productivity gradients in the Canadian boreal with small-footprint Lidar. Environ.
The ability to map, quantify and monitor canopy height using Monit. Assess. 185, 6617–6634.
remotely sensed data coupled with in-situ approaches are critical to Bonan, G.B., 2008. Forests and climate change: forcings, feedbacks, and the climate
understanding and measuring forest ecosystem health and productivity, benefits of forests. Science 320 (5882), 1444–1449.
Boutsoukis, C., Manakos, I., Heurich, M., Delopoulos, A., 2019. Canopy height estimation
given the effects of near-term climate variability. This investigation from single multispectral 2D airborne imagery using texture analysis and machine
complements ongoing research to quantify tree height by providing learning in structurally rich temperate forests. Remote Sens. 11 (23), 2853.
updated information on canopy elevation for a subarctic locale at Box, J.E., Colgan, W.T., Christensen, T.R., Schmidt, N.M., Lund, M., Parmentier, F.J.W.,
Brown, R., Bhatt, U.S., Euskirchen, E.S., Romanovsky, V.E., Walsh, J.E., Overland, J.
increased spatial and temporal detail. Here, we utilized an innovative E., Wang, M., Corell, R.W., Meier, W.N., Wouters, B., Mernild, S., Mard, J.,
web-based tool to facilitate an analysis of forest canopy heights for Pawlak, J., Olsen, M.S., 2019. Key indicators of Arctic climate change: 1971–2017.
several sites of the boreal forest under the stewardship of the Kluane Environ. Res. Lett. 14 (4), 045010.
Brabets, T.P., Wang, B., and Meade, R.H. (2000). Environmental and hydrologic
First Nation, and to compare the data with existing terrestrial informa­ overview of the Yukon River Basin, Alaska and Canada. USGS Water-Resources
tion. This study underscores the significance and potential worth of Investigations Report 99-4204, Anchorage, Alaska.
employing the use of this digital platform as an avenue for using ICESat- Bragg, D.C., 2014. Accurately measuring the height of (real) forest trees. J. For. 112 (1),
51–54.
2 data to address research questions based on community needs related Brahney, J., Clague, J.J., Edwards, T.W., Menounos, B., 2010. Late Holocene
to terrain and canopy height changes. Findings from our study provide paleohydrology of Kluane Lake, Yukon Territory, Canada. J. Paleolimnol. 44 (3),
critical insights into canopy height in the region, and have been incor­ 873–885.
Bratkovich, S., 1994. Flooding and its Effect on Trees: Information Packet. United States
porated into the most recent KFN Forest Resources Management Plan
Department of Agriculture, Forest Service, State and Private Forestry, Northeastern
(Kluane First Nation Forest Management Plan, 2022). Despite the Area.
challenges posed by in-situ data collection, establishing a rich set of both Brienen, R.J.W., Gloor, E., Clerici, S., Newton, R., Arppe, L., Boom, A., Bottrell, S.,
ground-based and space-based satellite data will facilitate local-scale Callaghan, M., Heaton, T., Helama, S., Helle, G., Leng, M.J., Mielikainen, K.,
Oinonen, M, Timonen, M., 2017. Tree height strongly affects estimates of water-use
continuous mapping of canopy height as well as future studies that efficiency responses to climate and CO2 using isotopes. Nat. Commun. 8 (1), 288.
aim to investigate the relationship between predictor variables and the Campbell, B.A., 2021. ICESat-2 and the trees around the GLOBE student research
changes in forest ecosystem in response to nutrient availability, campaign: looking at Earth’s tree height, one tree at a time. Acta Astronaut. 182,
203–207.
permafrost thaw, increased length of the growing period, drought, log­ Campbell, E.M., Antos, J.A., van Akker, L., 2019. Resilience of southern Yukon boreal
ging, windthrow, insect disturbances and the prevalence of wildfires. forests to spruce beetle outbreaks. For. Ecol. Manag. 433, 52–63.
Understanding the potential geographic variability in boreal forest Chadwick, K.D., Asner, G.P., 2016. Tropical soil nutrient distributions determined by
biotic and hillslope processes. Biogeochemistry 127 (2–3), 273–289.
canopy height and being able to easily access, quantify and visualize Chang, X., Jin, H., Zhang, Y., He, R., Luo, D., Wang, Y., Lü, L., Zhang, Q., 2015. Thermal
these data is essential to future land use management, hazard and risk impacts of boreal forest vegetation on active layer and permafrost soils in northern
mapping and to supporting community-led climate mitigation strategies Da Xing’anling (Hinggan) Mountains, Northeast China. Arct. Antarct. Alp. Res. 47
(2), 267–279.
that accurately address the dynamics of probable alterations in forest Chesnokova, A., Baraër, M., Laperrière-Robillard, T., Huh, K., 2020. Linking mountain
ecosystems at the community and regional scales. glacier retreat and hydrological changes in southwestern Yukon. Water. Resour. Res.
56 (1), e2019WR025706.
Clague, J.J., 1981. Landslides at the south end of Kluane Lake, Yukon Territory. Can. J.
CRediT authorship contribution statement
Earth Sci. 18 (5), 959–971.
Clague, J.J., Luckman, B.H., Van Dorp, R.D., Gilbert, R., Froese, D., Jensen, B.J.,
Ravi Darwin Sankar: Writing – review & editing, Writing – original Reyes, A.V., 2006. Rapid changes in the level of Kluane Lake in Yukon Territory over
the last millennium. Quat. Res. 66 (2), 342–355.
draft, Validation, Resources, Methodology, Investigation, Formal anal­
Cook, B.I., Seager, R., Williams, A.P., Puma, M.J., McDermid, S., Kelley, M.,
ysis, Data curation. Finella Pescott: Writing – review & editing, Vali­ Nazarenko, L., 2019. Climate change amplification of natural drought variability:
dation, Resources, Methodology. Kate Hicks: Writing – review & the historic mid-twentieth-century North American drought in a warmer world.
editing, Validation, Resources, Methodology. Maribeth S. Murray: J. Clim. 32 (17), 5417–5436.
Crookshanks, S., Gilbert, R., 2008. Continuous, diurnally fluctuating turbidity currents in
Writing – review & editing, Supervision, Resources, Project adminis­ Kluane Lake, Yukon Territory. Can. J. Earth Sci. 45 (10), 1123–1138.
tration, Investigation. Cruikshank, J., 2020. Melting glaciers and emerging histories in the Saint Elias
Mountains. Indigenous Experience Today. Routledge, pp. 355–378.
Dandabathula, G., Sitiraju, S.R., Jha, C.S., 2021. Retrieval of building heights from
Declaration of competing interest ICESat-2 photon data and evaluation with field measurements. Environ. Res. 1 (1),
011003.
The authors declare the following financial interests/personal re­ Dearborn, K.D., Danby, R.K., 2018. Climatic drivers of tree growth at tree line in
Southwest Yukon change over time and vary between landscapes. Clim. Change 150
lationships which may be considered as potential competing interests: (3–4), 211–225.
Ravi Sankar reports statistical analysis was provided by Longwood Duncanson, L., Neuenschwander, A., Hancock, S., Thomas, N., Fatoyinbo, T., Simard, M.,
University. If there are other authors, they declare that they have no Silva, C.A., Armston, J., Luthcke, S.B., Hofton, M., Kellner, J.R., Dubayah, R., 2020.
Biomass estimation from simulated GEDI, ICESat-2 and NISAR across environmental
known competing financial interests or personal relationships that could gradients in Sonoma County, California. Remote Sens. Environ. 242, 111779.
have appeared to influence the work reported in this paper. Ellwood J.R. and Nixon, J.F., 1983. Observations of soil and ground ice in pipeline trench
excavations in the south Yukon. In: Proceedings, 4th International Conference on
Permafrost, Fairbanks, Alaska. July 1983, National Academy Press, Washington, D.
Data availability
C., 1:278-282.
Enterkine, J., Campbell, B.A., Kohl, H., Glenn, N.F., Weaver, K., Overoye, D., Danke, D.,
Data will be made available on request. 2022. The potential of citizen science data to complement satellite and airborne lidar

11
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

tree height measurements: lessons from The GLOBE Program. Environ. Res. Lett. 17 Laxton, N.F., Burn, C.R., Smith, C.A.S., 1996. Productivity of loessal grasslands in the
(7), 075003. Kluane Lake region, Yukon Territory, and the Beringian ``production paradox’’.
Environment Canada. (2022). Hourly data report: weather station: Burwash A, Yukon. Arctic 9 (2), 129–140.
Retrieved from https://climate.weather.gc.ca (Accessed December 06, 2023). Lefsky, M.A., 2010. A global forest canopy height map from the moderate resolution
Fagua, J.C., Jantz, P., Rodriguez-Buritica, S., Duncanson, L., Goetz, S.J., 2019. imaging spectroradiometer and the geoscience laser altimeter system. Geophys. Res.
Integrating LiDAR, multispectral and SAR data to estimate and map canopy height in Lett. 37 (15).
tropical forests. Remote Sens. 11 (22), 2697. Lefsky, M.A., Hudak, A.T., Cohen, W.B., Acker, S.A., 2005. Patterns of covariance
Fernandez-Diaz, J.C., Velikova, M., Glennie, C.L., 2022. Validation of ICESat-2 ATL08 between forest stand and canopy structure in the Pacific Northwest. Remote Sens.
terrain and canopy height retrievals in tropical Mesoamerican forests. IEEE J. Sel. Environ. 95 (4), 517–531.
Top. Appl. Earth Obs. Remote Sens. 15, 2956–2970. Li, W., Niu, Z., Shang, R., Qin, Y., Wang, L., Chen, H., 2020. High-resolution mapping of
García, M., Saatchi, S., Ustin, S., Balzter, H., 2018. Modelling forest canopy height by forest canopy height using machine learning by coupling ICESat-2 LiDAR with
integrating airborne LiDAR samples with satellite Radar and multispectral imagery. Sentinel-1, Sentinel-2 and Landsat-8 data. Int. J. Appl. Earth Obs. Geoinf. 92,
Int. J. Appl. Earth Obs. Geoinf. 66, 159–173. 102163.
Givnish, T.J., Wong, S.C., Stuart-Williams, H., Holloway-Phillips, M., Farquhar, G.D., Lin, X., Xu, M., Cao, C., Dang, Y., Bashir, B., Xie, B., Huang, Z., 2020. Estimates of forest
2014. Determinants of maximum tree height in Eucalyptus species along a rainfall canopy height using a combination of ICESat-2/ATLAS data and stereo-
gradient in Victoria, Australia. Ecology 95 (11), 2991–3007. photogrammetry. Remote Sens. 12 (21), 3649.
Griesbauer, H.P., Green, D.S., 2012. Geographic and temporal patterns in white spruce Lindeman, R.H., Merenda, P.F., Gold, R.Z., 1980. Introduction to Bivariate and
climate-growth relationships in Yukon, Canada. For. Ecol. Manag. 267, 215–227. Multivariate Analysis. Scott: Foresman and Company, Glenview, IL, p. 119.
https://doi.org/10.1016/j.foreco.2011.12.004. Lisein, J., Pierrot-Deseilligny, M., Bonnet, S., Lejeune, P., 2013. A photogrammetric
Harris, S.A., 1987. Altitude trends in permafrost active layer thickness, Kluane Lake, YT. workflow for the creation of a forest canopy height model from small unmanned
Arctic 40 (3), 179–183. aerial system imagery. Forests 4 (4), 922–944.
Hawkins, B.A., Field, R., Cornell, H.V., Currie, D.J., Guégan, J.F., Kaufman, D.M., Kerr, J. Liu, A., Cheng, X., Chen, Z., 2021. Performance evaluation of GEDI and ICESat-2 laser
T., Mittelbach, G.G., Oberdorff, T., O’Brien, E.M., Porter, E.E., 2003. Energy, water, altimeter data for terrain and canopy height retrievals. Remote Sens. Environ. 264,
and broad-scale geographic patterns of species richness. Ecology 84 (12), 112571.
3105–3117. Liu, M., Popescu, S., Malambo, L., 2020. Feasibility of burned area mapping based on
Heginbottom, J.A., 1995. Canada-Permafrost. The National Atlas of Canada, Map MCR ICESAT-2 photon counting data. Remote Sens. 12 (1), 24.
4177F 1: 1,750,000 Scale. Natural Resources Canada, Geological Survey of Canada, Lu, J., Zhang, L., 2011. Modeling and prediction of tree height–diameter relationships
Ottawa. using spatial autoregressive models. For. Sci. 57 (3), 252–264.
Hirata, Y., 2004. The effects of footprint size and sampling density in airborne laser MacArthur, R.H., MacArthur, J.W., 1961. On bird species diversity. Ecology 42 (3),
scanning to extract individual trees in mountainous terrain. Int. Arch. Photogramm., 594–598.
Remote Sens. Spat. Inf. Sci. 36 (Part 8), W2. Main, B., Copland, L., Smeda, B., Kochtitzky, W., Samsonov, S., Dudley, J., Skidmore, M.,
Hogg, E.H., Wein, R.W., 2005. Impacts of drought on forest growth and regeneration Dow, C., Van Wychen, W., Medrzycka, D., Higgs, E., 2023. Terminus change of
following fire in southwestern Yukon, Canada. Can. J. For. Res. 35 (9), 2141–2150. Kaskawulsh Glacier, Yukon, under a warming climate: retreat, thinning, slowdown
Hudak, A.T., Lefsky, M.A., Cohen, W.B., Berterretche, M., 2002. Integration of lidar and and modified proglacial lake geometry. J. Glaciol. 69 (276), 936–952.
Landsat ETM+ data for estimating and mapping forest canopy height. Remote Sens. Markus, T., Neumann, T., Martino, A., Abdalati, W., Brunt, K., Csatho, B., Farrell, S.,
Environ. 82 (2–3), 397–416. Fricker, H., Gardner, A., Harding, D., Jasinski, M., Kwok, R., Magruder, L., Lubin, D.,
Hugenholtz, C.H., Lewkowicz, A.G., 2002. Morphometry and environmental Luthcke, S., Morison, J., Nelson, R., Neuenschwander, A., Palm, S., Popescu, S,
characteristics of turf-banked solifluction lobes, Kluane Range, Yukon Territory, Zwally, J., 2017. The ice, cloud, and land elevation satellite-2 (ICESat-2): science
Canada. Permafr. Periglac. Process. 13 (4), 301–313. requirements, concept, and implementation. Remote Sens. Environ. 190, 260–273.
Hunter, M.O., Keller, M., Victoria, D., Morton, D.C., 2013. Tree height and tropical forest Matasci, G., Hermosilla, T., Wulder, M.A., White, J.C., Coops, N.C., Hobart, G.W.,
biomass estimation. Biogeosciences 10 (12), 8385–8399. Bolton, D.K., Tompalski, P., Bater, C.W., 2018. Three decades of forest structural
James, D. (2017). Kluane Lake West Project Yukon Mining Exploration Program Target dynamics over Canada’s forested ecosystems using Landsat time-series and lidar
Evaluation 16-018. plots. Remote Sens. Environ. 216, 697–714.
Jones, M.W., Kelley, D.I., Burton, C.A., Di Giuseppe, F., Barbosa, M.L.F., Brambleby, E., McKinley, D.C., Ryan, M.G., Birdsey, R.A., Giardina, C.P., Harmon, M.E., Heath, L.S.,
Hartley, A.J., Lombardi, A., Mataveli, G., McNorton, J.R., Spuler, F.R., Wessel, J.B., Houghton, R.A., Jackson, R.B., Morrison, J.F., Murray, B.C., Pataki, D.E., 2011.
Abatzoglou, J.T., Anderson, L.O., Andela, N., Archibald, S., Armenteras, D., A synthesis of current knowledge on forests and carbon storage in the United States.
Burke, E., Carmenta, R., Chuvieco, E., Clarke, H., Doerr, S.H., Fernandes, P.M., Ecol. Appl. 21 (6), 1902–1924.
Giglio, L., Hamilton, D.S., Hantson, S., Harris, S., Jain, P., Kolden, C.A., Kurvits, T., Miyamoto, Y., Griesbauer, H.P., Scott Green, D., 2010. Growth responses of three
Lampe, S., Meier, S., New, S., Parrington, M., Perron, M.M.G., Qu, Y., Ribeiro, N.S., coexisting conifer species to climate across wide geographic and climate ranges in
Saharjo, B.H., San-Miguel-Ayanz, J., Shuman, J.K., Tanpipat, V., van der Werf, G.R., Yukon and British Columbia. For. Ecol. Manag. 259 (3), 514–523. https://doi.org/
Veraverbeke, S., Xanthopoulos, G., 2024. State of wildfires 2023–2024. Earth Syst. 10.1016/j.foreco.2009.11.008.
Sci. Data 16 (8), 3601–3685. Moles, A.T., Warton, D.I., Warman, L., Swenson, N.G., Laffan, S.W., Zanne, A.E.,
Jucker, T., Bongalov, B., Burslem, D.F., Nilus, R., Dalponte, M., Lewis, S.L., Phillips, O.L., Pitman, A., Hemmings, F.A., Leishman, M.R., 2009. Global patterns in plant height.
Qie, L., Coomes, D.A., 2018. Topography shapes the structure, composition and J. Ecol. 97 (5), 923–932.
function of tropical forest landscapes. Ecol. Lett. 21 (7), 989–1000. Molinier, M., López-Sánchez, C.A., Toivanen, T., Korpela, I., Corral-Rivas, J.J.,
Kashian, D.M., Romme, W.H., Tinker, D.B., Turner, M.G., Ryan, M.G., 2006. Carbon Tergujeff, R., Häme, T., 2016. Relasphone—Mobile and participative in situ forest
storage on landscapes with stand-replacing fires. Bioscience 56 (7), 598–606. biomass measurements supporting satellite image mapping. Remote Sens. 8 (10),
Khalsa, S.J., Nandigam, V., Borsa, A., Phan, M., & Lopez, L. (2019). Rapid Visualization 869.
and Analysis of ICESat-2 Data using an Intuitive GUI and JupyterHub Notebooks. Muller, J.E., 1967. Kluane Lake Map-Area, Yukon Territory. Commission Géologique du
Khalsa, S.J.S., Borsa, A., Nandigam, V., Phan, M., Lin, K., Crosby, C., Fricker, H., Baru, C., Canada.
Lopez, L., 2022. OpenAltimetry-rapid analysis and visualization of spaceborne Mulverhill, C., Coops, N.C., Hermosilla, T., White, J.C., Wulder, M.A., 2022. Evaluating
altimeter data. Earth. Sci. Inform. 15 (3), 1471–1480. ICESat-2 for monitoring, modeling, and update of large area forest canopy height
Klein, T., Randin, C., Körner, C., 2015. Water availability predicts forest canopy height at products. Remote Sens. Environ. 271, 112919.
the global scale. Ecol. Lett. 18 (12), 1311–1320. Nandy, S., Srinet, R., Padalia, H., 2021. Mapping forest height and aboveground biomass
Kluane First Nation Final Agreement (2003). Minister of Indian Affairs and Northern by integrating ICESat-2, Sentinel-1 and Sentinel-2 data using Random Forest
Development. Retrieved from https://www.aadnc-aandc.gc.ca/DAM/DA algorithm in northwest Himalayan foothills of India. Geophys. Res. Lett. 48 (14),
M-INTER-HQ/STAGING/texte-text/al_ldc_ccl_fagr_ykn_klu_fia_1330357096505_eng. e2021GL093799.
pdf. Narine, L.L., Popescu, S., Zhou, T., Srinivasan, S., Harbeck, K., 2019. Mapping forest
Kluane First Nation Forest Management Plan (2022). Retrieved January, 12, 2024 from aboveground biomass with a simulated ICESat-2 vegetation canopy product and
https://fortitudeconsulting.ca/wp-content/uploads/2022/11/kfn_frm_report_fin Landsat data. Ann. For. Res. 62 (2), 69–86.
al_small-1.pdf. Neuenschwander, A., Guenther, E., White, J.C., Duncanson, L., Montesano, P., 2020.
Korhonen, L., Korhonen, K., Rautiainen, M., Stenberg, P., 2006. Estimation of forest Validation of ICESat-2 terrain and canopy heights in boreal forests. Remote Sens.
canopy cover: a comparison of field measurement techniques. Silva Fennica 40, Environ. 251, 112110.
577–588. https://doi.org/10.14214/sf.315. Neuenschwander, A., Pitts, K., 2019. The ATL08 land and vegetation product for the
Kozlowski, T.T., 1984. Plant responses to flooding of soil. Bioscience 34 (3), 162–167. ICESat-2 mission. Remote Sens. Environ. 221, 247–259.
Krebs, C.J., Boonstra, R., Boutin, S., Sinclair, A.R., Smith, J.N., Gilbert, B.S., Martin, K., Neuenschwander, A.L., Magruder, L.A., 2016. The potential impact of vertical sampling
O’Donoghue, M., Turkington, R., 2014. Trophic dynamics of the boreal forests of the uncertainty on ICESat-2/ATLAS terrain and canopy height retrievals for multiple
Kluane Region. Arctic 71–81. ecosystems. Remote Sens. 8 (12), 1039.
Krebs, C.J., Boonstra, R., Cowcill, K., Kenney, A.J., 2009. Climatic determinants of berry Neumann, T.A., Brenner, A., Hancock, D., Robbins, J., Saba, J., Harbeck, K., Gibbons, A.,
crops in the boreal forest of the southwestern Yukon. Botany 87 (4), 401–408. Lee, J., Luthcke, S.B., Rebold, T., 2020. ATLAS/ICESat-2 L2A global geolocated
Lambers, H., Chapin, F.S., Pons, T.L., 2008. Plant Physiological Ecology, 2. Springer, photon data, version 5. NASA National Snow and Ice Data Center Distributed Active
New York, pp. 11–99. Archive Center, Boulder, CO. https://doi.org/10.5067/ATLAS/ATL03.005.
Lang, N., Jetz, W., Schindler, K., Wegner, J.D., 2023. A high-resolution canopy height Notarnicola, C., 2022. Overall negative trends for snow cover extent and duration in
model of the Earth. Nat. Ecol. Evol. 7 (11), 1778–1789. global mountain regions over 1982–2020. Sci. Rep. 12 (1), 13731.
Parker, G.G., Harding, D.J., Berger, M.L., 2004. A portable LIDAR system for rapid
determination of forest canopy structure. J. Appl. Ecol. 41 (4), 755–767.

12
R.D. Sankar et al. Trees, Forests and People xxx (xxxx) xxx

Parrish, C.E., Magruder, L.A., Neuenschwander, A.L., Forfinski-Sarkozi, N., Alonzo, M., Smith, C.A.S., Webb, K.T., Kenney, E., Anderson, A., Kroetsch, D., 2011. Brunisolic soils
Jasinski, M., 2019. Validation of ICESat-2 ATLAS bathymetry and analysis of of Canada: genesis, distribution, and classification. Can. J. Soil Sci. 91 (5), 695–717.
ATLAS’s bathymetric mapping performance. Remote Sens. 11 (14), 1634. Yukon Ecoregions Working Group, 2004. Yukon coastal plain. In: Smith, C.A.S.,
Peetoom, K., 2018. The Alaska Highway Gas Pipeline Easement: A Kluane First Nation Meikle, J.C., Roots, C.F. (Eds.), Ecoregions of the Yukon Territory: Biophysical
Case Study. University of Northern British Columbia. Properties of Yukon Landscapes. Agriculture and Agri-Food Canada, Summerland,
Pierce, A.D., Farris, C.A., Taylor, A.H., 2012. Use of random forests for modeling and British Columbia, pp. 63–72. PARC Technical Bulletin No. 04-01.
mapping forest canopy fuels for fire behavior analysis in Lassen Volcanic National Smithwick, E.A., Harmon, M.E., Domingo, J.B., 2007. Changing temporal patterns of
Park, California, USA. For. Ecol. Manag. 279, 77–89. forest carbon stores and net ecosystem carbon balance: the stand to landscape
Popescu, S.C., Zhou, T., Nelson, R., Neuenschwander, A., Sheridan, R., Narine, L., transformation. Landsc. Ecol. 22 (1), 77–94.
Walsh, K.M., 2018. Photon counting LiDAR: an adaptive ground and canopy height Sothe, C., Gonsamo, A., Lourenço, R.B., Kurz, W.A., Snider, J., 2022. Spatially continuous
retrieval algorithm for ICESat-2 data. Remote Sens. Environ. 208, 154–170. mapping of forest canopy height in Canada by combining GEDI and ICESat-2 with
Potapov, P., Li, X., Hernandez-Serna, A., Tyukavina, A., Hansen, M.C., Kommareddy, A., PALSAR and sentinel. Remote Sens. 14 (20), 5158.
Pickens, A., Turubanova, S., Tang, H., Silva, C.E., Armston, J., Dubayah, R., Blair, J. Statistics Canada (2021). Kluane First Nation. Aboriginal Population Profile. 2021
B., Hofton, M., 2021. Mapping global forest canopy height through integration of Census. Statistics Canada Catalogue no. 98-510-X2016001. Ottawa. Retrieved
GEDI and Landsat data. Remote Sens. Environ. 253, 112165. November 13, 2023 from https://www12.statcan.gc.ca.
Rahman, M.F., Onoda, Y., Kitajima, K., 2022. Forest canopy height variation in relation Sun, T., Qi, J., Huang, H., 2020. Discovering forest height changes based on spaceborne
to topography and forest types in central Japan with LiDAR. For. Ecol. Manag. 503, lidar data of ICESat-1 in 2005 and ICESat-2 in 2019: a case study in the Beijing-
119792. Tianjin-Hebei region of China. For. Ecosyst. 7 (1), 1–12.
Reid, K.A., Reid, D.G., Brown, C.D., 2022. Patterns of vegetation change in Yukon: recent Tao, S., Guo, Q., Li, C., Wang, Z., Fang, J., 2016. Global patterns and determinants of
findings and future research in dynamic subarctic ecosystems. Environ. Rev. 30 (3), forest canopy height. Ecology 97 (12), 3265–3270.
380–401. Tchebakova, N.M., Parfenova, E.I., Korets, M.A., Conard, S.G., 2016. Potential change in
Roman, L.A., Scharenbroch, B.C., Östberg, J.P., Mueller, L.S., Henning, J.G., Koeser, A.K., forest types and stand heights in central Siberia in a warming climate. Environ. Res.
Sanders, J.R., Betz, D.R., Jordan, R.C., 2017. Data quality in citizen science urban Lett. 11 (3), 035016.
tree inventories. Urban For. Urban Green 22, 124–135. Thomas, R.Q., Hurtt, G.C., Dubayah, R., Schilz, M.H., 2008. Using lidar data and a
Saliu, I.S., Wolswijk, G., Satyanarayana, B., Fisol, M.A.B., Decannière, C., Lucas, R., height-structured ecosystem model to estimate forest carbon stocks and fluxes over
Otero, V., Dahdouh-Guebas, F., 2020. A dataset of tree heights in mangrove and non- mountainous terrain. Can. J. Remote Sens. 34 (sup2), S351–S363.
mangrove trees in Malaysia derived from multiple measurement methods. Data Torita, H., Masaka, K., 2020. Influence of planting density and thinning on timber
Brief. 33, 106386. productivity and resistance to wind damage in Japanese larch (Larix kaempferi)
Sanborn, P.T., Jull, A.J.T., 2010. Loess, bioturbation, fire, and pedogenesis in a boreal forests. J. Environ. Manag. 268, 110298.
forest-grassland mosaic, Yukon Territory, Canada. In: 19th World Congress of Soil Turkington, R., John, E., Watson, S., Seccombe-Hett, P., 2002. The effects of fertilization
Science, Soil Solutions for a Changing World. World Congress of Soil Science and herbivory on the herbaceous vegetation of the boreal forest in north–western
Published on DVD, Brisbane, Australia. Canada: a 10-year study. J. Ecol. 90 (2), 325–337.
Sawada, M., Johnson, P.G., 2000. Hydrometeorology, suspended sediment and Urbazaev, M., Cremer, F., Migliavacca, M., Reichstein, M., Schmullius, C., Thiel, C.,
conductivity in a large glacierized basin, Slims River, Yukon Territory, Canada 2018. Potential of multi-temporal ALOS-2 PALSAR-2 ScanSAR data for vegetation
(1993–94). Arctic 53 (2), 101–117. height estimation in tropical forests of Mexico. Remote Sens. 10 (8), 1277.
Scudder, G.G.E., 1997. Environment of the Yukon. Insects of the Yukon. Biological Wang, C., Zhu, X., Nie, S., Xi, X., Li, D., Zheng, W., Chen, S., 2019. Ground elevation
Survey of Canada (Terrestrial Arthropods), Ottawa, pp. 13–57. accuracy verification of ICESat-2 data: a case study in Alaska, USA. Opt. Express 27
Shufan, W., Chun, L., 2022. Urban forest canopy height modeling using spaceborne laser (26), 38168–38179.
ICESAT-2 LIDAR. In: IOP Conference Series: Earth and Environmental Science, 1004. Wilson, N.J., 2019. Seeing Water Like a State?”: indigenous water governance through
IOP Publishing, 012023. Yukon first nation self-government agreements. Geoforum 104, 101–113.
Shugar, D.H., Clague, J.J., Best, J.L., Schoof, C., Willis, M.J., Copland, L., Roe, G.H., Xu, M., Xiang, H., Yun, H., Ni, X., Chen, W., Cao, C., 2019. Retrieval of forest canopy
2017. River piracy and drainage basin reorganization led by climate-driven glacier height jointly using airborne LiDAR and ALOS PALSAR data. J. Appl. Remote Sens.
retreat. Nat. Geosci. 10 (5), 370–375. 14 (2), 022203.
Simard, M., Pinto, N., Fisher, J.B., Baccini, A., 2011. Mapping forest canopy height Zhang, J., Huang, S., He, F., 2015. Half-century evidence from western Canada shows
globally with spaceborne lidar. J. Geophys. Res. 116 (G4), G04021. forest dynamics are primarily driven by competition followed by climate. Proc. Natl.
Sintayehu, D.W., Belayneh, A., Dechassa, N., 2020. Aboveground carbon stock is related Acad. Sci. 112 (13), 4009–4014.
to land cover and woody species diversity in tropical ecosystems of Eastern Ethiopia. Zhang, J., Nielsen, S.E., Mao, L., Chen, S., Svenning, J.C., 2016. Regional and historical
Ecol. Process. 9 (1), 1–10. factors supplement current climate in shaping global forest canopy height. J. Ecol.
Smith, B., Fricker, H.A., Holschuh, N., Gardner, A.S., Adusumilli, S., Brunt, K.M., 104 (2), 469–478.
Csatho, B., Harbeck, K., Huth, A., Neumann, T., Nilsson, J., Siegfried, M.R., 2019. Zhu, J.L., Shi, Y., Fang, L.Q., Liu, X.E., Ji, C.J., 2015. Patterns and determinants of wood
Land ice height-retrieval algorithm for NASA’s ICESat-2 photon-counting laser physical and mechanical properties across major tree species in China. Sci. China
altimeter. Remote Sens. Environ. 233, 111352. Life Sci. 58, 602–612.
Smith, C.A.S., Meikle, J.C., & Roots, C.F. (2004). Ecoregions of the Yukon Territory: Zhu, X., Nie, S., Zhu, Y., Chen, Y., Yang, B., Li, W., 2023. Evaluation and comparison of
biophysical properties of Yukon landscapes. Agriculture and Agri-Food Canada, ICESat-2 and GEDI data for terrain and canopy height retrievals in short-stature
PARC Technical Bulletin No. 04-01, Summerland. British Columbia, 197–206. vegetation. Remote Sens. 15 (20), 4969.

13