Digitalization and Classification of Cesare Battisti’s Atlas of 1915

Abstract

The paper deals with an automated methodology for the digital acquisition of thematic information from historical maps in order to use them for spatial analysis in a GIS software. This methodology has been applied to an early XIX c. map in order to assess the historical changes in the forest coverage in Trentino. Specifically, a tailored Object Based Image Analysis (OBIA) and filtering procedure has been applied to digitize and georeference Cesare Battisti’s map of forest density published in his atlas “Il Trentino. Economic Statistical Illustration” from 1915. According to the historical ecology approach, forest history can be analyzed and evaluated with the use of historical documentary sources. Following this approach, historical cartography is a precious information tool, and in many respects unique, through which it is possible to reconstruct the evolution of the forest cover of a given territory. Trentino, in particular, has a rich heritage of historical maps from which to draw useful information for the construction of a qualitative and quantitative diachronic picture of the evolutionary dynamics of wooded areas. In these territories, forest management is a topic of great importance both for its socio-economic implications and for the more strictly environmental ones, connected to the increasingly urgent need to implement mitigation and adaptation policies towards climate change. Thus, the paper presents the historical maps and illustrates the methodology used for the digitisation. Data extracted by the historical sources have been compared with the current one in order to identify changes in forest density in the last century.

Modeling of Forest Landscape Evolution at Regional Level: A Foss4g Approach

“…The analysis of selected landscape metrics provided preliminary results about the forest distribution and the pattern of recolonization during the last 155 years. While the classification methods described in (Gobbi et al, 2019b) and (Zatelli et al, 2022) are able to achieve high accuracy (Tables 5 and 6), there is still the need for manual calibration of some parameters. This is mainly due to the heterogeneity of historical maps, both in the original maps and in the digitization process (resolution, color depth and image compression), and in the different features of the ortho-photos, especially in terms of resolution and number of bands.…”

“…The classification of areas containing hatching has been carried out by using additional bands, such as texture maps and an high pass filter map on a specific band (Zatelli et al, 2022). Given the large number of maps which have been processed, all the procedures have been scripted, mostly in python.…”

“…Given the large number of maps which have been processed, all the procedures have been scripted, mostly in python. Details of the OBIA classification and of the filtering procedures can be found in (Gobbi et al, 2019b) and (Zatelli et al, 2022).…”

“…It has been transformed into a ETRS89 UTM32N (EPSG 25832) datum. The scanning resolution is 72 DPI, resulting in a ground resolution of 54.48 m in the east and 79.47 m in the north direction (Zatelli et al, 2022). The 1936 Italian Kingdom Forest Map (IKFM, "Carta Forestale della Milizia Forestale del Regno d'Italia del 1936") represents the forest extent and main species composition in the whole Italian territory and parts of some surrounding countries in 1936 (Ferretti et al, 2018).…”

“…The forest class from the LULC maps has been analyzed to identify trends. The "Battisti" map for year 1915 reports forest density for each district (Figure 7) and not forest coverage (Zatelli et al, 2022), therefore it is not directly comparable with the other maps. The overall forest coverage for each year is reported in Table 7 and Figure 14.…”