ARTICLE IN PRESS Urban Forestry & Urban Greening 9 (2010) 139–148

Contents lists available at ScienceDirect

Urban Forestry & Urban Greening journal homepage: www.elsevier.de/ufug

A study on village forest landscape in small island topography in Okinawa, Japan Bixia Chen a,1, Yuei Nakama b,n a b

School of Economics, Fujian Normal University, China Department of Subtropical Agricultural Science, Faculty of Agriculture, University of the Ryukyus, Japan

a r t i c l e in fo

Keywords: East Asia Feng Shui Forest composition Small islands Village landscape

abstract Forest landscape based on Feng Shui concepts in East Asia deserves research focus for its cultural and ecological contexts. How to contain the wind is the primary principle of Feng Shui practice in small island villages. To protect from strong wind, house-embracing Fukugi (Garcinia subelliptica Merr.) tree lines have been planted around the hamlets and along the coastline in small islands. After Tonaki and Bise villages, we continued to study the actual forest structure, e.g., the forest layout, composition, and density, and further discussed the regeneration and management of Feng Shui trees on small islands. Another objective of this study was to compare the features of the house-embracing Fukugi trees in Aguni Island to those in the two former survey sites of Tonaki Island and Bise village. House-embracing Fukugi trees along the village borderlines and in the north were thicker in Tonaki and Bise villages. In contrast, Fukugi trees on Aguni Island are laid out almost single file. Differences in village topography might contribute to the difference of layout. Hamlets in Aguni are located on the southern part of the island, backed by the high land, and somewhat far from the seashore, while Tonaki and Bise villages are located close to the sea. Fukugi tree density in Aguni was much lower, while, the mean DBH was higher than those in Tonaki and Bise. A large number of small trees were found in the lower storey in the previous survey sites in Tonaki and Bise, where the villages were built on sandy soil. In contrast, small trees were few due to the regular cutting in Aguni Island. Routine chopping and cutting of extended branches were still conducted once a year on Aguni Island. The Shimajiri Mahji soil might also contribute to the few lower storey trees. The biggest tree on Aguni island is estimated to be about 296 years old. The old trees might have been planted prior to the building of Goban villages. Different to Tonaki and Bise, hamlets on Aguni might not be Goban villages. It also suggests that Fukugi trees might have been planted prior to 1737, though central government started to recommend to plant a Fukugi tree belt around every house, every village, and along the coastline in 1737. & 2009 Elsevier GmbH. All rights reserved.

Introduction With economic development and urbanization, the forestland in the world, in particular, urban and sub-urban green space, has been decreasing rapidly. Natural forests have been the focus of environmental conservation. Urban forests and sub-urban forests have attracted research interest quite recently. The economic, ecological, social and cultural significance of urban and sub-urban forests have been highly valued in previous studies. Vegetation can moderate harsh urban climates, e.g., by cooling the air through shading, reducing wind speed impacts and minimizing

n

Corresponding author. E-mail address: [email protected] (Y. Nakama). 1 JSPS Postdoctoral Fellow, Faculty of Agriculture, University of the Ryukyus.

1618-8667/$ - see front matter & 2009 Elsevier GmbH. All rights reserved. doi:10.1016/j.ufug.2009.12.004

pollution (Konijnendijk, 1999). Forests help to control soil erosion. The level of biodiversity of green areas, close to where people live, is often surprisingly high. In contrast to urban forest research in the western world, forest use in East Asia still is little mentioned. Forest landscape based on Feng Shui concepts in East Asia deserves research focus for its cultural and ecological contexts. Some studies mention that trees have been prevalently used as Feng Shui symbols in mainland China (Graham, 1961; Feuchtwang, 1974; He, 1990), Hong Kong (Webb, 1995), and Korea (Shibuya, 2003; Whang and Lee, 2006). Villages were built by the slope sides in the new territories in Hong Kong. Feng Shui forests behind the villages have been well preserved to embrace the villages (Anderson, 1996; Yip, et al., 2004). Greenery on the Dragon Mountain in classic Feng Shui has frequently been cited (He, 1990; Wang, 1992). A few old big

ARTICLE IN PRESS 140

B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

evergreen trees are prevalent in Southern China (Graham, 1961; Feuchtwang, 1974). Big trees have been planted at the entrance of the village with the aim of holding back Bad Energy. House-embracing Garcinia subelliptica Merr. trees (J. Fukugi Yashikirin) in Okinawa are unique in East Asia (Chen, 2008). ) was used in old Actually, the term ‘‘yashiki Ho:go’’ ( documents during the Ryukyu Kingdom Period. It is assumed that the word ‘‘yashikirin’’ had not been used until Meiji Period in Okinawa. Ho:go is a special word related to Feng Shui. Fukugi yashikirin in Okinawa was planted based on Feng Shui concepts in order to contain living energy, while, a shelter belt was planted surrounding the house for the purpose of protecting from wind or snow in mainland Japan. A Ryukyu Island forest landscape is set up based on Feng Shui concepts. House-embracing trees are a critical part of village landscape, and a part of land-use system closely related to agricultural society. However, Fukugi trees have been lost in many islands with the building of concrete houses. Thus, it is an urgent issue to preserve the existing house-embracing trees and record the local knowledge of forest management and regeneration in small island topography. We have previously published our research results on the house-embracing Fukugi trees on Tonaki Island and in Bise village on the northern part of Okinawa Island (Chen et al., 2008). This study continues to clarify the actual forest structure, e.g., forest layout, composition, and density, and further discusses the regeneration and management of woodland in another small island of Aguni Island. Another objective of this study is to compare the features of the house-embracing Fukugi trees on Aguni Island to those in the two former survey sites of Tonaki Island and Bise village.

Feng Shui and national policy Feng Shui was originated in mainland China and had a wide influence on neighboring countries and regions in East Asia. It is assumed that Feng Shui was first introduced to Okinawa, in the Kingdom of Ryukyu period, by the first Chinese immigrants in the 14th century (Machida and Tsuzuki, 1993). During the 17th and 18th centuries, Feng Shui was adopted as a national policy and was applied to many aspects of the building of houses, villages, graveyards, and to mountainous forest management. According to Yasyu Nakamatsu (1977), a planned Ryukyu village layout including the planning of coastline, farming lands, and satoyama, with a center of the so-called Go Board Pattern village building, was built under the influence of Feng Shui concepts. Satoyama is the area of secondary forest hills closely related to traditional agricultural society. It is a land-use mosaic mainly composed of woodlands, paddy fields, crop fields, grasslands, and settlements (Ichikawa et al., 2006). Village reform, including land tenure reform by the Kingdom of Ryukyu from 1730s to 1750s, was also influenced by Feng Shui. During these decades, Fukugi trees (G. subelliptica) were planted to embrace every village house. Forest belts of pine (Pinus luchuensis Mayr) trees or Adan (Pandanus odoratissimus L.) were planted in the coastline. Forest belts were also planted to surround several hamlets. Somayama ( , Timber Mountain, soma is a Japanese-created kanji made up of two parts, , that means the designated mountains that can produce timber only for sacred places such as temples, shrines, and kingdom castle.) was managed based on the Feng Shui. The Wooded Mountain Methods J. Sanrinshinpi) by Sai On, a member of the Council of ( ), the highest governing body of Ryukyu at that time, Three ( who was notable for his great contributions to both diplomacy and domestic management, recorded the forestry regulations in

detail. At that time, a Feng Shui master was appointed to judge the village Feng Shui and maintain the village landscape. Ryukyu island Feng Shui village landscape Okinawa Prefecture, Japan’s southwestern most prefecture, is located at latitude 24–281 north and longitude 123–1321 east. It consists of 49 inhabited and 110 uninhabited islands. The islands are scattered over an area extending 1000 km from east to west and 400 km from north to south. The climate of Okinawa is influenced by the latitude, the surrounding ocean, the monsoon, the Black Current, and typhoons (Glacken, 1960). It has a subtropical climate with considerable humidity. It has a long typhoon season extending from June to November. The Ryukyu Kingdom was an independent kingdom, which ruled most of the Ryukyu Islands from the 15th century to the 19th century. Japan failed to exercise political control over the islands until the seventeenth century. The Ryukyu Kingdom was a Chinese tributary state during much of its existence and, therefore, had a much closer relationship with China than it with Japan. The Ryukyu Kingdom lost its independence when the Satsuma armies invaded the islands in 1609. Under the policy of Ryukyu Shobun (literally, the ‘‘disposition of the Ryukyu Kingdom’’), the Japanese government officially dissolved the kingdom in 1875, and incorporated the islands into the Japanese nation-state as ‘‘Okinawa Prefecture’’ four years later. Feng Shui concepts have played a vital role in shaping rural landscapes in China. With the introduction and development of Feng Shui in the Ryukyu Kingdom, Feng Shui has been used to assess the landscape and guide the building of the capital and the choice of village site and its planning on the small islands. Distinctions between mainland Feng Shui and Ryukyu island landscapes have been summarized (Chen, 2008). Topography has been the first factor in village location in China, Hong Kong, and Korea. The Form School of Feng Shui observes land formation and terrain and then determines location and orientation. Surrounding hills are desired to accumulate Qi for a favorable living site. Feng Shui practice primarily applies the repairing principle (C. Bu; K. Bi-bo: O. Ho:go). In mainland China, artificial buildings such as a bridge, a pagoda or a pavilion, and a grove of trees were built in the lowest site, Water Gate, to retain the fortune that would otherwise leave with water. Bi-bo in Korean means a supplementary measure to repair the defective Feng Shui landscape of certain vital sites such as Water Gate (K. Su-gu) or Dragon Mountain. The Ryukyu Feng Shui village is laid out with the village houses in the center in contrast to mainland China, which emphasizes landform village houses are encircled by multilayer forest belts of yashikirin, and kusatimui (preserved natural woods in the higher mountain in the back of a village, usually in the north) along with planted forest belt in the front to embrace the village and coastline preserving forest belts.

House-embracing Fukugi trees The house-embracing Fukugi tree is the one landscape factor most worth noticing in Okinawa. A well-ordered tree line was planted to surround all village houses. As is well known, the two basic principles of Feng Shui are to contain the wind and to accumulate the water. On small islands of Okinawa Prefecture, which are exposed to summer typhoons and winter monsoons, how to contain the winds must have been the top concern of Feng Shui practice. Tree lines were laid out to enclose the house in order to contain the strong winds.

ARTICLE IN PRESS B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

House-embracing forest, as a Feng Shui practice, has played an important role in island people’s lives. A questionnaire survey reveals that inhabitants held a high consciousness of Fukugi trees as providing shelter from typhoons, as scenic landscape, and as shading to cool the air in summer (Nakama and Koki, 2002). There is no agreement on when such a house-embracing tree landscape was formed. Here, we still adhere to the Nakamastu’s statement that Fukugi trees were planted in connection to jiwarisei (the practice of rural villages periodically redistributing land) in 1737. Garcinia subelliptica is considered to be distributed only in the Phillipines, Taiwan and Okinawa, with a few naturally distributed in southern Okinawa (Hatsushima, 1975). Fukugi is still little studied. The layout of house-embracing Fukugi trees on Tonaki Island and Bise village on the northern part of Okinawa have been surveyed (Chen et al., 2008). We continued to survey on Aguni Island, which also has well-preserved house-embracing trees. Different from the two former survey sites, local people conduct regular pruning of trees. A general layout of trees and village roads, as well as the characteristics of forest belts, will be discussed in this study.

Survey site Aguni (Fig. 1) is a tiny island 12 km in circumference, about 57 km northwest of Naha, Okinawa. It is located at latitude 261340 north and longitude 1271130 east. Aguni is triangular (Photo 1), with a total area of 7.64 km2. Ryukyu limestone, which is coral limestone formed and pushed up in the Pleistocene, extends mainly through the central part and the southern end of the island, forming terraces. Lava is found in the west of the island. Surface soils are mainly Shimajiri Mahji (dark red soil or Ryukyu limestone soil, neutral to weakly acidic), and a small area of Kunigami Mahji (red and yellow soil or other stone soils, strong acidic). Recently formed coral reef surrounds the island. Sugar

141

cane and sweet potato are widely planted. Two clustered hamlets of East and West villages are located on the center of the island, and another hamlet of Hama near the coast. The island is usually short of fresh water resources. As of November, 2007, the village had an estimated population of 874, and a density of 114 people per km2. A large number of villagers moved to work in Naha before WWII. Archaeological studies found that ancient people lived on the high mesa near the coast easily accessible to the sea in the gathering and hunting society (1984). People moved to the flat lowlands only several hundreds of years ago and reclaimed agricultural lands nearby (see Photo 2).

Methods We chose to survey four sites (see Fig. 2) with the best preserved house-embracing trees. Fukugi trees were planted in four directions around the house to shape embracing protection for the house. People cut big trees to build a house, also planted seeds to grow them, and removed the unwanted saplings; thus, a proper density and layout of Fukugi trees came into being under management. Since WWII, many trees have been cut after the building of concrete houses. These four survey sites have standing trees in four directions around the house. First, every tree taller than 1 m was numbered, and then tree height and diameter at breast height (DBH) were measured and recorded. The estimated tree age was calculated based on the formula [Age (years)= DBH(cm)/2  8 correlation coefficient=0.41] by Hirata (2006). The numbers of sprouts and saplings were also counted to understand the tree regeneration. Sprouts from the rootstock and the stumps were counted. Young trees less than 1 m tall were counted as seedlings. It is assumed that Fukugi house-embracing trees were cut to build the houses. The number of stumps was also counted. We also recorded the actual location of every tree in the woodland. We reproduced the actual layout of the trees with HO

Fig. 1. Location of Aguni Island. Aguni Island is small triangle island about 57 km northwest of Naha, Okinawa. Source: In Ryukyu Islands (1953) by United States Civil Administration No.4.

ARTICLE IN PRESS 142

B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

Photo 1. Aerial Photo of Aguni Isl. Source: The photo was provided by Aguni Village Office.

Photo 2. A panoramic view of hamlets in Aguni Isl. from the mesa in the west. Green trees, most of which are Fukugi trees, surround the houses inside the hamlets.

CAD software in the figures. On the basis of its DBH, the size of every tree was drawn in a circle of proportional size.

Results and discussion Layout of house-embracing Fukugi trees Altogether 564 trees in four sites were surveyed. From the figures drawn by CAD software, we can see that house-embracing Fukugi trees on Aguni Island are laid out almost in single file (see

Fig. 3). All surveyed houses were facing SSW, with a wide open space in the front. Tree lines were much thicker in the borderline of the village, in particular, those either facing the coast or standing in the north. We found that there was usually a single file tree line in the center of the village, while there were several tree lines in the borderline on Tonaki Island and Bise village. In contrast, trees were laid out in a wellordered line both inside the village or in the borderline of the village on Aguni. Such a difference might be attributed to differences in the village location. Aguni Island has an area more than twice that of Tonaki Island. The hamlets are located at the southern part of the

ARTICLE IN PRESS B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

143

No. 2

No. 3

No. 1 No. 4

Fig. 2. Location of survey sites. Source: Map of Aguni Island was provided by Aguni Village Office.

Fig. 3. Layout of house-embracing Fukugi (Garcinia subeliptica) trees in Aguni Isl. Scale= 1:600.Notes: small circles refer to house-embracing Fukugi trees. They were drawn on basis of the size of the trees’ DBH.

island, with the high land in the north, and relatively far from the ocean. The surveyed houses of Nos. 1, 2, 3, 4 are about 350, 280, 640, and 440 m away from the seashore, respectively.

According to Chorography of Aguni Village, the people originally lived in the high land in the west of the island. Several utakis were located in this area. Utaki is always a sacred site covered with

ARTICLE IN PRESS 144

B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

thick old forests, where local people praying to gods and their ancestors for a good harvest, a good catch of fish, fortune, and voyage safety. Utaki is always the original habitat of local people. People moved to the flat land and built the new hamlets in the current, more convenient place. The hamlets in Aguni were not included as a Goban village, also called a jiwari Village, which was newly built after 1737. A Goban village features grid order roads. Jiwari, was a collectively owned land system in modern Ryukyu. Under the jiwari system, all land is collectively owned, used, and managed. The land was allocated and redistributed in a certain period from 2 to 30 years, differing in hamlets. Thus, different to the previous two surveyed villages of Tonaki and Bise, Aguni village might have been built earlier than a Goban village. The descriptive data of surveyed Fukugi trees are listed in Table 1. The mean heights of house-embracing Fukugi trees were 457, 523, 521, and 536 cm. The mean DBHs were 13, 16, 17, and 16 cm. The mean estimated ages of surveyed trees were 54, 65, 68, and 64 years old. The oldest trees surveyed were 172, 184, 200 and 191 years old in the four surveyed houses. In our later survey of all huge Fukugi trees in the whole village, we found the biggest tree had a DBH of 740 cm, and was estimated to be about 296 years old. These huge trees might have been planted before 1737, the start of jiwarisei. Thus, people might have planted Fukugi trees as windbreak. However, after 1737, Fukugi trees were recommended by the central government, and became a specific part of village landscape in modern Ryukyu. The distribution frequency curves of tree height and DBH are shown in Figs. 4 and 5. The height frequency curves show that the majority of Nos. 1 and 3 were about 1–2 and 7–8 m. The majority of Nos. 2 and 4 were about 2–3 and 6–7 m in height (See Figs. 4, 6). In general, all surveyed trees have a two-storied structure of about 2 and 7 m. The distribution frequency curves of DBH show that the majority of trees were less than 5 cm (See Figs. 5, 7). Trees have an even distribution under 35 cm.

The mean height of house-embracing Fukugi trees on Aguni Island is not largely different from those in Tonaki and Bise. The mean heights of Fukugi trees surveyed were 512, 407, and 552 cm in Aguni, Tonaki, and Bise, respectively. It may be assumed that houseembracing Fukugi trees in Okinawa have been chopped to maintain a certain height. The highest trees have been kept less than 8 m tall. Before World War II, house-embracing forest was considered as common village property, and forest management such as cleaning and pruning were regularly organized in the Ryukyuan Islands. ), which was a compulsory labor service system before Buyaku ( the Meiji period, conducted the forest management. On Octobet 1st of the lunar calendar, there is a traditional ) in Aguni, when villagers pray to ritual called Kamama:i: ( the Fire God for fire prevention. On the same day, the branches of house-embracing trees that reach far enough that cover the village roads are to be cut and the tops of trees are trimmed. According to the local people, the tops are chopped once every several years. About 1 m higher than the roof seems to be desired by the local inhabitants. Bise and Tonaki have become two famous tourist sites of house-embracing Fukugi trees. In particular in Bise, Fukugi trees grow very tall without regular cutting of the tree tops. Tourists seldom arrive on Aguni Island, and the island people seem to have maintained their traditional life and cut the branches and the tree tops to acquire a pleasant living space with the Fukugi tree line. The mean DBH of surveyed trees on Aguni Island was the biggest among the three surveyed villages. The mean DBH of Fukugi trees in Aguni Island was 16 cm. In contrast, the mean DBHs in Tonaki Island and Bise village were 10 and 11.5 cm, respectively. The village location and routine chopping might have contributed to the relatively bigger size on average. Houses we surveyed are quite far from the sea. In contrast, the west side of Bise village is close to the sea. Houses on the eastern part of Tonaki Island are also facing the sea.

Table 1 Descriptive data of house-embracing Garcinia subelliptica trees. Surveyed house No.

No.1 No.2 No.3 No.4 Total

132 127 114 191 564

Tree height (cm)

DBH (cm) of trees

Estimated age of the trees (year)a

Mean

Maximum

Mean

Maximum

Mean

Maximum

457.4 523.4 521.2 536.3 511.9

946 835 992 1197 1197

13.4 16.3 17 16 15.7

43 45.9 50 47.8 50

53.6 65.2 68 64 62.8

172 183.6 200 191.2 200

The estimated tree age was calculated based on the formula [Age (years) =DBH(cm)/2  8 correlation coefficient =0.41] by Hirata (2006).

120 100 Tree number

a

Total amount

80 60 40 20 0 0

1

2

3

4

5 No.1

6 7 Height (m) No.2

No.3

No.4

8

9

10

11

12

Total

Fig. 4. A frequency distribution curve of the height of house-embracing trees by houses

13

ARTICLE IN PRESS 145

B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

There were 42, 14, 1, and 12 seedlings found in the surveyed houses, The ratios of stump number to the stand tree number were 0.2, 0.3, 0.4, and 0.4. The large number of stumps suggests that routine cutting has been conducted in the house-embracing woodlands. Small trees less than 1 m high were least found on Aguni Island among these surveyed villages. Only one was found in house No.3. The regeneration feature of few seedlings in Aguni is greatly different from Tonaki Island and Bise village. The large number of

Density and regeneration Table 2 shows the stand tree density on Aguni Island. The number of Fukugi trees taller than 1 m in every meter was counted as density. The tree densities were 1.6, 1.6, 1.8, and 1.8. The number of sprouts and seedlings were counted, respectively, to observe management and regeneration in houseembracing woodlands. A large number of stumps were found in the surveyed houses, namely, 29, 39, 42, and 68.respectively.

160 140

Tree number

120 100 80 60 40 20 0 5

0

10

15

20 No.1

25 DBH (cm) No.2

No.3

30 No.4

35

40

45

50

Total

Fig. 5. A frequency distribution curve of DBH of house-embracing trees by houses

130 Mean = 511.9 120

Std. Dev. = 254.241 N = 564

110 100

Tree number

90 80 70 60 103

50 92

97

92

40 30

60 42

20 32 24

10

16

0

5

1

0 0

100

200

300

400

500

600

700

800

900 1000 1100 1200

Height (cm) Fig. 6. Frequency of the height of house-embracing Fukugi trees in Aguni Isl.

ARTICLE IN PRESS 146

B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

140 Mean = 15.62 Std. Dev. = 11.06 N = 564

120

Tree number

100

80 143 60

84

40

75 67

63

64 43

20

14 7

4

0 0

5

10

15

20

25 30 DBH (cm)

35

40

45

50

Fig. 7. Frequency of DBH of house-embracing Fukugi trees on Aguni Isl.

Table 2 Density and regeneration of house-embracing G. subelliptica

No.1 No.2 No.3 No.4 Total

Area (m2)a

Perimeter (m)b

c

d

e

f

g

h

i

688 579.9 601.4 1120.1

81.49 68.83 57.33 94.56 302.21

132 113 102 170 517

1.6 1.6 1.8 1.8 1.7

285 167 152 211 815

42 14 1 12 69

29 39 42 68 178

22% 35% 41% 40%

179 244 265 303

a

Area of each residence unit surveyed. b Perimeter of each residence unit. c Refers to the total number of Garcinia subelliptica trees taller than 1 m. d = Garcinia subelliptica tree number/length of Garcinia subelliptica tree lines (m). e Refers to number of the sprouts grow from the rootstocks. f Refers to the seedling lower than 1 m. g Refers to number of cut stumps. h Refers to ratio of cut stumps to stand trees. i Refers to number of the sprouts grow from the cut stumps.

seedlings amount to 360, 280, and 213 in surveyed houses Nos. 6, and 7 in Tonaki Island and Type 3b in Bise village. Regular cleaning and cutting might have contributed to few seedlings. Shimajiri Mahji soil might be another factor in the low regeneration. A comparison of the three survey sites might suggest that the sandy land is the most proper for the regeneration and growth of Fukugi trees. Another layout of Fukugi trees directly related to few seedlings was that the branches lower than 2 m were kept to shape a natural fence, while, the branches among the low and high layers were cut to allow the wind and light.

Thus, a unique system of two-layer branches was shaped on Aguni (see Photo 3). The fewer small trees and large number of stumps might be attributed to the bigger mean DBH on Aguni Island.

Management Routine cutting of extended branches and chopping were still conducted on Aguni Island. Two villagers in each hamlet were hired to conduct the management of all house-embracing Fukugi trees. Branch cutting is conducted once a year. With the increase in use of automobiles, the village roads were broadened, and the branches facing the roads were cut for smooth traffic. However, overcutting without concern for a scenic landscape was also found (see Photo 4). Thus, a re-evaluation of the landscape beauty of house-embracing trees must be conducted in order to maintain its charm. In contrast, there exists difficulty in house-embracing tree management on Tonaki Island and in Bise village. About half of the houses on Tonaki Island were abandoned during these decades. The Fukugi woodlands around these abandoned houses became thicker with a large amount of small trees. These small trees block the winds and become the shelter of mosquitoes and other insects. In Bise village, the upper branches of Fukugi trees cross. It is difficult to tell from an aerial photo where the village roads are. Fukugi trees were better maintained on Aguni Island. A younger population composition and more convenient transportation might contribute to the better condition of forests.

ARTICLE IN PRESS B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

147

Photo 3. A two-storeyed branches of house-embracing G. subelliptica trees. Fukugi trees planted hundreds of years ago become big and the lower branches are kept to hide the garden from the outsiders, while the middle parts are emptied for ventilation.

Photo 4. A routine cutting even chopped the high branches along the roadside. When it comes to tree maintenance inside the hamlets, more emphasis is sometimes placed on smooth traffic over scenic landscape.

Conclusion Forest landscape in Okinawa, planned and based on Feng Shui concepts has adapted to the local climate and tree plantation used to repair the defective landscape. In contrast to the mountainous village site choices, how to guide the strong wind has been the primary principle in small island Feng Shui practice. To protect from

strong winds, house-embracing trees were planted in the hamlets in small islands and in the newly built hamlet on the coastline. Following Tonaki and Bise villages, house-embracing Fukugi trees on Aguni Island were surveyed. About 564 house-embracing Fukugi trees were surveyed in four sites. The features of the house-embracing Fukugi trees are summarized as follows. The mean tree height ranged from 457 to 536 cm.

ARTICLE IN PRESS 148

B. Chen, Y. Nakama / Urban Forestry & Urban Greening 9 (2010) 139–148

House-embracing Fukugi trees in the village borderlines and in the north were thicker in Tonaki and Bise village than those on Aguni. The difference in village location might have contributed to the difference of layout. Hamlets on Aguni are located on the southern part of the island, backed by the high land, and somewhat far from the seashore. The mean DBH ranged from 13 to 17 cm. The mean estimated tree age ranged from 54 to 68 years old. The oldest tree was estimated to be 200 years old in the four surveyed houses. The biggest tree in the whole island had a DBH of 740 cm, and is estimated to be about 296 years old. The old trees might have been planted prior to the building of a Goban village, which was said to be newly built after around 1737. Different from Tonaki and Bise, hamlets on Aguni might not be Goban villages. It might suggest that Fukugi trees were planted prior to 1737; however, Fukugi trees have been recommended to be planted around the whole hamlet since 1737. The density of stand trees, which were taller than 1 m was estimated to be about 1.6–1.8 in every meter of woodland. There were few small trees less than 1 m tall, but a large number of stumps were found in the woodlands. A large number of small trees were found in the previous survey sites in Tonaki and Bise, where the villages were covered with sand soil. In contrast, small trees were few due to the regular cutting on Aguni Island. Routine chopping and cutting of extended branches are still conducted once a year on Aguni Island. The Shimajiri Mahji soil might also contribute to the few lower storey trees. A routine cutting of saplings and chopping of the tree tops continued even after WWII in Aguni Island, Bise and Tonaki have become two most famous tourist sites for house-embracing trees in recent decades. In particular, the Fukugi tree top has not been chopped since WWII in Bise village, and the trees have become taller. In contrast, house-embracing trees have been maintained as a sound living habitat on Aguni. Frequent tree chopping, branch cutting, sapling cutting are conducted, and a proper density and height have been kept. House-embracing Fukugi tree landscape on Aguni Island represents an ideal living space. This work is a continuation of our previous surveys in Tonaki and Bise villages. Results from Aguni Island also exemplify that natural endowments have a significant influence on local Feng Shui practice. When and how such Feng Shui villages with houseembracing Fukugi trees were built is left to our future study.

References Anderson, E.N., 1996. Ecologies of the Heart: Emotion, Belief, and the Environment. Oxford University Press, New York. Chen, B., 2008. A comparative study on the Feng Shui Village Landscape and Feng Shui Trees in East Asia – a case study of Ryukyu and Sakishima Islands. The Science Bulletin of the Faculty of Agriculture, University of the Ryukyus 55, 25–80. Chen, B., Nakama, Y., Kurima, G., 2008. Layout and composition of houseembracing trees in an island Feng Shui village in Okinawa, Japan. Urban Forestry & Urban Greening 7, 53–61. Feuchtwang, S.D.R., 1974. An Anthropological Analysis of Chinese Geomancy. Southern Materials Center, INC, Taiwan, pp. 261. Glacken, C.J., 1960. The Great Loochoo: A Study of Okinawan Village Life. Graham, D.C., 1961. Folk Religion in Southwest China. The Smithsonian Institution, Washington DC. Hatsushima, S., 1975. Flora of the Ryukyus. Okinawa Biological Education and Research Society, Naha (in Japanese). He, X., 1990. The Source of Feng Shui. South-East University Press, Nanjing in Chinese. Hirata, E., 2006. On the estimation of the age of an old Garcinia subelliptica tree. In: NPO Body Corporate of Yamabiko (Ed.), On the Garcinia Subelliptica Trees in Okinawa. Okinawa Green Promotion Committee, Naha, pp. 41–46 (in Japanese). Ichikawa, K., Okubob, N., Okuboa, S., Takeuchi, K., 2006. Transition of the satoyama landscape in the urban fringe of the Tokyo metropolitan area from 1880 to 2001. Landscape and Urban Planning 78, 398–410. Konijnendijk, C.C., 1999. Urban forestry in Europe: a comparative study of concepts, policies and planning for forest conservation, management and development in and around major European cities. Doctoral Dissertation. Research Notes No. 90. Faculty of Forestry, University of Joensuu. Machida, M., Tsuzuki, A., 1993. An introduction to Feng Shui village-studies on Kitaki Mountain Feng Shui. Ryukyu University. Law Faculty Bulletin (History and Geography) 36, 99–213 in Japanese. Nakama, Y., Koki, Z., 2002. The significance of housing shelter forest and regional inhabitants’ consciousness in island regions. Journal of the Japanese Society of Coastal Forest 2 (1), 5–11. Nakamatsu, Y., 1977. The ancient Village: A Folklorist Study of Okinawa. Okinawa Times Press, Naka (in Japanese). Shibuya, S., 2003. Forest and Bi-bo in rural Korea – an explanation from Feng Shui and woodland management. In: Ishihara, H. (Ed.), Rural Space Research, G: Rural Culture and Space Recognition. Daimeido, Tokyo, pp. 413–425 (in Japanese). Wang, Q. (Ed.), 1992. Research of Feng Shui theory. Tianjing University Press, Tianjing (in Chinese). Webb, R., 1995. The Village Landscape. In: Hase, P.H., Sinn, E. (Eds.), Beyond the Metropolis: Villages in Hong Kong. Joint Publishing, Hong Kong, pp. 37–47. Whang, B.-C., Lee, M.-W., 2006. Landscape ecology planning principles in Korea Feng Shui, Bi-bo woodlands and ponds. Landscape Ecology Engineering 2 (2), 147–162. Yip, J.K.L., Ngar, Y.N., Yip, J.Y., Liu, E.K.Y., Lai, P.C.C., (Plant Working Group, Agriculture, Fisheries and Conservation Department), 2004. Venturing Fung Shui Woods. Cosmos Books Ltd., Hong Kong.