Take-home points

Ordinations

Unconstrained ordination

Not a great fit and no positive results re: dissimilarity and time-since-fire.

NMDS ordination with Canberra distance measure. First two axes explain 43% of total variation.
MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 MDS9 MDS10
Eigenvalue 1.21 0.45 0.42 0.33 0.27 0.23 0.20 0.17 0.14 0.10
Proportion Explained 0.31 0.12 0.11 0.08 0.07 0.06 0.05 0.04 0.04 0.03
Cumulative Proportion 0.31 0.43 0.53 0.62 0.69 0.75 0.80 0.84 0.88 0.90
Unconstrained ordination (NMDS with Canberra distance matrix).

Unconstrained ordination (NMDS with Canberra distance matrix).

Constrained ordination

Using a constrained ordination with time-since-fire as a constraint and variation conditioned by Year provides a better picture of dissimilarity among fire management histories (TSF) but whether these groups are meaningfully different is difficult to support statistically. In all of the analyses, the year effect is the most explanatory and TSF is marginally significant at best.

Differentiation between communities by TSF in Constrained Analysis of Proximities (CAP) with Canberra distance measure. Despite the groupings, the test does not perform well: The model is only marginally significant in permutational ANOVA (F=1.19, P=0.08) and first two axes account for only 17% of variation.

Differentiation between communities by TSF in Constrained Analysis of Proximities (CAP) with Canberra distance measure. Despite the groupings, the test does not perform well: The model is only marginally significant in permutational ANOVA (F=1.19, P=0.08) and first two axes account for only 17% of variation.

Testing constrained ordination

Despite the nice groups, little statistical support for the constrained model:

Variation explained by constrained Canberra ordination. The a priori constraint, TSF, explains little variance (CAP1-4 = 25% variance, while first unconstrained axis alone (MDS1) = 22%.)
CAP1 CAP2 CAP3 CAP4 MDS1 MDS2 MDS3 MDS4 MDS5 MDS6
Eigenvalue 0.29 0.23 0.16 0.10 0.69 0.31 0.28 0.25 0.17 0.15
Proportion Explained 0.09 0.08 0.05 0.03 0.22 0.10 0.09 0.08 0.06 0.05
Cumulative Proportion 0.09 0.17 0.22 0.25 0.47 0.57 0.66 0.74 0.80 0.85 
## 
## Permutation test for capscale 
## 
## Blocks:  strata 
## Permutation: free
## Number of permutations: 999
##  
## Call: capscale(formula = ghop.spp ~ ghop.d$TSF +
## Condition(ghop.d$Year), distance = "canb", sqrt.dist = FALSE,
## metaMDSdist = FALSE)
## Permutation test for all constrained eigenvalues
## Pseudo-F:     1.185857 (with 4, 14 Degrees of Freedom)
## Significance:     0.082

Dissimilarity via perMANOVA

Marginal support at best for TSF in permutational MANOVA to test for meaningful differences among TSF groups. Year is significant.

## Blocks:  strata 
## Permutation: free
## Number of permutations: 999
## 
## Type II tests
## Response: ghop.spp
##           Sum Sq Mean Sq Df      F Pr(>F)    
## Year      0.6711 0.67111  1 4.5206  0.001 ***
## TSF       0.7828 0.19569  4 1.3182  0.037 *  
## Year:TSF  0.8257 0.20642  4 1.3905  0.067 .  
## Residuals 1.4846 0.14846 10                  
## Total     3.8918         19                  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1