123
Figure 8. Panorama of the mountains from Lecco (Italy)
The purpose of this research was to investigate and to
pursue studies and hypothesis regarding some questions
about this panorama. For example, the point(s) of view of
the panorama were localized with a lot of difficulties. The
hypothesis that this panorama was obtained from
successive surveys on the area is uncertain, because
signs of any principal or secondary geodetic stations in
Lecco were not found.
Moreover the panorama points are out of the ability and
the artistic taste of the author (the shades were obtained
by the watercolor) and besides his knowledge of the area.
These elements are important to identify the author. In
fact references to this panorama were found in a lot of
papers of Carlini written during the period of the Second
Topographic Map of Lombardy, but the notes on the
drawing don't seem to belong to him.
In order to identify the author of this panorama, many
documents of other important surveyors of that period
were analyzed. In the books of geodetic measurements
between 1803 and 1807, the name of the young
engineering Giuseppe Bovara was many times repeated.
His diary, recently found, confirms both his presence
during the geodetic campaigns and his hard work made
with the astronomers of Brera. In 1810, Bovara
specialized in architecture in Rome. He also drew a Map
of the Naples Gulf, showing his strong interest in the
cartographic representation. After leaving Rome, he lived
and worked in Lecco and in its neighborhood. Bovara was
also a big friend of Carlini during all his life.
Analyzing the panorama of Lecco, it seems clear that the
notes have been written by Bovara, instead of by Carlini,
and other coincidences (described later in the paper)
confirm that Bovara was the author.
Anyway there are some other questions to answer to, so
the ancient drawing has been analyzed and compared to
the modern photogrammetric survey.
5. PRELIMINARY ANALISYS
In order to find the point(s) of view, some tops of
mountains and some places were recognized in the
panorama and compared with the measurements made
on the Topographic Map. The points were recognized
using different kind of maps at different scales: the Map of
Touring (scale 1:200.000), the Regional Technical Map of
Lombardy (scale 1:10.000), the Thematic Map of Lecco
District (scale 1:50.000) and, for the historical buildings in
the city, the first Map of Italy (scale 1:100.000) realized by
IGM in the 1888.
In order to make a mathematical analysis, a reference
system on the Regional Technical Map of Lombardy was
assumed. The origin was placed in the intersection of the
cartographic grid at the coordinates 5.078.000 m (North),
1.531.000 m (East) of the sheet B4d4 Lecco (x grows
towards West and y grows towards North). This reference
system was used for the measurements made on the
maps; on the contrary, the measurements on the
panorama, were referred to the first top of the Mont.
Barro, positioned in the left part of the panorama.
The two measurements were compared using a least
square linear regression between azimuth angles and
linear distances: d = k+sQ, where d represents the
horizontal distance measured on the panorama in mm
and 0 the cartographic azimuth. Analogous considerations
were made for the elevation angles a. The reference
system for the elevation angles was the surface of the
sea. As before, the measurements of the elevation angles
from the map were compared to the vertical distances h
measured on the panorama, according to the least square
linear regression: h = k+sa.
Changing the point of view, the residuals for 36 points,
estimated using these relations, were obviously variable.
In fact the correct localization of the some points on the
panorama depended on the selection of the point station.
The point of view placed in Lecco (near Belvedere Street)
provided the least standard deviation. The optimal
solution showed the point of view coordinates, in the local
reference system: x = 911 m, y = 550 m., z = 230 m, the
panorama scales for the azimuth angles: k = 12.1 mm, s =
271 mm, the panorama scales for the elevation angles: k
= 40.5 mm, 5 = 226.7 mm, and a standard deviation of
17.1 mm.
Notice that no points were rejected, because a robust
down - weighting procedure for outlier identification was
set up. Furthermore the preliminary information,
concerning the elevation of the lake (and of the candidate
point(s) of view), was introduced in the system with a very
small weight.
Observing the residuals, they had a systematic behavior.
In fact the drawing was made on two papers and every
paper seemed to have its own point station. Therefore
analyzing separately the two parts of the drawing, two
different points of view, which minimized the mathematical
solution, were found.
The first solution, obtained for the section with the Mont.
Barro, using only 20 points, showed the point of view
coordinates, in the local reference system: * = 860 m, y =
696 m., z = 216 m, the panorama scales for the azimuth
angles: k = 9.5 mm, s = 274.0 mm, the panorama scales
for the elevation angles: k = 15.9 mm, s = 275.4 mm, and
a standard deviation of 8.4 mm. The second solution,
related to the section with the Mont. S. Martino, obtained
with 16 points, showed the point of view coordinates, in
the local reference system: x = 716 m.y = 184 m., z = 210
m, the panorama scales for the azimuth angles: k = 21.5
mm, s = 342.7 mm, the panorama scales for the elevation
angles: k = 54.6 mm, 5 = 197.3 mm, and the same
standard deviation, because the solution is unique.
After this analysis, the great reduction of the standard
deviation confirmed that the panorama was likely realized
from two points of view in Lecco.
