2. AVAILABLE DATA
In recent years, Italy and Switzerland have produced elevation
data in different reference frames, produced with different
technologies, resolutions, and accuracies. The first aim of the
WP2 was the collection of the data and metadata for the study
areas. In particular, this phase involved the research and the
collection of local DTMs, i.e. digital terrain models with a
medium or high resolution (at least 50 meters), developed by
local authorities (Regions, National Geodetic Authorities), with
a national or regional extension:
e official DTMs of Lombardy and Piedmont Italian Regions
and Switzerland with resolutions of 20, 50 and 25 meters
respectively,
e DTM with a 10 meters planimetric resolution for the eastern
area of Piedmont,
e Lidar DTM (Wehr et al, 1999, Brovelli et a1.,2004) that
covers the catchement area of the Como Lake with a
planimetric resolution equal to 2 meters,
e high resolution Lidar DTM (1 meter), produced by the
Ministry of Environment, which covers the Lombardy and
Piedmont main hydrographic basins.
The available DTMs are in different reference frames: the
Lombardy DTM is in the Italian Roma 40, Gauss Boaga (R40-
GB) reference frame (West zone), while the other DTMs are
provided in ETRF89-IGM95. Some of them are in UTM
coordinates (the two DTMs of Piedmont and the Lidar DTM of
Como Lake), the others in geographic coordinates (the Swiss
DTM and the HR Lidar DTM). Figure 1 shows the distribution
of the DTMs collected over the project area of interest.
8°E 8°30°E SE 9°30'E 10°E
/ | *
/ S
46°30'N |‘ s ; nan t |
« Ó
N Si: m
OA . PT a s
TS t.
46°N
45°30'N
[I HR hydrographic DTM
[SC] LR Switzerland DTM
LR Piedmont DTM
L771 LR Lombardy DTM
45°N
Figure 1. Distribution of the local DTMs for the area of interest
of the HELI-DEM project
DTM A(qm) | Sim) | C (km)
IT Piedmont 2.5 50 7194
IT Lombardy 5-10 20 23870
CH 1.5-3 25 —5400
IT Lidar ~1 1 4110
Table 2. The available data. A: nominal vertical accuracy. S:
spatial resolution. C: covered area within the Helidem region
The main goal of the HELI-DEM project is the creation of a
multiresolution digital terrain model unified for the alpine and
subalpine area between Italy and Switzerland. Therefore some
conclusions about the final resolution and accuracy of the
resulting model can be drawn depending on the DTMs
available. To this purpose, Figure 1 shows a graphical
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B4, 2012
XXII ISPRS Congress, 25 August - 01 September 2012, Melbourne, Australia
representation of the maximum resolution available for the
provinces involved in the project.
3. CROSS-VALIDATION OF DTMS
To cross validate the DTMs, two different analyses have been
carried out. The first is the comparison between crossborder
DTMs with similar planimetric resolution where they overlap,
ie. in the borders between their domain. The second
comparison has been done between low and high resolution
DTMs where the second one exist. Several cross-validations
between different couples of DTMs have been carried out. As
case studies, in the following sections only some comparisons
will be reported:
e Piedmont (res. 50 m) and Switzerland (res. 25 m) DTMs,
e Lombardy (res. 20 m) and Switzerland (res. 25 m) DTMs,
e Lombardy (res. 20 m) and HR Lidar (res. 1 m) DTMs.
The first and second comparisons are example of cross-
validation between similar DTMs, while the third is a
comparison between LR and HR DTMs.
Since these DTMs are not all in the same reference frame,
before doing the cross-validation, some datasets have been
transformed into a new reference frame. To do that the program
GK2CNV which allows to transform a list of points or a grid
from one original reference/coordinate frame to another one has
been implemented.
The GK2CNV program, developed in FORTRAN language
under the GNU General Public License, allows to transform
spatial data between different reference frames (ETRF89,
ETRF2000, R40-GB, ED50), according to Italian specifications
realized by the Italian Military Survey Office (IGM — Istituto
Geografico Militare, Altamimi and Boucher, 2001, Donatelli et
al, 2002). The transformations are based on latitude and
longitude shift parameters, gridded for the entire national
territory with a spatial resolution of 5' in latitude and of 7'30"
in longitude. The vertices (more than 800) used for the
calculation of the national transformation parameters between
the different frames belong to the National geodetic network.
In the most common case, in which the point to be transformed
does not coincide with a grid node, the shift value is obtained
interpolating (using a bilinear interpolation algorithm) the four
nodes closest to the point.
GK2CNV accepts as input ASCII list of coordinates, a single
DTM (ESRI or GRASS ASCII grid format) or a list of DTMs.
The transformed output of the program is always a list of
coordinates: it is worth to remember that, even if the input is on
a regular grid, the transformed points are no more on a grid
although regularly distributed.
In the following the cross-validations, the implemented
procedures and the results will be presented.
3.1 Cross-validation of similar resolution DTMs
All the LR DTMs have similar resolutions and accuracies: no
one of them can be considered as truth and they have been
compared on some points that do not coincide with the nodes of
one of the original DTMs. The comparison points have been
randomly generated in ETRF89 with a uniform spatial
distribution within the overlapping area of the DTMs. Their
density is about 1point/400 m”.
For each point two elevations have been obtained, through exact
bicubic interpolation from the nodes of the Switzerland DTM
(hs) and then from those of the Piedmont DTM, hp (if the point
is in the overlapping area between Switzerland and Piedmont)
or Lombardy DTM, h, (if viceversa the point is in the
overlapping area between Switzerland and Lombardy). For each
