Resolving clustering issues in historical datasets for modelling
In this short introductory webinar, we will discuss the issue of clusters in historical datasets that can skew your spatial model. We will outline the problem and discuss the 2 most used techniques you can use the solve the problem with spatial modelling in mind. Spatial modelling in VECMAP itself is not discussed in this webinar but will be the topic for a future webinar.
Spatial modelling using heterogeneous data: VECMAP to the rescue
Whilst VECMAP provides the means to produce well-structured vector datasets, many vector and host distribution datasets are heterogenous and need quite a lot of pre-processing before they can be used for modelling. In our second VECMAP webinar, Dr William Wint, a researcher at the Environmental Research Group Oxford, presents examples of what can be done to convert and enhance ‘training’ distribution datasets. and therefore maximising results from VECMAP spatial distribution modelling.
Using VECMAP to model historical data sets the case of the mosquito-borne SINDBIS
For our third VECMAP Webinar Ruut Uusitalo, a doctoral researcher at the University of Helsinki will present her approach and findings comparing results from VECMAP and the biomod2 package in R when investigating the spatial distribution of the mosquito-borne Sindbis virus causing Pogosta disease in Finland.
Achieving the best balance for sample sizes with VECMAP
This webinar will be given by Guy Hendrickx from Avia-GIS and will focus on sample sizes: how many samples are needed to prepare for a robust spatial model? Not much has been written on this topic, whilst this is of critical importance. At Avia-GIS we have been working on providing an answer that we would like to discuss with you.