User Tools

Site Tools


en:tbrda_examples

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
en:tbrda_examples [2018/04/22 00:17]
David Zelený
en:tbrda_examples [2018/04/22 00:38]
David Zelený
Line 7: Line 7:
 [[{|width: 7em; background-color:​ white; color: navy}tbrda_exercise|Exercise {{::​lock-icon.png?​nolink|}}]] [[{|width: 7em; background-color:​ white; color: navy}tbrda_exercise|Exercise {{::​lock-icon.png?​nolink|}}]]
  
 +FIXME (the end of Ex1)
 ==== Example 1: Vltava river valley dataset ==== ==== Example 1: Vltava river valley dataset ====
 In this example, we will apply constrained ordination on [[en:​data:​vltava|Vltava river valley dataset]]. We will ask how much variance in species composition can be explained by two variables, soil pH and soil depth. Both are important factors for plant growth, and moreover in the study area they are somewhat correlated (shallower soils have lower pH, since prevailing geological substrate is acid). In this example, we will apply constrained ordination on [[en:​data:​vltava|Vltava river valley dataset]]. We will ask how much variance in species composition can be explained by two variables, soil pH and soil depth. Both are important factors for plant growth, and moreover in the study area they are somewhat correlated (shallower soils have lower pH, since prevailing geological substrate is acid).
Line 47: Line 48:
 </​code>​ </​code>​
  
-The two variables explain 8.9% of variance (the row ''​Constrained''​ and column ''​Proportion''​ in the table above, can be calculated also as the sum of eigenvalues for the contrained axes divided by total variance (inertia): (0.04023+0.02227) /​0.70476=0.08868. The first constrained axis (RDA1) explains 0.04023/​0.70476=5.7% of variance, while the second (RDA2) explains 0.02227/​0.70476=3.2%. Note that the first unconstrained axis (PC1) represents 0.07321/​0.70476=10.4% of total variance, which is more than both explanatory variables together. This means that the dataset may be structured by some strong environmental variable(s) different from pH and soil depth (we will check this below).+The two variables explain 8.9% of variance (the row ''​Constrained''​ and column ''​Proportion''​ in the table above, can be calculated also as the sum of eigenvalues for the contrained axes divided by total variance (inertia): (0.04023+0.02227) /​0.70476=0.08868. The first constrained axis (RDA1) explains 0.04023/​0.70476=5.7% of variance, while the second (RDA2) explains 0.02227/​0.70476=3.2%. Note that the first unconstrained axis (PC1) represents 0.07321/​0.70476=10.4% of total variance, which is more than both explanatory variables together; the first two unconstrained explain (0.07321+0.04857)/​0.70476=17.3%. This means that the dataset may be structured by some strong environmental variable(s) different from pH and soil depth (we will check this below).
  
 Let's see the ordination diagram: Let's see the ordination diagram:
Line 55: Line 56:
 {{:​obrazky:​tb_rda_vltava_ordiplot12.png?​direct|}} {{:​obrazky:​tb_rda_vltava_ordiplot12.png?​direct|}}
  
 +What may be those environmental variables associated with unconstrained axes? The ''​vltava.env''​ dataset contains a number of other measured variabels which we may fit as supplementary to the first and second unconsrtrained axis to see which of them is most related to which of them. But here we will do an alternative thing: we will use mean Ellenberg indicator values (mEIV) calculated for each plot based on the species composition and tabulated Ellenberg species indicator values (ecological optima of species along several main environemntal gradients). This approach will illustrate the situation as if in the field we meausured only soil pH and depth (relatively easily obtained variables), and we use these indirect estimates to get idea about which other factors may be important.
  
 +<code rsplus>
 +ordiplot (tbRDA, choices = c(3,4), type = '​n'​)
 +points (tbRDA, choices = c(3,4), display = '​sites',​ pch = as.character (vltava.env$GROUP),​ col = vltava.env$GROUP)
 +ef <- envfit (tbRDA, vltava.env[,​23:​28],​ choices = c(3,4))
 +plot (ef)
 +</​code>​
 +{{:​obrazky:​tb_rda_vltava_ordiplot32_ellenberg.png?​direct|}}
en/tbrda_examples.txt · Last modified: 2018/04/22 00:38 by David Zelený