Plant invasions in different habitat types
Habitat types received relatively higher anthropogenic activities are more vulnerable to plant invasions (Table 5). Propagule pressure may be the main mechanism promoting plant invasions in habitats with relatively high anthropogenic activities (Lockwood et al. 2005).
For example, achenes of Bidens pilosa L. var. radiata may adhere to clothing and disperse with human activities. Besides, resource
availability after disturbance may facilitate plant invasions as well (Davis et al. 2000). Highest naturalized species frequency, cover, and species percentage found in the cemetery habitats may be this case, since it usually has annual disturbances, including uprooting,
prescribed fire, tramping, and herbicide application, during the memorial month in April.
Although anthropogenic activities may facilitate plant invasions, natural disturbances may also induce invasions. Riparian habitat, which suffers from frequent natural disturbances, is the case (Table 5;
Hood and Naiman 2000). Anthropogenic disturbances may introduce exotic species to riparian zones, and natural disturbances may
facilitate their spread (Hood and Naiman 2000). On the other hand, the low invasion in forest habitat may partly due to it generally received relatively lower anthropogenic activities, since forests were still retained natural plant community comparable to abandoned field, crop field, or cemetery. Lower anthropogenic activities may correlate to lower propagule pressure and lower anthropogenic disturbance (Alston and Richardson 2006), which may result in decreased plant invasions.
Beside natural or anthropogenic disturbances, environmental factors, such as light intensity may function as well. Forest was the only low light availability environment in this study, but the
dominant naturalized plants in other habit types (e.g. Bidens pilosa var. radiata) are often the species adapt to open canopy with full sunlight (Table 4). Even if these species enters forest, they may have less competition advantage due to environment intolerance
(Williamson 1996). But we have to be cautious, once the propagule pressure from surrounding habitats are large enough, even the species with poor adaptation can also invade forest (Rejmánek 1989). For example, Bidens pilosa var. radiata was the third dominant species in
forest (Table 4), this demonstrated that the environmental intolerant naturalized species can still invade once the propagule pressure is large enough. Fewer exotic plants have good adaptation to forest (e.g.
Setaria palmifolia and Elephantopus mollis), and these species have
relatively fewer propagule pressure than species which have good adaptation to open canopy with full sunlight (e.g. Bidens pilosa var.
radiata and Ipomoea cairica); plant invasions may have time lag
(Banasiak and Meiners 2009) in this case. There were also studies indicate that plant invasion were serious in forest with higher anthropogenic disturbance (Vitousek 1990, Meyer and Florence 1996); But in this study, we did not distinguish between forests that received higher or lower anthropogenic activity thus we cannot examine the effects of different anthropogenic activities on plant invasion in forest. In Taiwan, the invasion process in forest still needs to be further addressed.
According to the native—exotic richness relationships at different habitats (Table 6 and Table 7), no negative relationships were
demonstrated. This means no habitats demonstrated biotic resistance.
These results may suggest that even the lowest invaded habitat
type—forest, should aware of further invasion in the future.
The effects of anthropogenic and environmental factors on native and naturalized biodiversity
The positive correlations between anthropogenic factors and naturalized biodiversity and dominance (Table 10), implied the facilitative effects of anthropogenic activities on plant invasions.
Among anthropogenic factors, landscape heterogeneity as well as habitat fragmentation are both associated with plant invasion, they may influence habitat diversity, propagule pressure, dispersal, and resource availability (Deutschewitz et al. 2003, Kumar et al. 2006);
for example, habitat diversity may support higher naturalized species richness. These may explain the increased plant invasion with higher landscape heterogeneity in this study.
The negative effects of anthropogenic activities on native communities (Table 10) are probably due to habitat destruction (Vitousek et al. 1997). Exploitation intensity is the anthropogenic factor that can reflect habitat destruction since it mainly comprised the built-up area percentage (Table 8). This is also the only factor
significantly detrimental to native species richness (Table 10).
Propagule pressure may be elevated by anthropogenic activities and result in plant invasions afterwards (Lockwood et al. 2005).
Although propagule pressure cannot be measured directly, factors used in this study including landscape heterogeneity, fragmentation, built-up area percentage, and road length can all be the proxy measures of propagule pressure (Chytry et al. 2008) and these were positively correlated to naturalized species richness (Table 8 and Table 10).
Furthermore, anthropogenic disturbance (e.g. habitat destruction) may also enhance plant invasion, since naturalized species richness
percentage were positively correlated to exploitation intensity (Table 8 and Table 10).
The results of elevation-temperature factor on biodiversity (Table 10) were mainly affected by anthropogenic activities.
According to Hsieh (2002), highest native plant richness situated at low elevation in Taiwan. However, the highest native richness
appeared at higher elevation in this study, this result can be attributed to lower anthropogenic activities at higher elevation (Table 11).
Lower anthropogenic activities cause less habitat destruction to
native community. The results demonstrated that naturalized species richness decrease with increasing elevation (Table 10), which is consistent with previous studies (Stohlgren et al. 2002, Pauchard and Alaback 2004, Becker et al. 2005, Daehler 2005, McDougall et al.
2005). This may be relevant to four main factors previous studies proposed (Pauchard et al. 2009): (1) adaptation of naturalized plants to abiotic conditions, (2) disturbances especially anthropogenic
disturbances, (3) biotic resistance of native community, (4) propagule pressure of naturalized plants. In this study, the results can mainly be attributed to (2) disturbance and (4) propagule pressure, these factors are highest at lowland where anthropogenic activities are high (Table 11). Also, the results can be attributed to poor adaptation of
naturalized plants at higher elevation. Naturalized plants may have poor adaptation at higher elevation because they have higher chance to be transported to lowland and adapted to there (Becker et al. 2005).
After exotic plants adapted to lowland, they became less adapted to higher elevation and may result in lower invasion at higher elevation.
Also, most naturalized plants in Taiwan comes from tropical zones (Wu et al. 2004); fewer naturalized plants comes from temperate
zones. Which suggest that fewer species have good adaptation at higher elevation comparable to lower elevation. We did not think abiotic conditions be harsh enough to constraint naturalized species (McDougall et al. 2005) in this study because the highest elevation only reaches 2000m (Fig. 4) where are no snow and still have long growing seasons.
The results of precipitation factor on biodiversity (Table 10) were not consistent with most previous studies (e.g. Dukes and Mooney 1999). Most previous studies situated in arid or semi-arid zones, and demonstrated that areas with higher precipitation were more invaded.
Nevertheless, there were also explanation about wet terrestrial habitats and its invasibility (Rejmánek et al. 2005): “Wet terrestrial habitats do not provide resources – mainly light – for invaders
because of fast growth and high competitiveness of resident species.”
This explanation based on the fact that net primary production increase with precipitation (e.g. Williams et al. 2005), but according to the study in Hawaii (Schuur and Matson 2001, Austin 2002), net primary production had no observed increase with precipitation when annual rainfall higher than 2000mm. Even more, primary production
may be lower at highest rainfall. In this study, average annual precipitation higher than 3000 mm (Fig. 3) where increase in precipitation cannot increase primary production any further. The explanation about precipitation—productivity does not fit in this study. Instead, reasonable explanation is that characteristics of naturalized plants had poor adaptation to highest rainfall area in Taiwan. Poor adaptation of naturalized plants to extremely wet habitats may lower the plant invasions in Taiwan but this hypothesis needs to be further examined.
Native—exotic richness relationships
The negative native—exotic richness relationships in lower anthropogenic activities group at fine scale (Table 13 and Table 15) may be explained by biotic resistance: the diversity of native
community can provide resistance to exotic plant invasion (e.g.
Levine and D'Antonio 1999). Biotic resistance is the most possible process among three processes contributed to negative native—exotic richness relationships (Table 1) because (1) the negative relationships were consistently appeared in plot groups with lower anthropogenic activities which may not simply due to statistical artifact (2) the plant
invasions were lowest in plot groups with lower anthropogenic groups where invasional meltdown was not possible to take place.
The positive relationships were much more complicated to explain. First, biotic resistance disappeared at higher anthropogenic activities. Second, some abiotic factors affected both native and naturalized plants at higher anthropogenic activities (may
simultaneously have higher landscape heterogeneity). Although landscape heterogeneity provides more habitat diversity and can comprise more species richness (Kumar et al. 2006), this is not the case in this study because landscape heterogeneity only had positive relationships with naturalized plants but had no relationships with native plants (Table 10). There was no evidence that landscape heterogeneity favors both native and naturalized plants. No directly effects of landscape heterogeneity can explain positive relationships.
Here, I propose two compatible hypotheses that may operate indirectly to result in positive relationships. First one is propagule pressure. Different landscape heterogeneity may have different propagule pressure (higher heterogeneity may have higher
disturbance and enhance immigration rate, Fridley et al. 2007), and
result in the same directional change of native and naturalized plants.
But the magnitudes of landscape heterogeneity and propagule
pressure are not exactly equals. Propagule pressure affected by many factors including plant distribution, distance to river, plantation, anthropogenic activities facilitate dispersal, and habitat
destruction …(Lonsdale 1999, Chytry et al. 2008). Landscape heterogeneity only reflects part of propagule pressure; this is why positive relationships occur at higher landscape heterogeneity and no evidence that landscape heterogeneity can favor both native and exotic plants. The second explanation is the unsaturated native communities. The niche saturation effect (Moore et al. 2001, Gilbert and Lechowicz 2005, Fridley et al. 2007, Stohlgren et al. 2008) proposed that when niche is saturated then no empty niche exists and immigration is difficult. In the contrary, sites with higher landscape heterogeneity are harder to be fully occupied (unsaturated) by relatively constrained species pool (Fridley et al. 2007). If the niche is unsaturated, native species and naturalized species are easily to immigrant and richness may increase simultaneously. Some evidence in this study can support this hypothesis (Table 16). Integrated high
anthropogenic activities with low elevation group have lower total cover percentage and native cover percentage than integrated low anthropogenic activities with high elevation group. Indicate the plant community is less saturated at higher anthropogenic activities. These two explanations: propagule pressure and unsaturated communities may be the cause of positive native—exotic relationships at sites with higher landscape heterogeneity.
The results of native—exotic richness relationships at different environmental groups (Table 14) suggest that higher precipitation may help native community to limit exotic plant invasion. And may imply that biotic resistance was not the reason why plots with higher elevation were less invaded. Anthropogenic activities may still be the most important determinant on plant invasion at higher elevation in this study.
The results of native—exotic richness relationships at different spatial scale were consistent with previous studies (Richardson and Pysek 2006): positively relationships mainly at broad scale and negative relationships at fine scale (Table 14 and Table 15). The explanation should be careful as Fridey et al. (2007) concluded:
“natively rich ecosystems are likely to be hotspots for exotic species, but the reduction of local species richness can further accelerate the invasion of these and other vulnerable habitats.”
The mechanisms affect native—exotic richness relationships at different plot characteristics and different spatial scales are similar:
“biotic” or “abiotic”. Negative relationships result from biotic resistance; positive relationships result from abiotic factor (e.g.
landscape heterogeneity). Also, disappear of negative relationships may suggest the disappearance of biotic resistance and may promote further plant invasion.
Conclusion
In summary of all the results and discussions, the most important process of plant invasion comes from anthropogenic activities which lead to increase propagule pressure. Also, anthropogenic activities cause disturbance to native community which decrease native biodiversity and dominance. This made biotic resistance disappear and resource availability increase. In this case, the relative dominance of naturalized species increased considerably.
This can explain that some habitats (e.g. cemetery and plots with high exploitation intensity) were seriously invaded. The areas where contain many different habitat types inevitably cause habitat
destruction to native species, so the heterogeneity cannot enhance native diversity. But the area with higher landscape heterogeneity will be much unlikely to be saturated and have open opportunity for native and naturalized species to immigrant. This may interact with propagule pressure to result in positive native—exotic richness relationships. Environmental factors have relatively smaller effects than anthropogenic activities, but the effects were still significant.
The processes may be relevant to adaptation and competition.