Program

 

SCHOOL PROGRAM

Monday 26 August 2019

Introduction to the Course and training objectives (school directors).

Basic of statistical models and sampling. Prof. Acutis, Dr. Perego. First class: ANOVA (one way, factorial), regression (linear, non linear and multiple). Second class: Sampling size, number of replication, and sampling design. Practical: use of Anova, regression and general linear models, mixed model. Sample size determination and definition of the number of replication, complex sampling designs (e.g . latin hypercube).

Tuesday 27 August 2019

Quantitative methods in Literature search and analysis: “Meta-analysis for environmental science”.
Dr. Valkama. First class: introduction on literature analysis, Second class: Meta-analysis.  Practical: Creation of a literature database and applications using “OpenMEE” software.

Wednesday 28 August 2019

Topography as source for agro-environmental information. Spatial mixed model. Prof. Märker, Prof. Dario Sacco, Dr. Schillaci, Dr. Veronesi. First class: SAGA GIS terrain analysis, environmental process modelling such as soil erosion and storm flow. Second class : Spatial mixed models  Practical: terrain analysis and erosion modelling using GIS.  Examples and applications of spatial mixed model in R.

Thursday 29 August 2019

Machine learning and Geostatistics for environmental modelling using reference databases. Dr. Veronesi, Dr. Schillaci, Dr. Saia. First class: soil and climate spatial databases available, digital soil mapping and local uncertainty for environmental mapping; Second Class: Geostatistics and Machine learning as a tool for environmental mapping Practical: examples and applications using R and GIS.

Friday 30 August 2019

Digital soil mapping for spatial assessment of the agro-environment. Dr. Schillaci, Dr. Lipani, Dr Saia, prof. Acutis and prof. Märker: First class: Dataset assessment and covariate selection (DEM, Remote sensing covariate, regression based methods, LASSO etc.). Second Class: Digital soil mapping of soil properties (Boosted regression trees, Random Forest); Practical: examples and applications using R and GIS.