Botanical Studies (2011) 52: 503-516.
ECOLOGY
Diversity of the alpine vegetation in central Taiwan is affected by climate change based on a century of floristic inventories
Chang-Hung CHOU1*, Tsurng-Juhn HUANG1, Yen-Ping LEE2, Chi-Yuan CHEN3, Tsai-Wen HSU4, and Chih-Hui CHEN4
1Research Center for Biodiversity and Graduate Institute of Ecology and Evolutionary Biology, China Medical University, Taichung, Taiwan
2Dongshih Forest District Office, Forestry Bureau, Council of Agriculture, Executive Yuan, Taichung, Taiwan
3 Graduate Institute of Bioresources, National Pingtung University of Science and Technology, Pingtung, Taiwan
4Taiwan Endemic Species Research Institute, Chi-chi, Nantao, Taiwan
(Received December 2, 2010; Accepted May 16, 2011)
ABSTRACT. Floristically, Taiwan is a very rich island due to her varied geography, topography and habitats. Through extended historical research involving past and present botanical inventories of the central mountains, particularly of the Hehuanshan area of Taiwan, we are now able to examine the floristic composition of four elevation zones, 2,000-2,500 m, 2,500-3,000 m, 3,000-3,500 m, and 3,500-3,950 m. We selected four study sites, namely Shinjenkan (SJK) at 2,250-2,585 m, Shihmenshan (SMS) at 3,000 m, Hehuan East Peak (HEP) at 3,401 m, and Hehuan Major Peak (HMP) at 3,408 m, and determined their a, p, and y diversities along with their Simpson's diversity indices. Our results clearly showed that the species richness (a diversity) was significantly high and decreased as the altitude increased. Coincidently, Simpson's diversity index at 2,250 m was significantly high at 0.85, drastically decreased to 0.17 at 3,145 m, and continued to decrease to 0.10 at both sites at 3408 m. On the other hand, by comparing plant distribution data collected over a century for the Hehuanshan, Alishan and Yushan areas, we were able to predict that plants would migrate mostly from a lower to a higher elevation when the global temperature increases. For instance, in the Hehuanshan area, 16 species would move towards higher elevations and seven species would remain in their original zone. In the Alishan area, seven species would migrate and four would remain in their original zone, and in Yushan, 15 species would migrate out of their zone and five would remain. Of all of the species, at least six risk extinc­tion, since their expected migration would take them far beyond the limits of the land available above 3,950 m. It is concluded that the alpine vegetation will be redistributed, many plant species will move towards a higher elevation and, eventually, at least six plant species (Anaphalis morrisonicola, Artemisia morrisonensis, Sweriia randaiensis, Hypericum nagasawai, Angelica morrisonicola, and Cirsium arisanensis) will become extinct.
Keywords: Alishan; Alpine vegetation; a, p, y diversity; Global warming; Hehuanshan; Plant migration; Simpson's diversity index; Yushan.
INTRODUCTION
Botanical inventories of the entire island, but particu­larly of the mountainous areas, were initiated by Japanese botanists in 1895. Pioneer Japanese botanists, namely Hayata, Kudo, Sasaki, Masamune, Yamamoto, Sato, etc. (Huang, 1993), paid great attention to mountainous regions such as Alishan, Yushan and Hehuanshan, all of which are above 2,000 m. A great number of specimens were col­lected and deposited in the herbaria of Tokyo University, Kyoto University (Chen, 1995) and Taihoku Empire Uni­versity (the present National Taiwan University) before World War II. After the war, Taiwanese botanists continu­ed the inventory. Taiwan is located in the subtropical and tropical region with great topographic variation ranging from sea level to 3,950 m above sea level (asl), and with more than 200 mountain peaks above 3,000 m. Due to the
Chen (1995) proposed that Taiwan has been affected by at least four ice ages, and that these have caused sig­nificant changes to the island's vegetation. Past PAGES (Past Global Changes, IGBP) studies on the effect of cli­mate change on vegetation in natural forests in Taiwan, including pollen analyses and geological variation (Tsu-kada, 1966; Su, 1984a, 1984b, 1985; Chen, 1995), have all suggested that climate change will lead to vegetation succession.
*Corresponding author: E-mail: choumasa@mail.cmu.edu. tw.
504
Botanical Studies, Vol. 52, 2011
variable climate and habitats, Taiwan possesses great plant diversity. Of the more than 4000 vascular plant species registered, more than 25% are endemic. The large amount of data accumulated over the past 100 years concerning the flora and climatic conditions in Taiwan now make the study of floristic changes related to climatic conditions possible.
vulnerability of tropical mountain endemic species due to climatic change, and concluded that endemic species in Madagascar distributed close to summits may be especial­ly vulnerable to extinction due to their upward movement. Few studies have addressed how climate change, and in particular changes in temperature and precipitation, affect the diversity of vegetation in mountain areas (Beerling and Woodward, 1994). We therefore attempt herein to assess the correlation between climate change and plant diver­sity in Taiwan based on one century of weather data and botanical records. However, it is rather difficult to address the subject in the lower elevations in Taiwan due to the many socio-economic factors affecting the natural vegeta­tion. Therefore, our analyses concentrate on remote areas, particularly in the central mountains above a 2,000-m ele­vation, such as Alishan, Hehuanshan and Yushan, with a special emphasis on the Hehuanshan region.
We used the following indices of diversity to understand changes in plant communities: a (alpha) diversity, p (beta) diversity and y (gamma) diversity. Alpha diversity refers to the number of species in a certain community; beta diver­sity represents the variability of species composition along an environmental or geographical gradient; and gamma diversity applies to larger geographical scales, referring to the number of species in a large region (Primack, 1998). These concepts are useful when studying biological diver­sity in relation to conservation biology in a community or region, such as in natural reserves or national parks. Com­bining botanical inventories and climatic information, we can use diversity indices to understand vegetation change relative to climatic or topographic changes.
MATERIALS AND METHODS
Study sites
Several recent international conferences on biodiversity informatics and the impact of climate change on life have taken place in different parts of the world (Dan BIF con­ference, 2008; ISGMB, 2008; ISGBHHW, 2009). Many
important questions were asked concerning the extent to which global warming will affect Earth's biodiversity, including the natural systems that sustain human societ­ies. It is very difficult to apply traditional experimental approaches to investigate the large-scale and long-term na­ture of the issues without involving biodiversity data and employing statistical analysis. Data from the Global Bio­diversity Information Facility, GBIF, have made it possible to assess the change in biodiversity as affected by climate change. It is beyond doubt that the substantial floristic documents covering a century of exploration from several data banks in Taiwan and Japan, as well as the climatic data from the Central Weather Bureau of Taiwan, which has been registered since the 1900s, allow us to investigate the aforementioned questions. In addition, the senior au­thor initiated a national thematic research program to ad­dress these issues in 2006 (2006-2009).
Taiwan is located to the southeast of China (21°53'50''-25°18'20'' N; 120。01'00"-121。59'15" E), and with the tropic of Cancer passing through Chia-Yi in southern Tai­wan; it is both subtropical and tropical. The topography of Taiwan ranges from sea level to 3,950 m above sea level, and has more than 200 peaks reaching above 3,000 m. The average temperature is 28°C in summer and 14°C in winter. The average annual precipitation is 2,515 mm, but the rain is often concentrated in the few months from July to September. A dry period from November to April is par­ticularly pronounced in southern Taiwan.
It is impossible to survey the floristic diversity of the entire alpine region; thus, we selected the Hehuanshan area (Nantou County) for detailed study. Four sites were selected along highway 14A at elevations of 2,550-3,500 m above sea level:
1. Shinjenkan (SJK) is located at 21.5 K ~ 22 K, Highway 14A, situated at 121。12'52.8" East and 24。06'46.5" North. The sampling site for shrubs is located at 2,585 m asl (above sea level) at 121°12'53" East and 24。06'49.3" North.
In fact, there has been an increasing awareness and evi­dence of global warming over the last century (Prabhakar and Shaw, 2008; CCSC, 2009), which is leading to climate change and temperature increases in the range of 0.6-1.5°C in many parts of the world. Negative impacts of climate change on ecosystems have also been demonstrated in Taiwan (Chou, 2009). The Intergovernmental Panel on Climate Change (IPCC) has proposed different scenarios (A1, A2, B2, etc.) that predict the future change in global temperature. In Taiwan, we have observed and analyzed climatic changes since the 1900s using data obtained from the Central Weather Bureau of Taiwan. If global warm-ing occurs in Taiwan, it is hypothesized that some plant species might migrate to higher elevations. Pearson and Raxworthy (2008) presented a case study on the extreme
2. Shihmenshan (SMS) is located at 34 K, Highway 14A,
at 3,145 m asl, 121。17'03.5" East and 24。08'48.1"
North.
3. Hehuan East Peak (HEP) is located at 3,401 m asl, 121。16'54.0" East, 24。08'11.3" North.
4. Hehuan Major Peak (HMP) is located at 3,408 m asl,
121。16'17.7" East, 24。08'30.6" North.
Botanical inventory of alpine vegetation from data banks in Taiwan
Several botanical databases are available from the Biodiversity Research Center of Academia Sinica, Tai­pei (BRC/AS) (http://dbIn.Sinica.edu.tw/textdb/hast_labelquery.php), National Taiwan University (http://tai2.
CHOU et al. ―Diversity of alpine vegetation affected by climate change
505
ntu.tw/webtaiprog/web.query-aspx), and Taiwan Forestry Research Institute (http://taif.tfri.gov.tw:8080/spec-query.aspx). We surveyed plant species distributed in the He-huanshan area, arbitrarily divided into 4 elevation zones (2,000-2,500 m, 2,500-3,000 m, 3,000-3,500 m, and above 3,500 m) and listed the plant species, genera, and families found within each zone. The plant species nomenclature follows Flora of Taiwan (2nd edition, Huang et al., 2002); in addition, a list of rare and endangered species was ob-tained from the database of the Council of Agriculture, Taiwan (Chou et al., 2009, unpublished report).
The expected elevation of migrated species was repre­sented by M (mean elevation in meter) plus SD (standard deviation). For example, Acer serrulatum was documented at the first time in 1984 at 2,200 m asl and was later docu­mented in 1999 at 2,400 m asl. From the two data, we obtained a mean elevation of 2,300 m and a standard de­viation (SD) of 141.42 m. From the mean and SD, it was deduced that the plant might move upwards to an expected elevation of 2,441 m. Following this computing process, we were able to assess the migration of each plant to a cer­tain altitude.
Diversity measurement
We determined the a, p, and y diversity of vegetation and used Simpson's diversity index to make a comparison between sampling sites and time. The equation of Simp­son's diversity index was described by Simpson (1949).
Estimation of available land area in the Hehuan-shan area
At the study sites, plants were sampled based on the botanical survey of quadrates (10 x 10 m each) and on and transect sampling (20 m long) at an interval of 100 m along Highway 14A from an elevation of 2,250 to 2,750 m.
Using ArcView software and a digital geographic model (Digital Terrain Model, DTM), and through com­munication with Prof. Chen at the National Pingtung Uni­versity of Science and Technology, Taiwan, the land area of Hehuanshan above 2,000 m was estimated by dividing into five elevational zones (2,000-2,250 m, 2,250-2,500 m, 2,500-2,750 m, 2,750-3,000 m, and above 3,000 m). The estimated available land area is expressed as a percentage (Chou et al., 2009, unpublished report).
Examination of plant specimens in herbaria in Japan and Taiwan
From the databases mentioned above, we downloaded relevant information for each species found at our selected study sites. We consulted the herbarium of Tokyo Universi­ty to add any information from early documents and recent findings that had not been uploaded to the databases. For the 367 plant species that were recorded in the Hehuanshan area, we carefully checked all vouchers deposited in the herbaria and the contents of several databases in Taiwan. We obtained a list of species found at the Shinjenkan site in different years and at different elevations (Chou et al. 2009, unpublished report, or see Supplementary Table 1).
RESULT AND DISCUSSION
Botanical diversity in the alpine area of Taiwan
We attempted to study the botanical diversity of alpine plants based on a century of botanical inventories docu­mented in the databases of various Taiwanese institutions. Appropriately 25,000 specimens have been collected in our study area above an elevation of 2,000 m, including 132 families, 557 genera, and 1442 species (Table 1). In the elevation zone of 2,000-2,500 m, we recorded 1266 species, 527 genera, and 131 families; in the 2,500-3,000 zone we recorded 829 species; in the 3,000-3,500 m zone we recorded 561 species; and above 3,500 m we recorded 252 species (Huang et al., 2002). It is clear that the num­ber of plant species decreases with increasing elevation.
Checking climatic data from the last century in Taiwan
Since 1895, weather stations have been established in most areas in Taiwan. Weather data from the last century were obtained from the Central Weather Bureau of Taiwan; however, the data from the Hehuanshan are incomplete due to the lack of a weather station there at the beginning of the 20th century. Instead, we obtained data from nearby stations at Alishan and Yushan, which are relatively close to the Hehuanshan site. The average annual temperature and precipitation since 1930 are presented in Figure 1.
At the same time, the available land decreases with al­titude. For example, in the Hehuanshan area, the available land in elevation zone C (2,500-2,750 m) corresponds to 94% of the land area in elevation zone D (2,750-3,000 m), and the area in zone E (3,000-3,421 m) is equal to 30% of the area in elevation zone D (2,750-3,000 m) (Table 2). Logistically, under global warming in the region, plant species would migrate upwards to the higher elevations in zone E, where the available land for plant growth is only 30% of the area previously occupied (Table 2); in other words, about 70% of plant species might not survive due to limited space. However, most endemic species may not be seriously affected by global warming. In the Hehuan-shan area, there are 480 endemic species, about one-third of the total number of species found in the mountainous
Statistical analysis
All species are listed in an Office Excel 2007 table ac­cording to altitude, and the elevations of each species' dis­tribution vs. year of each species' recording were analyzed statistically using standard deviation (SD) as described by Gomez and Gomez (1976). An adjusted SD was obtained according to the formula which was used by Jump et al. (2006):
506
Botanical Studies, Vol. 52, 2011
Table 1. A summary of the number of plant families, genera, and species distributed in four elevation zones of the Central moun­tain area of Taiwan based on a century of botanical inventories. Data are synthesized from Flora of Taiwan (2002).
region. We concluded that some species, even though they can migrate to higher elevations, may become extinct anyway due to the limited space and to the environmental stress in these areas (Chen, 1995). In Taiwan, vegetation is almost absent at elevations above 3,900 m, which is similar to that documented for the island of Madagascar (Pearson and Raxworthy, 2008).
Table 2. Estimated available land in the five elevation zones in the Hehuanshan area*.
Botanical inventory at the Shinjenkan study site
The Shinjenkan study site, located at a 2,250-m eleva­tion along Highway 14A, is a unique ecotone for alpine vegetation, featuring alpine herbaceous plants, shrubs, and woody plants. This transitional site lends itself to the study of vegetation composition and succession. The senior au­thor has taken students and researchers to this site since 1984 and has made an extensive survey of its botanical composition from 1984-1990 and 2007-2009. A complete list of the species encountered is presented in Supple-mentary Table 1. In total, 88 species were found at the Shinjenkan site, all of which were found in other parts of the Central Mountain alpine region. Thirty three of these species were recorded over 100 years ago (1905-2010), 25 species were recorded over 90 years ago (1911-2010), 15 species over 80 years ago (1922-2010), two species re-corded before 1940, and the remaining species were docu-mented after 1963 (Supplementary Table 1).
*Available land for plant growth as plants migrate upwards.
In addition, a comparison of the botanical composi­tions of pteridophytes, gymnosperms, and angiosperms (including dicotyledon and monocotyledon) reveals that there were 62 species in 1984, 90 species in 2007, and 83 species in 2008 (Table 3). The number of species appeared to increase from 1984 to 2008, and naturally, we cannot discount the hypothesis that this significant increase is due to global warming.
Botanical composition of the Hehuan East Peak
At the Hehuan East Peak study site, we made an in-
CHOU et al. ―Diversity of alpine vegetation affected by climate change
507
ventory of the seasonal species distribution for 2008 and 2009. Twenty-four species (Ainsliaea macroclinidioides, Carex spp., C. satzumensis, Deschampsia fhxuosa, Fes-tuca ovina, E. rubra, Gentiana arisanensis, G. davidii var. formosana, Hieracium morii, Hypericum nagasawae, Ju-niperus formosnus, J. squamata, Lycopodium lavatum, L. obscurum, L. pseudoclavatum, L. selago var. appressum, L. veitchii, Miscanthus transmorrisonensis, Platanthera brevicalcarata, Rhododendron pseudochrysanthum, Sol-idago virgaurea var. leiocarpa, Trichophorum subcapitat-um, Veratrum formosanum and Yushania niitakayamensis) were commonly found both years (Table 4). Of these, five species (Gentiana arisanensis, L. obscurum, M. transmor-risonensis, Trichophorum subcapitatum and Yushania niitakayamensis) were present every season, indicating their dominance. Some species may have died or were not found in the winter or spring. Thus, the a diversity at the site is about 24, a rather low value for species diversity.
Figure 1. Annual temperature and precipitation after subtract­ing the average temperature and precipitation for the years 1930­2008. Data obtained from the Central Weather Bureau of Taiwan.
Comparison of species diversity in the Hehuan-shan area
-0.6。C for every 100 m; thus, theoretically, the tempera­ture difference between Shihmenshan and Shinjenkan is about -3.6。C. The temperature difference would be caus­ative of plant diversity. It is not surprising, then, that the a diversity at the Shinjenkan site was significantly higher than that at the other three sites. Naturally, species richness (a diversity) decreases as elevation increases. Further­more, y diversity refers to the total number of species in a region or at larger geographical scales; thus, the y diversity of each site in the Hehuanshan area is also variable and decreases as elevation increases. In fact, at both the HHE and HHM sites, the number of species in winter and spring is exceedingly low, indicating that plant growth at higher elevations during the cold winter and early spring is diffi­cult. This is reflected by the p-diversity, which is expressed as the value of a/y, and represents the variability of spe­cies composition along an environmental gradient. When the temperature decreases, the p diversity decreases. We observed that the number of species along the elevation of Highway 14A decreases when the elevation increases. In
A botanical inventory of the Hehuanshan area was performed at four study sites (Shinjenkan, Shihmenshan, Hehuan East Peak, and Hehuan Major Peak). The total number of species and the a, p, y diversity are presented in Table 5. The alpha (a) diversity is related to the population concept of species richness and can be used to compare the number of species in particular places or ecosystems, such as forests (Primack, 1998). Among the four sites in the He-huanshan area, the a diversity was significantly higher at the Shinjenkan site as compared with the other three sites; this difference is presumably due to temperature at the lower elevation.
The altitudinal difference between Shinjenkan and Shi-hmenshan is about 600 meters, while that between Shin-jenkan and both the Hehuan East Peak (HHE) and Hehuan Major Peak (HMP) is about 1,000 m. In general, when the elevation increases, the temperature is usually reduced by
Table 3. Comparison of the botanical composition at the Shinjenkan site in different inventory years.
508
Botanical Studies, Vol. 52, 2011
Table 4. Dynamics of plant species present at the Hehuan East Peak site.
1 +, Indicates the presence of the species; ―, Indicates that the species was not found.
addition, from a comparison of the inventory data between the 1980s and 2000s, the number of species at the Shin-jenkan site appears to increase. Presumably, this is due to the invasion of species from elsewhere and/or is likely due to global warming.
jenkan would move upwards to the SMS site (3,145 m) and continue to move to the HHM and HHE sites (3,410 m). The migration of species is discussed in detail below.
The average number of species at each site was com­puted and the average of Simpson's index was obtained by the techniques described by Barbour et al. (1987). The species distributions at the four sites at different eleva­tions in Shinjenkan and Shihmenshan are given in Table 6. These data show that species diversity, as measured by Simpson's index, is significantly higher (0.85) at the SJK site, drastically decreasing to 0.17 at the Shihmenshan site and to 0.10 at both HHE and HHM. A significant negative correlation between elevation and Simpson's index was found (Figure 2). This finding agrees with our previous results (Table 5), concluding that the topographic gradient certainly leads to the temperature change that results in variable plant diversity.
If funds are limited and sites for conservation need to be prioritized, the Shinjenkan site should be considered highest priority, due to its the exceedingly high y diversity as compared with the other three sites.
Based on a, p, y diversity, the Shinjenkan site exhibits the greatest mountain biodiversity. It is also interesting to note that when elevation increases the plant diversity in terms of a and y decreases, indicating that temperature is a causative factor affecting plant populations. According to the IPCC, the temperature of the four sites simultaneously increased by 0.6。C, 1.5。C, and 3。C in the A1, B1 and B2 scenarios, respectively. Consequently, the plants in Shin-
CHOU et al. ―Diversity of alpine vegetation affected by climate change
509
Table 5. Variation of the diversity index of plants distributed at the four study sites with seasons.
1SJK: Shinjenkan (新人崗)；SMS: Shihmenshan (石門山)；HHE: Hehuan East Peak (合歡東峰)；HHM: Hehuan Major Peak (合歡主峰).
2 ― ： Not detected.
Table 6. Comparison of Simpson's diversity index of plants found at 4 sites in the Hehuanshan area.
Migration of plant species in the Central Moun­tains of Taiwan
in zone A. In zone B, Eupatorium formosanum is expected to migrate 419 m; thus, it will invade to 3,038 m asl in zone D, and Lyonia ovalifolia will remain in zone B. In zone C, all seven species are expected to migrate upwards. Of these, three species, Aster taiwanensis, Deutzia pulchra and Pinus taiwanensis, will migrate into zone D, while the remaining four species will migrate past zone D to zone E. In zone D, two species, Rosa transmorrisonensis and Hy-drocotyle setulosa, will remain, and the other seven spe­cies are expected to migrate upwards to zone E; however, only one species, Hemiphragma heterophyllum, will mi­grate downwards to zone C. In zone E, two species, Rho­dodendron rubropilosum and Rubus rolfei, will remain in the same zone, and Artemisia morrisonensis will migrate 578 m to reach 3,978 m, which is beyond the highest peak of Yushan. This species is thus expected to become extinct.
Plant distribution results for Alishan, using the same methodology, are given in Table 8. In zone A (2,000-2,250
According to the IPCC scenario A1, last century's global temperature increased by 0.6。C, which was coin­cident with the elevation change. An altitude increase of 100 meters is associated with a temperature decrease of -0.6。C. Thus, plants would have to migrate 100 meters or more upwards in order to compensate for the temperature change registered over the past century.
Regarding vegetation distribution in the Hehuanshan area, we arbitrarily divided the vegetation zone into 6 zones, namely zone A (2,250-2,500 m), zone B (2,500­2,750 m), zone C (2,750-3,000 m), zone D (3,000-3,250 m), zone E (3,250-3,500 m), and zone F (3,500-3,750 m). By the aforementioned analyses, those plants expected to migrate upwards, downwards, or to remain in the same zone are listed in Table 7. In zone A, Acer serrulatum is expected to migrate upwards only 141 m; thus, it remains
510
Botanical Studies, Vol. 52, 2011
m), three species (Lyonia ovalifolia, Rhododendron rubro-pilosum and Sedum morrisonense) are expected to migrate upwards to the higher zone B, C, and B, respectively; however, Aster taiwanensis appears to migrate downwards to zone A, and Cirsium arisanense remains in zone A. Re­garding species migration in zone B (2,250-2,500 m), two species (Deutzia pulchra and Angelica morrisonicola) will migrate downwards to zone A, one species Hemiphragma heterophyllum will migrate upwards to zone C (2,500­2,750 m), and the remaining three species will not migrate out of zone B. It is interesting to note that no data are available for zone C.
Figure 2. Comparison of the average number of species and Simpson's diversity index against elevation at each experimental site.
Regarding the species distribution in the Yushan area, six zones, namely, zone A (2,500-2,750 m), zone B (2,750-
Table 7. Plant species expected to migrate to different altitudes in the Hehuanshan area.
CHOU et al. ―Diversity of alpine vegetation affected by climate change
511
Table 8. Plant species expected to migrate to different altitudes in the Alishan area.
Table 9. Plant species expected to migrate to different altitudes in the Yushan area.
512
Botanical Studies, Vol. 52, 2011
3,000 m), zone C (3,000-3,250 m), zone D (3,250-3,500 m) and zone E (3,500-3,750 m), and zone F (3,750-4,000 m), were defined, and the details of plant migration are given in Table 9. In zone A, Aster taiwanensis is expected to mi-grate upwards to zone B and Eupatorium formosanum will migrate to zone C. In zone B, Picris hieracioides subsp. morrisonensis and Rubus rolfei will remain in the same zone. In zone C, two species (Rhododendron rubropilosum and Spiraea formosana) will migrate upwards to zone E, but Hemiphragma heterophyllum will migrate downwards to zone A. Rosa transmorrisonensis and Gaultheria itoana are expected to migrate upwards to the higher zone E. In zone D, three species (Anaphalis morrisonicola, Artemisia morrisonensis and Swertia randaiensis) are expected to migrate into higher elevations above zone F; thus, these species might be expected to become extinct. Two species (Gaultheria itoana and Triplostegia glandulifera) will mi-grate into zone E. In zone E, one species, Luzula taiwani-ana will remain in the same zone, but another species, Sedum morrisonense, will migrate upwards to the higher-elevation zone F. Finally, in zone F, two species (Dianthus pygmaeus and Pimpinella niitakayamensis) will remain in the same zone, but three other species (Hypericum naga-sawai, Cirsium arisanense and Angelica morrisonicola) are expected to migrate upwards to higher elevations above 3,950 m; thus, these three species might eventually become extinct. In conclusion, the number of migrating species in the Yushan area is expected to be lower than that in Alishan and Hehuanshan; however, six species are eventually expected to become extinct.
Jump and Penuelas (2005). We used similar approach to study the molecular adaptation of Alnus formosana and Pinus taiwanensis.
In addition to global warming, precipitation is also one of major factors that determine vegetation change. Although a weather station was not present in the He-huanshan area, we were able to use weather data from nearby stations at Alishan and Yushan. During the past century, the rainfall pattern has varied significantly in the mountainous area below an elevation of 2,000 m. Liu et al. (2009) showed that the frequency of small or medium rainfall (< 40 mm/each) significantly reduced but that of heavy rainfall (> 400 mm/each typhoon) increased by 50%. Liu et al. (2009) further indicated that the frequency of heavy rainfall will increase by 25% when the global temperature increases by 1。C. If that occurs in Taiwan, re­sulting disasters such as landslides in areas of heavy rain­fall will severely impact forest communities. Fortunately, this has not occurred very often in the Hehuanshan (HHS) area due to the dominant and dense vegetation growing there, such as Yushania niitakayamensis (bamboo plant) and Miscanthus transmorrisonensis (grassland vegeta­tion). In addition, a well-protected ecological conservation area has been defined in the Hehuanshan area as part of the Taroko National Park; thus, the Hehuanshan region has only been slightly affected by rainfall. Furthermore, Lin et al. (2010) developed a yearly warning index to assess the climatic impact on the water resources in Taiwan, which indicated that the water resources will increase in the unpopulated eastern region but will decrease over other densely-populated and densely-industrialized regions. However, they did not mention a change in the Hehuan-shan area. Presumably, the rainfall pattern has not changed significantly in the area above 2,000-m elevation over the last century; however, the greatest indication of climate change in the Hehuanshan region may be the temperature increase caused by global warming.
In all cases, plant migration toward higher elevations means a reduction in the land available for plant coloni­zation. Using ArcView software and a digital geographic model, the available land for plant growth above 3,000 m was predicted to be only 30% of that in the 2,750-3,000 m elevation zone (Chou, 2009, unpublished data). At higher elevations, the available habitat decreases, resulting in some species becoming extinct (Chou, 2009, unpublished report). Indeed, there is almost no vegetation distributed in areas above 3,900 m. In consequence, plants migrating to higher elevations above 3,900 m are likely to become ex­tinct eventually. Similarly, Peason and Raxworthy (2008) indicated that upslope displacement of species is one of the expected biological responses to climate change. They further reported that locally endemic species in Madagas­car such as amphibians and reptiles would be pushed up to summits and would be vulnerable to extinction as they reach the tops of mountains. Guisan (2008) also predicted the effect of climate change on alpine flora in the Swiss Alps and in other mountain ranges in Europe, and forecast the change of alpine species distribution and diversity.
Jump et al. (2006) suggested that a rapid increase in global temperature is likely to impose strong directional selection on plant populations, which was demonstrated by a unique study of Fagus sylvatica conducted in Spain. Evidence of the evolutionary effects of climate change on natural populations was reviewed by Thomas (2005) and
Acknowledgements. We appreciate the diligent work of former assistants, Ms. C. H. Huang and Mr. C. H. Liao, for their laboratory and field assistance during the course of this study. We also acknowledge the help of Dr. Hiroshi Ikeda and the staff at the Herbaria and databasess of Tokyo University, the National Taiwan University, and Academia Sinica, who allowed us to examine specimens and vouch­ers. We are indebted to Professor Henrik Balslev for his critical review and thorough editorial correction and to two other anonymous reviewers for their helpful criticism on the manuscript. The financial support received from the National Science Council of Taiwan (NSC 96-2625-Z-039-001; NSC 97-2625-M-039-001; NSC 97-2625-M-039-002; NSC 98-2911-I-039-001) to C. H. Chou is greatly appreciated.
LITERATURE CITED
Barbour, M.G., J.H. Burk, and W.D. Pitts. 1987. Terrestrial Plant
CHOU et al. ―Diversity of alpine vegetation affected by climate change
513
Ecology. 2nd Ed. The Benjamin/Cummings Publishing Com­pany, Inc. Menlo Park, Reading, Don Mills Workingham Amsterdam, Sydney, Singapore, Tokyo, Madrid, Bogota, Santiago San Juan.
Beerling, D.L. and F.I. Woodward. 1994. Climate change and the British scene. J. Ecol. 82: 391-397.
Chen, Y F. 1995. Flora of Vegetation in Taiwan. Vol. I. Yu Shan Publisher, Taipei, Taiwan, 303 pp.
Chou, C.H. et al. 2009. Impact assessment of climate change on biodiversity of alpine vegetation in Taiwan. Project Report of the National Science Council, Taipei, Taiwan.
Climate Change Science Compendium. 2009. United Nations Environment Programme. 69pp.
Gomez, K.A. and A.A. Gomez. 1976. Statistical procedures for agricultural research with emphasis on rice. The Interna­tional Rice Research Institute, Los Banas, Laguna, Philip­pines.
Guisan, A. 2008. Climate change impacts on alpine biota. Pro­ceeding of International Conference Biodiversity Informat­ics and Climate Change Impacts on Life. DanBIF, April 5-6, 2008. Copenhagen, Denmark.
Huang, T.C. 1993. Plant Taxonomy: Flora of Vascular Plants in Taiwan. Nantien Book Publisher, Taipei, 658 pp.
Huang, T.C. et al. (eds.) 2002. Flora of Taiwan, 2nd edition. Vol. 6. Dept. of Botany, National Taiwan University, Taipei. pp.
343.
Hsieh, C.F. 2002. (see Huang, T. C. eds. Flora of Taiwan) International Conference Biodiversity Informatics and Climate
Change Impacts on Life. 2008. April 5-6, 2008. DanBIF,
University of Arhaus, University of Copenhagen, Denmark. International Symposium on Global Mountain Biodiversity,
2008. June 7-10, 2008. China Medical University, Taichung,
Taiwan.
International Symposium on Global Biodiversity, Human and
Well-being, 2009. December 3-9, 2009. China Medical Uni­versity, Taichung, Taiwan.
IPCC. 1995. Climate change 1995: A report to the IPCC United
Nations Environmental Programme, Intergovernmental Panel on Climate Change.
Jump, A.S., J.M. Hunt, J.A. Martinez-Izquierdo, and J. Penuelas. 2006. Natural selection and climate change: temperature-
linked spatial and temporal trends in gene frequency in Fa-
gus sylvaticus. Mol. Ecol. 13: 3469-3480.
Jump, A.S. and J. Penuelas. 2005. Genetic effects of chronic habitat fragmentation in a wind-pollinated tree. Proc. Nat. Acad. Sci. USA 103: 8096-8100.
Lin, S.H., C.M. Liu, W.C. Huang, S.S. Lin, Z.H. Yen, H.R. Wang, J.T. Kuo, and Y.C. Lee. 2010. Developing a yearly warning index to assess the climatic impact on the water re­sources of Taiwan, a complex terrain island. J. Hydrol. doi: 10.1016/ j.jhdrol.2010.06.024.
Liu, S.C., C.J. Fu, C.J. Shiu, and F. Wu. 2009. Temperature dependence of global precipitation extreme. Geophysical Research Letter 36: L17702, doi: 10.1029/2009 GL040218, 2009.
Pearson, R. and C.J. Raxworthy. 2008. Extinction vulnerability of tropical mountain endemism to climate change: observa­tions and uncertainties. Proc. International Conference Bio­diversity Informatics and climate change Impacts on Life. DanBIF, April 5-6, 2008, Copenhagen, Denmark.
Prabhakar, S.V.R.K. and R. Shaw. 2008. Climate change adapta­tion implications for drought risk mitigation: a perspective for India. Climate Change 88: 113-130.
Primack, R.B. 1998. Esentials of Conservation Biology. Sinauer Associates. Publishers, Sunderland. Massachusetts, USA.
Simpson, E.H. 1949. Measurement of diversity. Nature 163:
688.
Su, H.J. 1984a. Studies on the climate and vegetation types of the natural forests in Taiwan (I). Analysis of the variation in climate factors. Quart. J. Chin. For. 17: 1-14.
Su, H.J. 1984b. Studies on the climate and vegetation types of the natural forests in Taiwan (II). Altitudinal vegetation zones in relation to temperature gradient. Quart. J. Chin. For. 17(4): 57-73.
Su, H.J. 1985. Studies on the climate and vegetation types of the natural forests in Taiwan (III). A scheme of geographical climate regions. Quart. J. Chin. For. 18(3): 33-44.
Thomas, C.D. 2005. Recent evolutionary effects of climate change. In TE. Lovejoy and Li Hannah (eds.), Climate change and Biodiversity, Yale University press, New Haven, Connecticut, pp. 75-88.
Tsukada, M. 1966. Late Pleistocene vegetation and climate in
Taiwan (Formosa) Proc. Nat. Acad. Sci. USA 55: 543-548.
514
Botanical Studies, Vol. 52, 2011
Supplementary Table 1. Species distributed at the Shin-Jen-Kan site and species collected and deposited in herbaria of Tokyo Uni­versity, National Taiwan University and Academia Sinica.*
CHOU et al. ―Diversity of alpine vegetation affected by climate change
515
Supplementary Table 1. (Continuing)
*Plant specimens collected before 1945 were obtained by Japanese botanists and vouchers were deposited in the herbarium of To­kyo University, while specimens collected after 1946 were obtained by local botanists and vouchers were deposited in the herbaria of the National Taiwan University or Academia Sinica, Taipei.
516
Botanical Studies, Vol. 52, 2011
台灣中部高山植物之植物歧異度受氣候變遷之影響
周昌弘1 黃琮竣1 李彥屏2 陳志遠3 許再文4 陳志輝4
1中國醫藥大學生物多樣性研究中心生態暨演化生物學研究所
2行政院農業委員會林務局
3國立屏東科技大學森林學系
4台灣特有生物研究保育中心
由於台灣地理位置、地形及棲地的變化造成豐富的植物歧異度，從過去百年及現今的植物調查，特
 別是在中央山脈的合歡山地區，我們得以了解植物的組成。在合歡山地區分爲四個高度帶，即2,000~
2,500公尺、2,500~ 3,000公尺、3,000 ~ 3,500公尺及3,500 ~ 3,950公尺，以了解其植物的組成，並選擇
新人崗（2,250公尺）、石門山（3,000公尺）、合歡山東峰（3,500公尺）及合歡山主峰（3,500公尺）的
樣區（10x10公尺），分別調查其植物組成，並由其數據計算其oc 、 p 、 Y及辛普森（Simpson's)歧異
度指數以比較上述硏究地之植物歧異度。其結果顯示a 、 p 、 y植物歧異度隨海拔高度之增加而遞減。
但在新人崗實驗地，其a歧異度隨年代而增加，辛普森指數隨海拔高度亦遞減，顯示氣候對合歡山地區
的植物歧異度有顯著的影響，從百年來的紀錄分析得知，在合歡山地區有16個物種會往較高的海拔移
出；7個物種留在同一海拔高度帶。在阿里山地區則有7個物種移出，4個物種留在同一海拔高度帶；
在玉山地區則有15個物種移出，5個物種留在原海拔帶，但至少有6種因移出的高度太大已超越3,950
公尺，故可能會滅絕。此結果顯示，若全球暖化發生於台灣則台灣高山植物將會有遷移的現象，會有相
當多的物種往更高的地區移動，最後導致至少6種植物種的滅絕。
關健詞：阿里山；高山植被；α、β、γ歧異度；全球暖化；合歡山；植物遷移；辛普森歧異度；玉山。