We will start with data exploration while using chironomid data collected from the Svratka River.
Import the species and environmetnal data.
Check the distribution of environmental variables and transform if necessary. hint
Use Cleveland dot plots, function dotchart() or a function multidot(), which is a simple for loop creating a dotchart for each variable in a given data set.
Check the relationships among environmental variables. hint
Use paired scatter plots pairs() for numerical variables and box plots for numerical variables vs. factors.
Check the distribution of species abundances, look for possible mistakes and correct them if they are mistakes. hint
For this one may convert the species data frame to a matrix to plot all values within one dotchart, or use function dtf.dotchart(), which gives a more visually appealing result. The abundance of two species (each in a different sample) is far much higher then the rest of the data. These are typos. When filling the database, zero was accidentaly pressed. The correct values are 127 (not 1027) and 217 (not 2107).
Calculate distance matrices from the species data for the following distance coefficients:
Euclidaen on ln(x+1) transformed data
Bray-Curtis distance on ln(x+1) transformed data hint
Use function vegdist() from the vegan package.
Compare these distance matrices using scatter plots. Which distances are more related given the data? hint
Use paired scatter plots.
anadat/cs/exercises/cv1.txt · Last modified: 2017/04/15 12:02 by vitek