Botanical Studies (2009) 50: 217-227.

*

Corresponding author: E-mail: esclove@ntu.edu.tw;

+886-2-33664640; +886-2-23697658.

INTRODUCTION

Invasive species have been considered as major threats

to the ecosystems of national parks by altering the soil

nutrients, taking over territories, forming monocultures,

changing species compositions, etc (Rogers and

Leathwick, 1996; Wuerthner, 1996; Jesson et al., 2000;

Evans et al., 2001; Wolf et al., 2004). The way in which

invasive species get into these native habitats are rarely

discussed (Gelbard and Harrison, 2005; Foxcroft et al.,

2007, 2008). Human activity may be to blame (Larson

et al., 2001). We believe that the recreation function

of national parks brings people to appreciate natural

Patterns of plant invasions in the preserves and

recreation areas of Shei-Pa National Park in Taiwan

Shan-Huah WU

1

, J. K. TSAI

2

, H. T. SUN

2

, C. F.

CHEN

2

, and Chyi-Rong

CHIOU

3,

*

1

Biodiversity Research Center, Academia Sinica, 128, Sec. 2, Academia Rd., Taipei 115 Taiwan

2

Department Life Science, National Taiwan Normal University, 88 Sec. 4, Ting-Chow Rd., Taipei 116, Taiwan

3

Department Forestry, National Taiwan University, 1, Sec. 4, Roosevelt Rd., Taipei 106, Taiwan

(Received June 25, 2008; Accepted November 27, 2008)

ABSTRACT.

Nature preserves in the national parks are usually adjacent to the recreation areas, where

most of the tourists visit. Although permits are required and only few small trials are available to enter

the preserves, species naturalized in the neighboring recreation areas may hitchhike across the borders. To

estimate the differences of plant invasions in neighboring preserves and recreation areas experiencing different

intensity of anthropogenic activities, we employed Wuling district (alt. 1,800-3,860 m), Shei-Pa National Park

in Taiwan as our study site. Our hypotheses were: (1) the recreation areas harbor more naturalized species,

and plant invasion patterns are different in these areas under various land management strategies; (2) species

inhabiting the preserves could be found in the recreation areas as well; (3) naturalized species of temperate

origins are dominant due to the temperate weather in the mountains. Total of 230 quadrats in one meter square

quadrats were randomly selected along the roads and trails in both areas. Naturalized species, relative cover,

elevation, and naturalness degree were obtained and analyzed. The results showed that the naturalized species

in both areas were herbaceous, originating from tropical and temperate Americas and Europe. Naturalized

floras of these two areas were presented by analogous dominant families, Asteraceae and Poaceae, and

dominant species, Bromus catharticus and Trifolium repens. However, the number and coverage of naturalized

species, £\ diversity, elevation, and naturalness degree, suggested different patterns of plant invasions of these

two areas. Recreation areas accommodated significantly more naturalized species and higher coverage, and

elevation was responsible for distinct patterns of plant invasions. Both of the preserves and recreation areas

in Wuling provided suitable habitats for similar naturalized floras; however, relatively more species harbored

by the later implied a source and sink relationships between these two areas. Furthermore, environmental

factors that change with the elevation, such as temperature, topography, and native vegetation, may contribute

to different patterns of plant invasions presented by preserves and the recreation areas in the subtropical

mountains.

Keywords: National park; Naturalized species; Nature preserve; Plant invasions; Recreation area; Shei-Pa

National Park.

wonders, but also increases chances of possible invasions,

since land use intensity has been contributive to plant

invasions in both recreation and protected areas of national

parks (Mckinney, 2002; Foxcroft et al., 2007). Although

the disturbance caused by visitors and transportation

infrastructures is mostly concentrated in the designated

recreation areas, these areas may serve as a base for

invaders to expand their territories into adjacent protected

areas, nature preserves, and cause problems (Wein et al.,

1992). Despite the endeavors scientists have devoted to

estimating plant invasions in national parks (Schwarz

and Wein, 1997; Jesson et al., 2000; Rowlands and Brian,

2001; Pauchard et al., 2003; Foxcroft et al., 2007, 2008),

patterns of plant invasions in neighboring recreation and

protected areas with different anthropogenic disturbances

are seldom reported.

ECOLOGY

218

Botanical Studies, Vol. 50, 2009

According to Py.ek et al. (2004), many environmental

and biological barriers have to be conquered for an

introduced species to invade new territories successfully.

Environmental factors, such as disturbance, usually

provide more opportunities for successful invasions

(Ross et al., 2002; Hendrickson et al., 2005; Alston and

Richardson, 2006). However, disturbance could be an

environmental barrier for introduced species to invade

places with low intensity of anthropogenic activities

or pressures of natural disturbances (Kang et al., 2007;

Paiaro et al., 2007; Theoharides and Dukes, 2007). On

the other hand, the invaders¡¦ successful establishment

and spreading may depend upon other factors as well,

such as biological traits and climates (Wu et al., 2004b;

Thuiller et al., 2006; Wilson et al., 2007). We assumed

that alien species successfully establishing in native

habitats might be considered as respective competitors and

deserve attentions. Besides, patterns of plant invasions

in the preserves might generate better understanding of

invasiveness.

The compositions of naturalized flora and dominant

naturalized species seemed to vary along the elevation

gradients in the central mountains in Taiwan (Wu et al.,

unpublished). Our previous study (Wu et al., 2004b)

suggested that dominant tropical American species in the

naturalized flora of Taiwan might be a result of similar

tropical and subtropical climates. Besides, noticeable

amount of temperate species from Europe might find

suitable habitats in the mountains where national parks and

natural reserves are located. This is worrisome since these

montane preserves with temperate to alpine climates are

usually immediately adjacent to the recreation areas, and

impacts of invasive species in natural reserves have been

shown in the boreal ecosystems (Pauchard et al., 2003;

Rose and Hermanutz, 2004). However, basic information

about plant invasions in these areas is anecdotal.

Shei-Pa National Park is a unique and popular

national park with 51 mountain peaks over 3,000 meters

in elevation located in northern Taiwan. The undulated

topography changing from about 1,000 to 3,886 meters,

together with climate variations, result in the diversified

habitats of Shei-Pa National Park harboring 61 endemic

and rare flora, such as Taiwan Sassafras (Sassafras

randaiense), Devol¡¦s Balsamine (Impatiens devolii),

Dumasia miaoliensis, Epilobium nankotaizanens, and

fauna, like Formosan Black Bear (Selenarctos tibetanus

formosanus), Formosan Macaque (Macaca cyclopsis

Swinhoe), Formosan Land-locked Salmon (Oncorhynchus

masou formosanus), Mikado Pheasant (Syrmaticus

mikado),and Swinhoe¡¦s Blue Pheasant (Lophura swinhoii)

(Wang, 1995). Despite its remote location and frequent

landslides on the main entrance ways, the major district of

Shei-Pa National Park, Wuling, is one of the most popular

destinations to the public and the most developed area

compared to other districts of this national park. A five star

resort, few hotels, vegetable farms, fruit and tea plantations

are the dominant scenes of this region. These crop fields

and plantations are directly adjacent to the preserves

and right on the side of Cijiawan Creek, the home of

endangered landlocked salmon, endemic Oncorhynchus

masou formosanum (Tseng, 1999). There is no restriction

for visitors to enter the recreation area, while permits are

required for accessing the neighboring preserves. The trail

and cabin system is well established in Wuling district,

which leads backpackers to the mountain peaks deep

inside the preserves from the recreation areas, in which the

ranger stations situated. In summer, the mountaineering

season, the cabins are usually full of hundreds of people

during holidays and long weekends. We were concerned

that invasive species may hitchhike into the roadless nature

preserves along the trails by mountain hikers (Harrison et

al., 2002).

Abandoned fields are becoming common recently in

the recreation areas in Wuling, and naturalized species

are usually seen and dominant in these fields. Following

the act of retreating farms and fruit plantations out of the

national park in recent years, many abandoned farms have

been scattering in the recreation areas of Wuling. Although

some of these fields are still under certain levels of

farming, the grounds are usually occupied by naturalized

species and persistent crops. Besides, the groundcover

of less managed fruit plantations is usually a mixture of

naturalized and native herbaceous species. Landslides and

disturbances of different intensities caused by frequent

visiting typhoons may also exacerbate plant invasions

since naturalized species usually favor disturbed areas

and they may get into the neighboring preserves by edge

effects (Lu and Keping, 2006).

The main purpose of our study was to approach and

compare the patterns of plant invasions in both preserves

and recreation areas, which represented two different

intensities of land management and use, in the Wuling

district of the Shei-Pa National Park. We hypothesized that

(1) the recreation areas should harbor more naturalized

flora, and plant invasion patterns are different in these

areas under various land management strategies; (2)

species inhabiting the preserves could be found in

the recreation areas as well; (3) naturalized species of

temperate origins are dominant due to the temperate

weather in the mountains. The field investigation was

conducted in summer, 2006.

MATERIALS AND METHODS

Study site

We chose Wuling district of Shei-Pa National Park

(24¢X23¡¦15.6" N latitude, 121¢X18¡¦08.0" E longitude) (Figure

1) for this study, since it is the only district with accessible

nature preserves and recreation areas experiencing

different levels of anthropogenic activities. Shei-Pa

National Park is located in northern Taiwan, and Wuling

is in the east of this national park with conservation

areas, special scenic areas, general protected areas, and

recreation areas. We grouped these four types of areas into

WU et al. ¡X Plant invasions in the areas of Shei-Pa National Park in Taiwan

219

Figure 1. Location and topography of the study site, Wuling,

Shei-Pa National Park, Taiwan (24¢X23¡¦15.6" N latit ude, 121¢X

18¡¦08.0" E longitude).

two major categories, recreation areas (general protected

and recreation areas) and preserves (conservation and

special scenic areas) for our study based on their land

management types.

The recreation areas are relatively smaller than adjacent

preserves possess major attractions and host majority of

the visitors. Elevations the preserves are approximately

1,800 to 3,850 meters. Annual average temperature is

about 16¢XC and the annual average precipitation is about

2,200 mm (Wang, 1995; Ou et al., 2006). The vegetations

of the recreation areas are gardens, abandoned fields,

vegetable farms, tea plantations, temperate cloud forests

(Quercus forests and Tsuga-Picea forests), and riparian.

The preserves are mainly covered by temperate cloud

forests, conifer forests, alpine grasslands, and Abies

woodlands (Hsieh et al., 1997). The alpine grassland, the

dominant vegetation over 3,000 meters in elevation, is

composed of very short bamboo, Yushania niitakayamensis

(Hayata) Keng f. (Wang, 1995).

The transportation system in the recreation area is

simply one county road together with a few outlets. Only

hiking trails lead to the preserves. The entrances of these

hiking trails to the preserves, however, are usually at the

ends of the outlets extending from the recreation areas.

No restriction is applied for entering the recreation areas,

while passes are required to approach the preserves.

Materials

To access the patterns of plant invasions, naturalized

species were utilized to present potential invaders, since

naturalization is the fundamental step for an introduced

species to invade new territories (Py.ek et al., 2004). Wu

et al. (2004a) was employed for identifying naturalized

species. Since the list was published a few years before

the study, relevant journal publications were also used as

supplementary.

Methods

Random sampling was implemented in this study for

plot selection. Approximately ten quadrates of 1 ¡Ñ 1 square

meter were randomly selected in every kilometer along the

roads and trails in both the recreation areas and preserves.

The distances of the quadrats to the roads ranged from one

to ten meters according to random number table. For each

quadrat, species and relative coverage (%) of naturalized

species were recorded. Furthermore, environmental

factors, including degree of naturalness, vegetation

type, and elevation were collected. Naturalness degree

simply classifies habitats into six levels according to the

vegetation types and anthropogenic utilizations: level

five represents native forests, four for native grasslands,

three for timber plantations, two for croplands and fruit

plantations, one for bare grounds, and zero indicates

constructions and infrastructure (Huang et al., 1999).

Data obtained in the field were complied and analyzed

for preservers (P) and recreation areas (R) for comparisons.

Simple statistics was performed to summarize basic

numerical data, such as numbers of family, genus, and

species. Additionally, abundance was estimated by the

total relative coverage (%) of each naturalized species,

and relative frequency was calculated by the percentage of

total plots divided by present plots. IVI (Importance value

index) values of naturalized species were estimated by half

of the sum of relative abundance and relative frequency.

Species with the highest abundance and IVI value were

both considered as the dominant species. As for the

origins, life forms and habit of most naturalized species,

we referred to Wu et al. (2004a).

Total coverage of naturalized species (SUM), number

of naturalized species (N), and six £\ diversity indices,

including Shannon-Wiener index (H¡¦), Simpson (£f),

evenness (E5) (Ludwig and Reynolds, 1988), species

richness (N1) and Berger-Parker dominance (BP Index)

(May, 1975) of each plot with naturalized species in

protected and recreation areas were calculated by SDR

(Species Diversity and Richness 4.1.2, 2006). A total of

230 plots were included, and obtained field data were

separated into two sets: one set for preserves; the other for

recreation areas, for further analysis. Multivariate Analysis

(MANOVA) was performed, and P<0.05 was used to

220

Botanical Studies, Vol. 50, 2009

determine the significance in all tests. Factor analysis (FA)

was also utilized to visualize and group the patterns of the

indices. Regression tree analysis was employed to access

the most significant variables separating patterns of plant

invasions in protected and recreation areas. Proportion of

reduction error (PRE) value, similar to R

2

in regression,

was used to determine the amount of variation explained

by the independent variable (Hansen et al., 1996). Factor

Analysis (FA), Multivariate Analysis (MANOVA), and

regression tree analysis performed by SYSTAT (version

11.0, Systat software Inc., Richmond, CA, USA. 2004).

RESULTS

The biological composition and biodiversity indices

differed between the preserves and recreation areas (Table

1). About 40% of the plots in the preserves did not have

any naturalized species, while around 10% of the plots in

the recreation areas were free of naturalized species. Five

biodiversity indices of these two areas were significantly

different, and two environmental factors, such as elevation

and naturalness degree, also showed significant difference

for the preserves and recreation areas (Table 1). All of the

Table 1. Numerical summary of plots, species composition, life forms, coverage (%), and £\ diversity indices of protected

and recreation areas. Summary statistics (mean with standard error in parentheses) by species number, coverage (%), £\

diversity indices, and environmental factors are included.

Protected areas

Recreation areas

Plot number

107

163

Plot with naturalized plant

83

147

Plot without naturalized plant

24

16

Family number

10

12

Genus number

16

25

Species number

21

32

Vine

0

1

Herbaceous

19

29

Shrubs

2

2

Trees

0

0

Dominant species

Bromus catharticus

Bromus catharticus

Trifolium repens

Trifolium repens

Dominant family

Asteraceae

Asteraceae

Dominant genus

Bidens

Bidens

Species number/plot

1.55 (0.11)

2.43 (0.09)****

Coverage sum (%)

1556.5 (15.565 m

2

)

5366 (53.66 m

2

)****

Maximum cover (%)

15.92 (2.80)

27.95 (2.10)****

Mean cover/plot (%)

18.76 (3.38)

36.50 (2.54)

Shannon-Weiner Index (H¡¦)

0.25 (0.04)

0.53 (0.03)****

Simpson Index (£f)

1.40 (0.09)

1.80 (0.06)***

Evenness (E5)

0.95 (0.03)

0.81 (0.02)****

Richness (N1)

1.37 (0.07)

1.84 (0.06)****

Berger-Parker dominance Index

0.87 (0.02)

0.76 (0.02)****

Elevation (m)

2148.22 (30.00)

1848.72 (22.54)****

Naturalness degree

3.02 (0.14)

4.39 (0.11)****

MANOVA: n = 270. Means with asterisks are significant from each other of the value of protected and recreation areas.

***: P<0.001; ****: P < 0.0001.

WU et al. ¡X Plant invasions in the areas of Shei-Pa National Park in Taiwan

221

the patterns of plant invasions of preserves and recreation

areas in (PRE = 0.72) (Figure 3B).

DISCUSSION

Plant invasions seem to be moderate in the Wuling

district. Compared to the 1,135 native species in this

region (Hsu, 1984), naturalized species only represented a

very small portion, approximate 4% of the total flora, half

of that of the whole island (about 8% of the Taiwan flora

is naturalized (Wu et al., 2004a). The difference might be

the results of geographical isolation or the climates of this

district. Situated in the central ridge of Taiwan, the only

one county road leading to this region is usually damaged

by frequent typhoons in summer. Besides, nearest city

connected by this only one wavy road is in about three-

hour driving distance. On the other hand, snows falling

on the mountains and harsh freezing winter in this region

might screen out the majority of naturalized species, which

dominant in the lowlands, from the tropics. Although the

proportion presented by naturalized species was small, it

naturalized species were herbaceous and shrub in both

areas; no naturalized trees were found (Table 1).

In terms of naturalized flora, the recreation areas tended

to harbor more families, genera, and species. However,

the dominant species Trifolium repen L. and Bromus

catharticus, followed by Conyza sumatranesis, and Bidens

pilosa var. pilosa were the same for these two types of

habitats (Table 2). The dominant family was Asteraceae

followed by Poaceae and Fabaceae for both areas (Figure

2A). Most of the naturalized species in both preserves and

recreation areas originating from tropical and temperate

regions, especially the Americas and Europe (Figure 2B).

The first two factors of factor analysis explained about

71% of the total variance (Figure 3A). Indices related to

the numbers of naturalized species, such as H¡¦ and £f, were

on the positive end of the first factor, while BP index was

on the opposite end. For factor two, elevation and total

coverage of the naturalized species were on the positive

end while naturalness degree and evenness (E5) were

on the other side. Regression tree analysis showed that

elevation was the most significant variable distinguishing

Table 2. Top ten dominant species of protected and recreation areas. Plant taxa are in the order of IVI values from high to

low. The family species belongs is indicated in parentheses. Both of the species with the highest IVI and relative cover (%)

are considered as dominant species with bolded numbers of IVI and relative cover (%).

Species (Family)

IVI

Relative cover (%)

Recreation areas

Bromus catharticus Vahl. (Poaceae)

23.42

12.26

Trifolium repens L. (Fabaceae)

20.90

14.92

Conyza sumatrensis (Retz.) Walker (Asteraceae)

10.72

4.39

Bidens pilosa L. var. radiata Sch. (Asteraceae)

9.41

4.11

Trifolium pratense L. (Fabaceae)

3.06

2.03

Veronica persica Poir. (Scrophulariaceae)

2.98

2.41

Crassocephalum crepidioides (Benth.) S. Moore (Asteraceae)

1.83

0.80

Lolium multiflorum Lam. (Poaceae)

1.77

1.08

Momordica charantia L. var. abbreviata Ser. (Cucurbitaceae)

1.75

1.41

Lepidium virginicum L. (Brassicaceae)

1.61

1.15

Protected areas

Bromus catharticus Vahl. (Poaceae)

11.43

5.87

Trifolium repens L. (Fabaceae)

10.43

8.78

Lolium multiflorum Lam. (Poaceae)

6.75

4.05

Conyza sumatrensis (Retz.) Walker (Asteraceae)

3.25

1.15

Bidens pilosa L. var. minor (Blume) Sherff (Asteraceae)

2.27

0.92

Bidens pilosa L. var. radiata Sch. (Asteraceae)

2.23

1.03

Crassocephalum crepidioides (Benth.) S. Moore (Asteraceae)

1.67

0.32

Trifolium dubium Sibth. (Fabaceae)

1.33

0.73

Ageratum conyzoides L. (Asteraceae)

1.11

0.81

Geranium carolinianum L. (Geraniaceae)

0.64

0.19

222

Botanical Studies, Vol. 50, 2009

is very difficult to testify whether Taiwan is more resistant

to plant invasion as Wu et al. assumed without supporting

field data (Wu et al., 2004b; Daehler, 2006).

The same dominant family, genera, and species shared

by preserves and recreation areas may be explained by

neighboring and stepping stone effects. The number of

naturalized species was significantly less in the preserves

than in the recreation areas (Table 1), nevertheless, all

the naturalized species present in the preserves were in

the flora of recreation areas. This, furthermore, supported

our idea that the naturalized flora of the protected areas is

a fraction of the naturalized flora in the recreation areas.

Besides, this result also implied that naturalized species

may come from the neighboring recreation areas where

tourists stop by before visiting the preserves. These shared

species should have broader niche breadth (Richards et al.,

2006).

The dominant species, Trifolium repens, and Bromus

catharticus Vahl (Table 2), of both areas might find the

temperate habitats in the central mountains of Taiwan

similar to their home ranges in Europe and temperate

South America (Richardson and Thuiller, 2007). Although

the nitrogen fixing function of T. repens may alter soil

nutrient gradients (Warren, 2000; Evans et al., 2001), its

impacts were restricted since it was currently only present

in places located in relatively developed areas. On the

contrary, B. catharticus, the only species surviving in

alpine vegetations up to 3,800 meters in elevation, has

Figure 2. A, Dom inant families of the naturalized flora in the protected areas and recreation areas; B, Origins of naturalized

species in the protected (P) and recreation areas (R).

Figure 3. A, Factor a nalysis for numbe r of naturalized species, total cover of naturalized species, the biodiversity indices,

elevation, and naturalness degree of the protected and recreation areas. Percent of total variance explained is 42% for factor (1),

29% for factor (2), and 15% for factor (3); B, Regression tree for number of naturalized species, total cover of nat uralized spe-

cies, the biodiversity indices, elevation, and nat uralness degree across the a rea types (protected and recreation areas), includ-

ing means, standard deviation (SD) and sample size (n) for each break. PR a reas indicated protected and recreation areas. PRE

(0.72) indicates the proportion of reduction in error.

WU et al. ¡X Plant invasions in the areas of Shei-Pa National Park in Taiwan

223

adapted very well to harsh montane environments. Bromus

catharticus has been shown to possess various advantages

in growing and population maintaining, such as high

phenotypic plasticity to various environments, large seed

production, seed dormancy, low temperate tolerance,

and high genetic variation (Mauromicale and Cavallaro,

1996; Wolff et al., 1996; Pistorale et al., 1999; Wolff et al.,

2001; Aulicino and Arturi, 2002). Beside the respectful

distributions in alpine habitats of B. catharticus, it often

formed monoculture and occupies most of the abandoned

fields in the recreation areas as well. The distribution

o f B. catharticus in the preserves was, nevertheless,

scattered and not continuous. More specifically, it

appeared at almost every stop and cabin on the hiking

trails. However, not a single individual was found in the

short bamboo (Yushania niitakayamensis) grasslands. We

suggested that the bamboo grasslands may function as a

barrier (Perelman et al., 2003) for plant invasions since

they did not accommodate any other naturalized species,

either. Although the distribution and dominance of B.

catharticus seemed to be threatening, we know nothing

about this species unfortunately. It is very worrisome that

B. catharticus may continue to spread and alter the alpine

ecosystems in the future, because their flexibility and

plasticity may be induced through interactions between

local alpine environments in terms of adaptation (Mooney

and Cleland, 2001). Further studies regarding its impacts

and associations with native species are immediately

needed. Furthermore, the possible defense mechanism of

Y. niitakayamensis may generate better understanding of

biological resistance (Chou and Yang, 1982; Eyini et al.,

1989).

Compared to the naturalized flora of Taiwan, Wuling

accommodated relatively more temperate species, although

species originating from the tropics were still presented

at certain proportions (Figure 2B). This supported our

expectation that the mountain areas in Taiwan may harbor

relatively more temperate species due to similar climates

(Pino et al., 2005). Tropical species, such as Conyza

sumatrensis and Bidens pilosa L. var. minor (Blume)

Sherff. still deserve further attentions, since they also

thrived in both preserves and recreation areas. Although

warm and humid weather in summer in this area may

provide suitable habitats, freezing temperate and relatively

harsh environments in winter still resulted in surviving

pressures to tropical species. Adaptation mechanisms of

these species deserve more attentions (Parker et al., 2003).

The composition of species and £\ diversity indices

supported our hypothesis that preserves and recreation

areas, two land management types, presented different

patterns of plant invasions (Table 1). It is not surprising

to observe more naturalized species and total cover

(SUM) of naturalized species in the recreation areas. In

addition, the differences of biodiversity indices strongly

suggested two different patterns of plant invasions in these

two areas. Significantly higher H¡¦, and £f of recreation

areas implied the composition of naturalized species was

more diversified, and species coexistence was common

according to E5 (Table 1). On the contrary, significantly

lower evenness and higher BP index of the preserves

revealed that the composition of naturalized species

was relatively uneven and dominated by few species in

the preserves. We suspected that different land uses and

native species community were relevant to these patterns.

Abandoned crop fields and plantations were the major

habitats for native and naturalized species in the recreation

areas, and chances for both species were assumed equal

when the fields were available. Therefore, relatively

even species composition could be a result of stochastic

effects. On the contrary, relatively undisturbed native plant

community formed by short bamboo and alpine species

left scattered openings for naturalized species to stand.

Only few species with formidable invasiveness could

possibly establish colonies after interacting with local

species and environments.

The environmental factors changing along the elevation

might be the causes of different invasion patterns in the

preserves and recreation areas (Figures 3A and B). The

result of factor analysis implied that elevation was the

factor correlated with total cover, since elevation was

categorized in the same factor with total cover of the

naturalized species, while the first factor of the factor

analysis was basically comprised of all the variables

relevant to species number. This pattern, furthermore, was

confirmed by the regression tree analysis (PRE = 0.72).

However, the naturalness degree, which we employed

to indicate the land cover types and probable land use

intensity, showed a reversed pattern (Table 1). It did

not seem to be reasonable that the naturalness degree

of the preserves was significantly lower than that of

the recreation areas, but we suggested that was a result

of certain areas of bare ground of gravels in the alpine

zone, which was categorized as zero. Although species

number together with a few more biodiversity indices

were the major factors that separated the plant invasion

patterns of the preserves and recreation areas. The most

important factor resulting in these different invasion

patterns was elevation (Baret et al., 2004; Becker et

al., 2005; McDougall et al., 2005). Different elevations

separated almost every subsets of the regression tree on

the recreation area side, and we were clear that the higher

the elevation, the fewer the naturalized species. This result

had nothing to do with disturbance levels alone, although

the land management and anthropogenic activities were

definitely different in the recreation areas and preserves.

However, it pointed out that the factors changing along the

elevation (similar to the latitude effect), such as climate,

dominant native species, dominant vegetation type,

temperature, disturbance, may be important in terms of

resistance (Py.ek, 1998; Cantero et al., 2003; Arevalo et

al., 2005; Gelbard and Harrison, 2005; Richardson et al.,

2005).

Our results basically supported all of our hypotheses,

although elevation, instead of nature degree, was the factor

224

Botanical Studies, Vol. 50, 2009

contributing to different plant invasion patterns of plant

invasions in protected and recreation areas. These different

patterns of plant invasions provided valuable information

of mountains in subtropical regions, where studies on

plant invasions are seriously rare. However, dominance

of Bromus catharticus in the preserves, especially in the

montane ecosystems, definitely requires further ecological

studies and deliberate attentions, since these habitats are

home to 25% endemic species and several species which

survived the last ice age (Hsieh et al., 1997). Furthermore,

we suggest that monitoring systems should be established

at the entrances of preserves, since few species have

managed to infect the protected areas from adjacent

recreation areas.

Acknowledgements. This study was sponsored by Shei-

Pa National Park, Construction and Planning Agency,

Ministry of the Interior, in 2006. Many thanks to park

rangers and staffs for technical support and assistance.

Furthermore, we thank Drs. Chang-Fu Hsieh and Chi-Jung

Chiu for technical supports; Wei-Chun Cheng, Yu-Ting

Chung, Chia-Long Huang, Chwen-Ling Kuo, Pei-Chun

Liao, Cheng-Fan Yang, and Sheng-Jye Wu for help with

fieldworks, and Hsuan-Jung Su for manuscript revision.

Moreover, we are very grateful for the valuable comments

and suggestions provided by two deliberate anonymous

reviewers.

LITERATURE CITED

Alston, K.P. and D.M. Richardson. 2006. The roles of habitat

feature, disturbance, and distance from putative source

populations in structuring alien plant invasions at the urban/

wildland interface on the Cape Peninsula, S outh Africa.

Biol. Conserv. 132: 183-198.

Arevalo, J.R., J.D. Delgado, R. Otto, A. Naranjo and J.M. Sa-

las M and Fernandez-Palacios. 2005. Distribution of alien

vs. native plant species in roadside communities along an

altitudinal gradient in Tenerife and Gran Canaria (Canary

Islands). Perspect. Plant Ecol. Evol. Syst. 7: 185-202.

Aulicino, M.B. and M.J. Arturi. 2002. Phenotypic diversity in

Argentinian populations of Bromus catharticus (Poaceae).

Genetic and environmental com ponents of quantitative

traits. N. Z. J. Bot. 40: 223-234.

Baret, S., S . Maurice, T. Bourgeois and D. Stras berg. 2004.

Altitudinal variation in fertility and vegetative growth in

the invasive plant Rubus alceifolius Poiret (Rosaceae). On

Reunion island. Plant Ecol. 172: 265¡V273.

Bec ker, T., H. Di etz , R. Bi llet er, H. Busc hma nn, a nd P.J .

E dwa rds . 2005. Alti tudinal dis tri but ion of a lien plant

s p ec ies i n th e S wi ss Alps . P e rs pec t. P l ant Ec ol ogy,

Evolution and Systematics 7: 173¡V183.

Byers, J.E. and E.G. Noonburg. 2003. Scale dependent effects

of biotic resistance to biological invasion. Ecology 84:

1428-1433.

Cantero, J.J., J. Liira, J.M. Cisnero, J. Gonzalez, C. Nunez, L.

Petryna, C. Cholaky, and M. Zobel. 2003. Species richness,

alien species and plant traits in Central Argentine mountain

grasslands. J. Veget. Sci. 14: 129-136.

Chou, C.H. and C.M. Ya ng. 1982. Allelopathic research of

subtropical vegetation in Taiwan. II. Comparative exclusion

of understory by Phyllos tachys edulis and Cryptomeria

japonica. J. Chem. Ecol. 8: 1489-1508.

D¡¦Antonio, C.M. and T.L. Douley. 1995. Biological Invasions

as agents of change on the islands vers us mainland. In

P.M. Vitousek, L.L. Loope and H. Adersen (eds.), Island:

Biological Diversity and Ecosystem Function. Vol. 115.

Springer Inc, New York, pp. 103-121.

Daehler, C.C. 2006. Invasibility of tropical islands by introduced

plants: portioning the influence of isolation and propagule

pressure. Presila 78: 389-404.

Evans, R.D., R. Rimer, L. Sperry, and J. Belnap. 2001. Exotic

plant invasion alters nitrogen dynamics in an arid grassland.

Ecol. Appl. 11: 1301-1310.

Eyini, M., M. Jayakumar, and S. Pannirselvam. 1989.

Allelopathic effect of bamboo leaf extract on the seedling of

groundnut. Trop. Ecol. 30: 138-141.

Foxcroft, L.C., M. Rouget, and D.M. Richardson. 2007. Risk

assessment of riparian plant invasions into protected areas.

Conserv. Biol. 21: 412-421.

Foxcroft, L.C., D. M. Richardson, and J .R.U. Wils on. 2008.

Ornam enta l P lants as Invas ive Alie ns: Prob lem s a nd

Solutions in Kruger National Park, South Africa. Environ.

Manage. 41: 32-51.

Gelbard, J .L. and S. Harrison. 2005. Invasibility of roadless

grasslands: An experimental study of yellow starthistle.

Ecol. Appl. 15: 1570-1580.

Harrison, S., C. Hohn, and S. Ratay. 2002. Distribution of exotic

plants along roads in a peninsular nature res erve. Biol.

Invasions 4: 425-430.

Hendrickson, C., T. Bell, K. Butler, and L. Hermanutz. 2005.

Disturbance-enabled invasion of Tuss ilago farfara L. in

Gros Morne National Park, Newfoundland: management

implications. Natural Areas J. 25: 263-274.

Hsieh, C.F., W.C. Chao, C.C. Liao, K.C. Yang and T.H. Hsieh.

1997. Floristic composition of the evergreen broadleaved

forests in Taiwan. Natural History Research Special Issue 4:

1-16.

Hsu, K.S. 1994. Rare and endemic plants of Shei-Pa National

Park. Research report, Shei-Pa National Park, Construction

and Planning Agency, Ministry of the Interior, Taiwan, (In

Chinese)

Huang, T.C., J.T. Wu and C.F. Hsieh. 1999. S uggestions for

evaluation criteria and report forma t of environmental

impact assessment. Department of Botany, National Taiwan

University, Taipei, Taiwan, 68 pp. (In Chinese)

Jesson, L., D. Kelly and A. Sparrow. 2000. The importance of

dis persal, disturbance, and competition for exotic plant

invasions in Arthur¡¦s Pass National Park, New Zealand. N.

Z. J. Bot. 38: 451-468.

WU et al. ¡X Plant invasions in the areas of Shei-Pa National Park in Taiwan

225

Kang, M., J .J.H. Ciborows ki, and L.B. J ohnson. 2007. The

influence of anthropogenic disturbance and environmental

s ui tab ili ty o n th e di st rib uti on of t he n onin dige nous

Amphipod, Echinogammarus ischnus, at Laurentian Great

Lakes Coastal Margins. J. Great Lakes Res. 33: 198-210.

Larson, D.L., P.J. Anderson, and W. Newton. 2001. Alien plant

invasion in mixed-grass prairie: Effects of vegetation type

and anthropogenic disturbance. Ecol. Appl. 11: 128-141.

Levine, J.M. 2000. S pecies diversity and biological invasions:

relating local process to community pattern. Science 288:

852-854.

Loope, L.L. and D. Mueller-Dombois. 1989. Characteristics of

invaded islands, with special reference to Hawaii. In J.A.,

H.A. Mooney, F. DiCastri, R.H. Groves, F.J. Kruger, M.

Rejmanek and M. Williamson (eds.), Biological Invasions:

A Global Perspective. Scope 37. John Wiley and Sons, New

York, 550 pp.

Lu, Z. and M. Keping. 2006. Spread of the exotic croftonweed

(Eupatorium adenophorum) across southwest China along

roads and streams. Weed Sci. 54: 1068-1072.

Ludwig, J.A. and J.F. Reynolds . 1988. Statistical ecology: a

primer on methods and computing. Wiley-Inters cience,

New York, 368 pp.

Ma uromic ale , G. and V. Ca valla ro. 1 996. E ffe cts of se ed

osmopriming on germination of three herbage grasses at

low temperatures. Seed Sci. Technol. 24: 331-338.

May, R.M . 1975. Patterns of species abundance and diversity.

In M.L. Cody and J .M. Diam ond (eds.), Ec ology and

Evolution of Com munities . Harvard Universi ty Press ,

Cambridge, pp. 81-120.

McDougall, K.L., J.W. Morgan, N.G. Walsh, and R.J. Williams.

2005. Plant invasions in treeless vegetation of the Australian

Alps. Perspective Plant Ecol. Evol. Syst. 7: 159-171.

McKinney, M.L. 2002. Influence of settlement time, human

population, park shape and age, visitation and roads on the

number of alien plant species in protected areas in the USA.

Divers. Distrib. 8: 311-318.

Moon ey, H.A. and E.E . Cle land. 2001. T he evol utiona ry

impact of invasive species. Proc. Natl. Acad. Sci. USA 98:

5446-5451.

Ou, C.H., K.C. Lu, and Y.H. Tseng. 2006. Study on classification

and spatial patterns of Shei-Pa National Park. I. Research

report, S hei-Pa National Park, Construction and Planning

Agency, Ministry of the Interior, Taiwan. (In Chinese)

Paiaro, V., A. Mangeaud, and

E. Pucheta. 2007. Alien seedling

recruitment as a response to altitude and soil disturbance in

the mountain grasslands of central Argentina. Plant Ecol.

193: 279-291.

Parker, I.M., J. Rodriguez, and M.E. Loik. 2003. An evolutionary

approach to understanding the biology of invasions: local

adaptation and general-purpos e genotypes in the weed

Verbascum thapsus. Conserv. Biol. 17: 59-72.

P auc hard, A., P.B. Alabac k and E.G. Edlund. 2003. Pl ant

invas ions in protected areas at multiple s cales: L inaria

vulgaris (Scrophulariaceae) in the west Yellowstone area.

West. North Am. Naturalist 63: 416-428.

Perelman, S.B., S.E. Burkart, and R.J.C. Leon. 2003. The role

of a native tussock grass (Paspalum quadrifarium Lam.)

in structuring plant communities in the flooding Pampa

grasslands. Biodiversity Conserv. 12: 225-238.

Pino, J ., X. F ont, J. Carbo, M. Jove, and L. Pallares . 2005.

Large-scale correlates of alien plant invasion in Catalonia

(NE of Spain). Biodiversity Conserv. 122: 339-350.

Pistorale, S., R. Wolff, and O. Bazzigalupi. 1999. Dormancy

and seed germ ination in natural populations of Bromus

catharticus Vahl (Cebadilla criolla). J. Genet. Breed. 53:

47-55.

Py.ek, P. 1998. Alien and native species in Central European

urban floras: a quantitative comparison. J. Biogeogr. 25:

155-163.

Py. ek, P., D.M. Richa rds on, M. Rej manek, G.L. Webs ter,

M. Williamson, and J . Kirs chner. 2004. Alien plants in

c hec klis ts and flo ra s : towards bet ter co mm unica tion

between taxonomists and ecologists. Taxon 53: 131-143.

Rejmanek, M. 1996. Species richness and resistance to

invasions. In G.H. Orians, R. Dirzo, J.H. Cushman (eds.),

Biodiversity and Ecosystem Processes in Tropical Forests,

Vol. 122. Springer, New York, pp. 153-172.

Richards, C.L., O. Bossdorf, N.Z. Muth, J. Gurevitch, and M.

Pigliucci. 2006. Jack of all trades, master of some. On the

role of phenotypic plasticity in plant invasions. Ecol. Lett. 9:

981-993.

Richardson, D.M. and W. Thuiller. 2007. Home away from home

- Objective mapping of high-risk s ource areas for plant

introductions. Divers. Distrib. 13: 299-312.

Richardson, D.M., M. Rouget, S.J. Ralston, R.M. Crowling, B.J.

Van Rensburg, and W. Thuiller. 2005. Species richness of

alien plants in South Africa: environmental correlates and

the relations hip with indigenous plant s pecies richness.

Ecoscience 12: 391-402.

Rogers, G.M. and J.R. Leathwick. 1996. North island seral

tussock grasslands. 3. The influence of heather (Calluna

vulgaris) on rates of change from tus sock grass land to

shrubland. N. Z. J. Bot. 1996: 473-487.

Rose, M. and L. He rm anutz. 2004. Are boreal ecos ystem s

susceptible to alien plant invasion. Evidence from protected

areas. Oecologia 139: 467-477.

Ross, K.A., B.J. Fox, and M.D. F ox. 2002. Changes to plant

s pe ci es ri ch nes s i n fore s t fra gm en ts : fr agm en t a ge,

disturbance and fire history may be as important as area. J.

Biogeogr. 29: 749-765.

Rowlands, P.G. and N.J. Brian. 2001. Fishtail Mesa: A vegetation

resurvey of a relict area in Grand Canyon National Park,

Arizona. West. North Am. Naturalist 61: 159-181.

Schwarz, A.G. and R.W. Wein. 1997. Threatened dry grasslands

in the continental boreal forests of Wood Buffalo National

Park. Can. J. Bot. 75: 1363-1370.

Simberloff, D. 1995. Why do introduced species appear to

226

Botanical Studies, Vol. 50, 2009

devastate islands more than mainland areas. Pac. Sci. 49:

87-97.

Theoharides, K.A. and J.S. Dukes. 2007. Plant invasion across

s pace and time: factors affecting nonindigenous species

success during four stages of invasion. New Phytol. 176:

256-273.

Thuiller, W., D.M. Richardson, M. Rouget, S . P roches, and

J .R.U. Wilson. 2006. Interactions between environment,

s pecies traits, and human uses describe patterns of plant

invasions. Ecology 87: 1755-1769.

Tseng, C.S. 1999. Studies on population ecology of the

F ormosan L and-locked S almon, Oncorhynchus masou

formosanus. II. Research report, Shei-Pa National Park,

Construction and Planning Agency, Ministry of the Interior,

Taiwan. (In Chinese)

Vitousek, P.M., C.M. D¡¦Antonio, L.L. Loope, M. Rejmanek,

and R. Westbrooks. 1997. Introduced species: A significant

components of human-caused global change. N. Z. J. Ecol.

21: 1-16.

Wang, C.M. 1995. The ec ologi cal environme nt of Wuling

d i s t ri c t. R e s ea r c h r e po rt , S he i -P a Na t io na l P ar k,

Construction and Planning Agency, Ministry of the Interior,

Taiwan. (In Chinese)

Wa rren, J .M. 2000. the role of white c lover in the los s of

diversity in grassland habitat restoration. Restor. Ecol. 8:

318-323.

Wein, R.W., G. Wein, S. Bahret, and W.J. Cody. 1992.

Northward invasding non-native vascular plant species in

and adjacent to Wood Buffalo National Park, Canada. Can.

Field-Nat. 106: 216-224.

Wilson, J.R.U., D.M. Richardson, M. Rouget, .. Proche., M.A.

Amis, L. Henderson, and W. Thuiller. 2007. Residence time

and potential range: crucial considerations in modelling

plant invasions. Divers. Distrib. 13: 11-22.

Wolf, J.J., S.W. Beatty, and T.R. Seastedt. 2004. Soil

characteristics of Rocky Mountain National Park grasslands

invasded by Melilotus officinalis and M. alba. J. Biogeogr.

31: 415-424.

Wolff, R., L. Abbott, and S. Pi stora le. 1996. Re produc tive

behaviour of Bromus catharticus Vahl (Cebadilla criolla)

in natural and cultivated populations. J. Genet. Breed. 50:

121-128.

Wolff, R., L. Abbott, and S. Pi stora le. 2001. Reproduc tive

strategy of Bromus catharticus Vahl (Cebadilla criolla):

Phenotypic plasticity in natural population progenies. J.

Genet. Breed. 55: 67-74.

Wu, S.H., C.F. Hsieh, S.M. Chaw, and M. Rejmanek. 2004a.

Plant invasions in Taiwan: Insights from the flora of casual

and naturalized alien species. Divers. Distrib. 10: 349-362.

Wu, S.H., C.F. Hsieh, and M. Rejmanek. 2004b. Catalogue of

the naturalized flora of Taiwan. Taiwania 49: 16-31.

Wuerthner, G. 1996. Alien invasion. National Parks 70: 32-35.

WU et al. ¡X Plant invasions in the areas of Shei-Pa National Park in Taiwan

227