2019 FOSS4G Bucharest Talks speaker: Stefano Gobbi

The speaker's profile picture


Fine spatial scale modelling of Trentino past forest landscape (TRENTINOLAND): a case study of FOSS application

Stefano Gobbi(1,2,3), Maria Giulia Cantiani(1), Duccio Rocchini(1,2), Paolo Zatelli(1), Clara Tattoni(1), Marco Ciolli(1), Nicola La Porta(2,3).

1 Università degli Studi di Trento, Dipartimento di Ingegneria Civile, Ambientale e Meccanica, Trento, Italy
2 Fondazione Edmund Mach, San Michele all’Adige (TN), Italy
3Mountfor Research center, San Michele all’Adige (TN), Italy

Trentino is an Italian alpine region (about 6200 Km2) with a forest coverage exceeding 60% of its whole surface. In the past, forest landscape has changed dramatically, especially in periods of forest over-exploitation.
Previous studies in some Trentino sub-regions (Val di Fassa, Paneveggio) have identified these changes and the current trend of forest growth at the expenses of open areas, such as pastures and grasslands, due to the abandonment of rural areas. This phenomenon leads to the rapid Alpine landscape change and profoundly affects the ecological features of mountain ecosystems. To be able to monitor and to take future actions about this trend it is fundamental to know in detail the historical situation of the progressive changes on the land use that occurred over Trentino.
The work aims to comprehensively reconstruct the forest cover of whole Trentino at high resolution (5m x 5m pixels) using a series of maps spanning a long period, consisting in historical maps, aerial images, remote sensed information and historical archives. The datasets were archived, processed and analyzed using the Free and Open Source Software (FOSS) GIS GRASS and QGIS. Historical maps include “Atlas Tyrolensis” (dated 1770), “Theresianischer Kataster” (dated 1859) and Italian Kingdom Forest Map (IKFM) of 1936. The aerial imagery dataset includes aerial images taken in 1954, which have been orthorectified during this research, and orthophotos available for years 1973, 1994, 2000, 2006, 2010 and 2016. Remote sensed information includes Landsat and recent Lidar data, while historical archives consist mostly in Forest Management Plans available since around 1950.
The versatility of the wide variety of modules supplied from the FOSS GRASS and QGIS enabled to perform a diverse set of analysis and pre-processing (e.g.:orthorectification) on a heterogeneous dataset of input images. We will focus on the different strategies and methodologies implemented in the FOSS GIS used to process the various types of geographic data, challenges for the future of the research and the fundamental role of the FOSS systems in this process.
Quantifying forest change in the time-span of our dataset can be used to perform further analysis on ecosystem services, such as protection from soil erosion, and on modification of biome diversity and to create future change scenarios.