Residence time determines the distribution of alien plants - Invasive Plants: Ecological and Agricultural Aspects - page 69

Table 3. Significance (F, df, P) of explanatory variables in the minimal adequate model of the fre-
quency of alien species in the Czech Republic when regressed on the minimum residence time togeth-
er with other species traits (factors are in italics , other traits are standardized covariates).

Explanatory variable F df P

Grime life strategy 2.47 6, 203 <0.05

Origin 5.18 2, 199 <0.01

Dispersal mode 3.59 3, 200 <0.05

( Grime life strategy ) × (minimum residence time) 2.49 6, 203 <0.05

( Life history ) × (minimum residence time) 4.03 3, 200 <0.01

( Grime life strategy ) × (height) 2.86 7, 204 <0.01

(F = 14.52; df = 10, 207; P < 0.001), while all species traits not including
MRT together explained only 16.1% of variance (F = 3.62; df = 18, 215;
P < 0.001). This can be considered as a robust evidence of the overwhelming
effect of the residence time on current pattern in the distribution of alien spe-
cies. It must be kept in mind that the results reported here concern a single
region, but given that the simple regression of frequency on MRT for this data
set is well within the range of values found for other models summarized in
Table 1, it might be of general validity.
Data on naturalized aliens of the Czech Republic allow another view on the
relative importance of MRT. Williamson et al. [30] give the maximum rate of
spread for a subset of species that were found to have a straight section on a
logarithmic plot of the cumulative number of quadrats over time. The rate of
spread of an alien species in the invaded region is constrained by environmen-
tal factors, biological and ecological barriers as well as the resistance of local
plant communities to invasion; hence the variation in this parameter reflects the
relative importance of these factors. Multiple regressions, relating the distribu-
tion range to both explanatory variables yielded a significant relationship
(F = 7.39; df = 2, 24; P < 0.01) and explained 38.1% of the variance. Both
explanatory variables, i.e., the MRT (F = 14.16; df = 1, 25; P < 0.001;
R 2 = 36.5%) and the rate of spread (F = 8.87; df = 1, 25; P < 0.01; R 2 = 22.9%)
were significant and contributed to the current distribution range of naturalized
aliens. Path analysis, a convenient tool for evaluating the relative effect of these
two factors (Tab. 4), revealed strong direct effects of MRT, a positive one on
range (b 2 ) and negative on the rate of spread (a 1 ). The earlier the species
arrived, the wider range it occupies, and the later it arrived, the more slowly it
has been spreading. When summing the positive direct and negative indirect
effect of MRT (the latter manifested via the rate of spread) on range (b 2 +a 1 b 1 ),
the total effect of MRT on the current distribution appears to be weaker than
the total effect of the rate of spread (Tab. 4).
Residence time is therefore less important than the rate of spread in deter-
mining the present distribution range of naturalized Czech aliens. This result