Find the EGU on

Tag your tweets with #EGU16

SSS12.11/GM2.4

Learning from spatial data: representation, inference and modelling in earth and soil sciences (co-organized)
Convener: Sebastiano Trevisani  | Co-Conveners: Paulo Pereira , Jean Golay , Igor Bogunović , Marco Cavalli 
Orals
 / Wed, 20 Apr, 15:30–17:15  / Room -2.21
Posters
 / Attendance Wed, 20 Apr, 17:30–19:00  / Hall X1
Spatial and spatiotemporal data are crucial for the analysis and modelling of the processes of interest in Earth and Soil Sciences; the heterogeneity characterizing the typology and quality of available datasets coupled with the complexity of the studied phenomena require advanced mathematical and statistical methodologies in order to fully exploit the informative content at hand.
The session aims to explore the challenges and potentialities of quantitative spatial data analysis and modelling in the context of Earth and Soil Sciences. Studies presenting applied mathematical approaches according to an intuitive approach and highlighting the key potentialities and limitations are particularly appreciated. The main interest is toward studies applying techniques and methodologies that make the data “talk” to us about the studied geo-environmental processes and factors; from this perspective we refers to a broad suite of mathematical and statistical techniques such as (but not limited to!):
• Machine learning
• Statistical learning theory
• Geostatistics
• Geomorphometry and other GIS related techniques for terrain analysis
• Pattern analysis and recognition
• Expert systems (e.g., fuzzy systems) combining expert knowledge and spatial data
• Alternative techniques of representation of spatial data (e.g.. visualization, sonification, haptic devices, etc.)

The session aims to discuss three key elements of spatial analysis, emphasizing the connections between spatial data and geo-environmental processes and factors:
1) Analysis of sparse (fragmentary) spatial data for mapping purposes with evaluation of spatial uncertainty
2) Analysis and representation of exhaustive spatial data at different scales and resolutions (e.g., geomorphometry, pattern recognition, etc.)
3) Spatial modelling, possibly using the results from points 1 and 2, of the physicochemical processes and aspects of interest (e.g., surface flow processes, landslides susceptibility models, landscape evolution models, ecological modelling, etc.).
It is our intention to consider selected contributions to promote a special number in an international ISI Journal.