In this 2-days course you will learn how to manipulate spatial data only using the R-software. You will create, import, modify and map vector (shapefiles of points, lines or polygons) and matrix (raster) layers.
You will be able to create your own maps with colors depending on variables of interest (attributes), without forgetting to add legend, spatial scale and metadata, but also while adding external data from openstreetmap or googlemap.
Spatial data manipulation and modeling tools
We will teach you common spatial data manipulation tools:
Projection of spatial layers into different coordinate reference systems
Spatial joints or attributes joints
Buffer area around spatial features
We will also approach specific tools designed for raster and digital elevation models (DEM) analyses, like slope calculation, levelplot, hillshade, …
To go further
For specific needs, we would be able to approach spatial interpolation tools (kriging) or spatial predictive models (see this other course).
We can also propose the equivalent course using QGIS mapping software to show you its complementarity with R for spatial data manipulations.
Contact me for further information and on-demand course.
Generalized Linear Models and Species Distribution Modeling
This 2-days course will teach you how to predict species distribution using the R-software. This is an initiation to statistics, Generalized Linear Models (GLM) and to the use of R spatial libraries.
In this course, species distribution modeling relies on GLM that link a variable of interest (probability of presence, abundance, density, biomass) to environmental covariates.
We use a zero-inflated dataset, a dataset in which an important part of the data is absence of the species of interest. This is a good challenge for GLM modeling and, in ecology, this is rather the rule than the exception. This dataset allows to play with a variety of statistical distributions and tests.
Species distribution models couples outputs of GLM model to spatial information allowing to draw maps of distribution. During the course, you will use spatial data with spatial R libraries and functions to produce geo-referenced maps of predictions.
At the end of the course, you will be able to:
Explore data with different graphical analyses
Choose a statistical distribution adapted to your dataset
Fit a GLM model and analyze the outputs
Select the best model among different combinations of covariates
Use R spatial libraries to deal with spatial datasets
Use R spatial libraries to predict species distribution