cv projects publications python

Giovanni Caudullo

Senior spatial analyst


European Atlas of Forest Tree Species - link
August 2013 - March 2016
The project aims at modelling forest resources at European scale by exploiting the harmonised data on forest tree species observed pre-August 2013 sences as aggregated in the European Forest Data Centre (EFDAC) dataset by the E-Forest consortium of European National Forest Inventories, and in other smaller datasets. In particular, the focus has been on modelling architecture and implementation for estimating tree species distribution along with forest tree species habitat suitability. The resulting geo-data layers, metadata and general information on more than 100 forest tree species contribute to design the book European Atlas of Forest Tree Species. An online version of the Atlas has also developed, remaining up to date when improved datasets and modelling methodologies become available. The web platform has been designed to be a Semantic MediaWiki. I took part in the project operating with spatial datasets and forest tree species databases both with ArcGIS and GRASS GIS, organizing data for further statistical processing and producing thematic maps of results. As author, I have contributed in writing several chapters of the Atlas, describing the tree species with information on their ecology, usage and threats, and as editor, I reviewed the contributions of other authors, with checks on quality of references and with the harmonization of the texts.

Supporting EFSA Panel on plant health
October 2011 - Present
The JRC Forest Unit collaborates with the European Food Safety Authority (EFSA) supporting the assessment of the EU exotic plant pests with scientific opinions. In particular, on occurrence JRC provides European and worldwide maps, data and metadata of endangered plants. I worked on spatial data, harmonizing and processing different datasets to generate distribution maps of plant species classified as pest hosts. I contributed and supported the development of modelling procedure for pan European forest tree species suitability maps (1 km spatial resolution), under climate change scenarios, for the tree species requested by EFSA. I collaborated on designing the software tools for XML metadata generation and on populating them with required information.

Connectivity of Natura 2000 forest sites in the European Union - link
September 2012 - March 2014
The JRC, following the recently adopted Biodiversity Strategies to halt biodiversity losses, proposed a connectivity index based on structural and functional criteria for the dispersal of terrestrial wildlife between Natura 2000 network sites, in order to characterize the distribution of the sites in each country and to identify sites likely to be isolated. This index was computed with the free software Conefor Sensinode. The functional distance between sites was computed from a land use based friction map with the ''least cost path'' method, which provided the cheapest friction path from site to site through the landscape matrix. ''Movement resistance'' (or friction) values were assigned to every land use; low values for natural and semi-natural land cover, medium values for agricultural lands and highest values for artificial surfaces (urban, roads). The friction map was created from two layers. First, the Corine Land Cover (CLC) map of year 2006. Second, the main EU road network (highways, motorways, national roads), which represents a significant barrier to most natural wildlife movements, was obtained from the Open Street Map layer. This environmental modelling exercise allows the actual connectivity status of Natura 2000 network to be assessed and compared across countries in a generic manner. I developed the modelling procedure of spatial data for batch processing with Python scripting under ArcGis environment.

European Union Solidarity Fund
June 2013 - January 2014
The European Union Solidarity Fund (EUSF) was established in order to deal with major national disasters. It provides financial assistance to disaster-stricken States. Council Regulation (EC) No 2012/2002 establishes the rules and principles relating to intervention by the EUSF. JRC supplies an analytic and scientific support, in particular providing base and situation maps. I was involved as GIS modeller for developing an automatic procedure to compute the amount of economic losses caused by floods, combining water depths layers with Corine Land Cover and flood damage functions by country in Euro per square meter. The python script produced runs in ESRI environment.

PESETA II - link
May 2011 - May 2012
This project is a part of the JRC project PESETA (Projection of Economic impacts of climate change in Sectors of the European Union based on boTtom-up Analysis) coordinated by JRC/IPTS (the Economics of Climate Change, Energy and Transport unit), it involves several JRC Units and international research institutes. PESETA will contribute to the EU adaptation strategy and the EU coordinated responses to climate change. The IES's FOREST Action focused on the status and trends of European forests in relation to forest extent and fragmentation, forest fires, forest species distribution and dynamics, species habitat modelling and the influences of climate change. I was involved in the workgroup which task is developing a model of tree species habitat suitability (HS) on the basis of species presence datasets and several scenarios of climate change. HS models relating tree species distribution data with environmental characteristics on those locations. Hence HS model was used for predicting species distribution across landscape integrating three main components: a georeferenced sample of observed tree species in Europe (National Forest Inventories); a set of biophysical factors classified in three groups: bioclimatic, soil and topography; a model describing the relationship of the species presence with the biophysical factors. I covered the role of bioclimatic and topographic datasets processing and preparation before the geo-statistical analysis. Python scripts were developed for batch processing climatic time series both with ESRI libraries in ArcMap 10 and GDAL libraries on Unix OS. Grass GIS software was used to develop topographic indicators.

European forest pattern, fragmentation and connectivity - link
Semptember 2009 - March 2012
The project is based on GIS technologies with the scope of developing ecological models for the European forest pattern, mainly for the change analysis of morphological and connectivity indicators over time using European Corine Land Cover and high resolution European Forest Map 1990, 2000 and 2006. Standardized and easily repeatable methods across scales enabled to map the spatial pattern and the edge context of forest habitats, using Morphological Spatial Pattern Analysis (with GUIDOS software) and landscape mosaic indices. Connectivity has been assessed with indices based on habitat availability and topology, accounting for species dispersal capabilities and landscape permeability using Conefor software with Euclidean distance and least cost path modelling. Landscape-level forest pattern, fragmentation and connectivity analysis was performed over Europe at broad scale (CLC 25 ha) and medium scale (HR Forest Map 25 m) and over 60 ''European Biodiversity Observation Network'' square 1x1 km sample areas at local scale (vector 400 m2). All the analyses have been coded with Molder builder and Python scripts for the automation of the processes, statistical analysis was made in R environment (S+ software). The results were mapped and prepared for European Forest web portal ''EFDAC''.

EBONE pattern analysis - link
October 2009 - October 2011
JRC partecipated the FP7 project EBONE to address landscape pattern and connectivity of General Habitat Categories (based on life forms). Landscape-level forest pattern, fragmentation and connectivity analysis was performed over 60 square sample areas (1x1 km) in several European environmental zones. JRC group developed and applied indices and also set up a prototype map viewer which enables to view and query habitat maps and query indices.