Canopy heights obtained from
GLAS
footprints transmitted
with high energy consistently yield better comparative results
than those from footprints transmitted with low energy for
RH
ROS
, where results appear less conclusive for RH
100
.
The mean bias for low and high laser transmission energies
(−0.08 and 0.01, respectively for RH
ROS
, and 0.06 and 0.12,
respectively for RH
1 00
) support the hypothesis that systemati-
cally shorter canopy heights are noted for low energy laser
transmissions; these observations are consistent with
ALS
profile observations noted by Hopkinson (2007).
Site
Controls are filtered according to study sites, where the most
correlated
GLAS
height method and
ALS
pXX are illustrated
in Figure 8 as a function of location and
ALS
data source. Ac-
companying summary statistics for both
GLAS
height methods
and corresponding
ALS
pXX are shown in Table 7. A best
performing comparison is less useful in the context of
GLAS
sensitivities, as site characteristics cannot be controlled un-
like data selection by unique sampling attribute (i.e., trans-
mission energy, or time of data acquisition). Essentially, site
Figure 5. Stratified control comparisons for the best GLAS height method and ALS pXX as a function of ALS data source: (a) all returns,
(b) first returns, and (c) raster; and each GLAS laser number: i) 1, ii) 2, or iii) 3. Linear models and associated 95 percent confidence in-
tervals are also illustrated. Note: grey points indicate best results were achieved where ALS pXX was compared to RH
100
rather than RH
ROS .
T
able
4. S
ummary
S
tatistics
for
C
ontrols when
F
iltered
A
ccording
to
GLAS L
aser
N
umber
;
p
XX
is
the
M
ost
C
losely
R
elating
ALS H
eight
P
ercentile with
its
C
orresponding
GLAS H
eight
D
erivation
. T
he
B
est
O
verall
GLAS/ALS H
eight
C
omparisons
are
B
old
for
E
ach
ALS D
ata
S
ource
. N
ote
: L
N
o
is
L
aser
N
umber
.
RH
100
RH
ROS
Data L
No
N
RMSE
R
2
F
20
|F
B
| pXX
RMSE
R
2
F
20
|F
B
| pXX
Raster
1
24
1.11
18.42
0.04 0.25 0.12 p100 0.82
16.25
0.07 0.42 0.20 p90
2
0.96
14.55
0.22 0.50 0.01 p90 0.78
14.10
0.37 0.54 0.21 p90
3
0.92
7.23
0.42 0.83 0.08 p90
0.81
9.39
0.48 0.54 0.21 p90
All
1
24
1.17
19.89
0.00 0.21 0.17 p100 0.95
17.67
0.00 0.50 0.04 p99
2
0.98
14.54
0.11 0.54 0.00 p100 0.98
8.91
0.24 0.67 0.02 p99
3
1.04
5.62
0.60 0.83 0.04 p99
1.06
4.06
0.78 0.88 0.05 p95
First
1
24
1.25
21.16
0.04 0.25 0.24 p99 1.01
17.86
0.03 0.46 0.03 p95
2
0.98
14.54
0.11 0.54 0.00 p100 0.96
9.01
0.24 0.67 0.05 p99
3
1.02
5.48
0.60 0.83 0.02 p99
1.01
3.60
0.79 0.96 0.01 p95
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
May 2016
357
