Trees (2013) 27:343–358 DOI 10.1007/s00468-012-0775-7

REVIEW

Dendrochronology in Southeast Asia Nathsuda Pumijumnong

Received: 29 November 2011 / Revised: 5 September 2012 / Accepted: 11 September 2012 / Published online: 27 September 2012 Ó Springer-Verlag 2012

Abstract Dendrochronological research in Southeast Asia is under development; however, the amount of tress with potential for dendrochronological studies is restricted. For example, teak trees from India, Myanmar, Thailand, and Java are valuable dendrochronologic studies for ready climate response. Teak from Java is best suited for studying the El Nin˜o-Southern Oscillation and sea-surface temperatures, whereas Indian teak is used to reconstruct periods of drought in India. Further, Thai teak and Vietnamese cypress trees captured the long drought period that led to the demise of the Angkor reign (fourteenth–fifteenth century). Diverse techniques including anatomical observation, cambial markings, cell differentiation, and isotopic analysis prove the age and growth of invisible tropical tree rings. A number of invisible growth rings in trees from both tropical and subtropical forests have been identified, resulting in the advancement of dendrochronology. Climate change is a substantial challenge for most living things and natural resources. A greater understanding of tree species adaptation in this region is necessary. The understanding of long-term paleoclimate can be gained by researching old samples and archaeological materials from this region. Keywords Dendrochronology
Southeast Asia
Teak
Pine
Cypress
Ringless tree

Communicated by A. Braeuning. Special topic: Dendroecology in Asia. N. Pumijumnong (&) Faculty of Environment and Resource Studies, Mahidol University, Salaya, Phutthamonthon, Nakhon Pathom 73170, Thailand e-mail: [email protected]

Introduction The dendrochronology field began with studies by American astronomer A.E. Douglass in the beginning of the twentieth century (Douglass 1914), and continued in Europe at the end of the 1940s (Huber 1941). In Southeast Asia, literature shows that dendrochronology has been studied for some time (Coster 1927, 1928; Liese 1986). Problems with dendrochronology studies in tropical and subtropical forests are multifaceted, and have been outlined in three international workshops (Bormann and Berlyn 1980; Bass and Vetter 1989; Eckstein et al. 1995). The tropical and subtropical forests of Southeast Asia are some of the oldest ecosystems in the world, and heavy rain and high temperatures have a significant influence on this region. The area is home to various plant and animal species, making these tropical and subtropical forests immensely diverse. The influence of climate on tree growth is different for each species (e.g., Williams and Bunyavejchewin 2008); thus, in the same area and at the same time, a tree with leaves can be found next to a leafless tree. The temporal distance between leafless states of trees is often irregular, establishing a need for demonstrating the inter-annual yearly wood formation. Many studies have attempted to demonstrate this inter-annual yearly wood formation using various methods (Mariaux 1967, 1976, 1995; Shiokura 1989; Worbes 1983, 1984, 1986, 1989; Nobuchi et al. 1995). Logging and forest clearing for agriculture has, unfortunately, destroyed much of the subtropical/tropical forests, and studies aimed at understanding tree growth and age have largely been neglected. Instead, the enormous availability of tree species that produce durable wood has led to exploitation of forests in this region. Due to monsoons and high temperatures that result from intensive degradation of

123

344

these forests, environmental problems have developed in agriculturally demanding regions. Climate change and cyclic variations in climatic phenomena cause changes in rainfall patterns and increasing temperatures, all particularly evident in Southeast Asia. Many climate-driving factors inhibit the establishment of scenarios for future climate changes, particularly ocean-landmass distribution, the El Nin˜o-Southern Oscillation (ENSO), and the warm pool in the Indian Ocean. Nevertheless, regional climate data derived from instrumental records (e.g., in Thailand, climate data are digitally recorded from 1951 to the present) are useful for establishing growth-climate models, and the compilation of past climatic data obtained using dendrochronological methods has been initiated. This research review aims to explore the development and potential of dendrochronology in Southeast Asia and investigates its potential. This review is intended to cover only in the ‘‘Southeast Asia region.’’ A view of Southeast Asia as a region is attributed to strategy in World War II and formation of the Southeast Asia Military Command in 1943 (Roberts 2011). In this definition, Southeast Asia comprises 10 countries (Myanmar, Thailand, Laos, Cambodia, Vietnam, Malaysia, Singapore, Brunei, Philippines and Indonesia). This demarcation differs from the ecological boundary considered. The natural distribution of teak extends from India to the Philippines (Gyi and Tint 1985). We thus include Indian teak in this article. Teak is the first tree species in tropical and subtropical Southeast Asian climates that shows promise for dendrochronology (Pumijumnong 1995).

The first stage of dendrochronology in Southeast Asia (1915–1995) The distribution of natural teak (Tectona grandis L. F.) in South Asia (India) and Southeast Asia (Thailand, Myanmar, and Laos) is the most relevant source for dendrochronology in this area. The growth of teak in the Philippines and in Indonesia is thought to have originated from Indian seeds (Lamprecht 1986). In natural areas, teak trees are found in both mixed–deciduous forests and in monsoon climates. Teak is conducive to long-term historical studies because its timber is extremely durable. Teak possesses a naturally oily wood, making it suitable for construction of outdoor furniture. Teak is highly resistant to moisture, fire, acids and bases, and has an attractive straight grain. Since the twentieth century, the dendrochronology of teak has gained popularity. Teak grows at different elevations within a large distribution area in Southeast Asia: in north-to-central Thailand, the growth range is 100–900 m asl; in India, documented range is from 400 to 800 m asl to even 1,000 m asl in some areas;

123

Trees (2013) 27:343–358

in Myanmar, teak trees range from 900 to 1,000 m asl up to 1,400 m asl (Dahms 1989). The teak growth period in Java occurs during the rainy season, between November and May (Coster 1927). During dry season, teak sheds its leaves and new leaves bud at the beginning of the rainy season, a process crucial to the formation of distinct annual rings. Geiger (1915) and Coster (1927, 1928) have explored the relationship between tree-ring width and climate data in Java. Berlage (1931) was the pioneering scientist who established a 400-year-long Java teak tree chronology. By correlating chronology data with climate data, Berlage found a correlation between teak-ring width and the June to October period (i.e., rainfall during the dry season). Additionally, Berlage determined that the number of rainy days, and not the total amount of rainfall, is the most influential factor affecting teak growth, directly influencing the current year and indirectly impacting the following year of teak growth. De Boer (1951), Jacoby and D’Arrigo (1990) and D’Arrigo et al. (1994) reanalysed teak chronology data published by Berlage (1931). De Boer (1951) compared teak chronology to the climate data in Jakarta from 1864 to 1929 and found that ring width was dependent on the number of cloudy and rainy days. This did not correlate with temperature, rainfall amount, or the number of dry months. Moreover, there exists a correlation between sunspots and the El-Nin˜o phenomenon. Murphy and Whetton (1989) compared the chronology of Java teak and the Southern-Oscillation Index (SOI), ultimately reconstructing the SOI using Java teak. A reanalysis and update of Berlage’s (1931) chronology to 1989 now exists, using samples from a teak plantation. The currently available teak chronology spans 416 years and is considered a record of the ENSO variability (Palmer and Murphy 1993). Jacoby and D’Arrigo (1990) recalculated the correlation between Java teak and climate data, confirming Java teak growth depends on the timing of the beginning rainfall. D’Arrigo et al. (1994) extended the Berlage chronology (1514–1929) by 62 years (up to 1991), and found a positive correlation between the chronology data and rainfall, and a negative correlation between the teak tree-ring index and sea surface pressure during the dry monsoon season (from approximately May to October, just prior to the growth period in Java). The first developed teak tree-ring index in Java was documented by Berlage (1931) and difficulty in synchronizing teak samples was highlighted by Palmer and Murphy (1993), who described that some disks at various stages were considered unsuitable, and some ring boundaries remain indistinct. These were subsequently rejected by Berlage. These studies highlight the difficulty in working with teak. Indian teak is confined primarily to the peninsular region, which is below 24°N latitude. The more important

Trees (2013) 27:343–358

teak forests, however, are in central and southern parts of India (Troup 1921). Indian teak has a long history due to the intensive forest trade during India’s British rule. Dietrich Brandis, appointed as Superintendent of Forests in India in 1856, was the most influential person in forest management. Under his guidance, working of India’s forests was transformed from material exploitation to obtain timber, to valuing forests as growing biological entities that should be handled in accordance with the principles of scientific forestry (Tiwary 2003). Long-term undertaking of teak tree-ring research has been performed in India despite limited natural distribution. Brandis (ca. 1850), who investigated the association between tree-ring width and climate, first investigated Indian teak (Liese 1986). Teak from the West Ghats exhibits a growth period that spans from the end of April to the beginning of October (Chowdhury 1940). By establishing a response function (Fritts 1976), Pant and Borgaonkar (1983) and Bhattacharyya et al. (1992) found no correlation between teak tree-ring width and current rainfall, but reported a correlation between ring width and rainfall during the previous October.

Progression of dendrochronology (1995–2011) Tree-ring width measurement: conventional techniques for dendrochronology Teak Teak trees are excellent subjects for conducting dendrochronological research and extending its development. An emphasis on Java teak is relevant for various reasons. The ENSO variability uses teak chronologies from Java and neighbouring countries. D’Arrigo et al. (2008) investigated the drought history of the area using nine drought-sensitive chronologies collected from Java teak tree rings with one dataset collected from corals. Using the teak chronology data, they also reconstructed the stream flow rates for the Citarum River in Java, Indonesia for the period 1759–2006 (D’Arrigo et al. 2009). Decreases in tree-ring width paralleled the reduced stream flow recordings, which are associated with the drought caused by the ENSO-related warming events in the tropical Pacific and Indian oceans due to positive dipole-type variability. Much dendrochronological research in Thailand has focused on teak, a representative favourable tree species until 1995. Pumijumnong et al. (1995a) established a teak tree-ring study in northern Thailand, and found that teak growth is controlled primarily by premonsoon rainfall that occurs from April to June. According to reconstructed April-to-June rainfall, the previous two decades were

345

moderately wet in northern Thailand (Pumijumnong et al. 1995b). Buckley et al. (2007a) conducted a teak study in the Mae Hong Son province in Northwestern Thailand, establishing that a 448-year-old teak chronology was sensitive to summer monsoon variations. Using variability of teak tree-ring morphology data, which depends on the rainfall and soil moisture conditions, Buckley et al. concluded that teak chronology data captured two prominent periods of drought: one in the early 1700s (in agreement with coral records from the tropical Pacific), and the second in the mid-1700s (in agreement with the known history of El Nin˜o events). Archaeological studies have been recorded in Thailand for many years (Gorman 1970; Sørensen 1973). There exists little information, however, about wooden findings including those made of teak. People inhabiting ancient settlements in the Pang Mapha district used teak wood from the Mae Hong Son province in Northwestern Thailand to make log coffins. The head-forms from teak logs were collected and subjected to dendrochronological techniques to prove archaeological hypotheses. The results of these analyses have elucidated the frequency of teak-log coffin use and have aided archaeologists in obtaining information regarding ancient woodcraft techniques (Wannasri et al. 2007). This research represents the first dendroarchaeological research in Thailand. However, 14C dating remains necessary to provide estimates of coffin age. Further work to extend tree-ring examinations will benefit both archaeology and dendroclimatology. Excellent research from Shah et al. (2007) used teak tree-ring data from Hoshangabad, India to reconstruct the June–September precipitation. This study demonstrated that teak from central India exhibits a strong correlation between the tree-ring index and both regional moisture variations and rainfall during the summer monsoon months. A direct effect of temperature on teak growth was not detected. Thus, teak tree-ring chronologies can be used as a high-resolution proxy to examine past rainfall and moisture conditions in central India (Ram et al. 2008). Southern Indian teak up to 523 years old has been discovered. The ring-width variations of this teak are sensitive to the Indian summer monsoon rainfall and the SOI. Periods of low tree-ring growth have been associated with El Nin˜o events since the late eighteenth century (Borgaonkar et al. 2010). Dietrich Brandis arrived in Myanmar in 1856 to take charge of the Bago forests. He introduced ring counting, estimating tree yields for the silvicultural system. Teak was the only commercial species available at that time (Gyi and Tint 1998). Pumijumnong et al. (2001) published the first teak tree-ring chronologies from Myanmar, and this group collected data from 187 living teak trees from various forest types, the oldest of which was 165 years old.

123

346

Interestingly, the rainfall from a single month (April) determined the growth of Myanmar teak; thus, Myanmar teak exhibits a similar growth pattern as Thai teak. Kyaw (2003) investigated the phenology and growth of Myanmar teak, revealing significant correlations between tree growth and total precipitation in the rainy season (May–October) for Mabein (0.29, p [ 95 %) and Kanbalu (0.37, p [ 95 %); however, the correlation between teak ring width and rainfall at Taungoo was not significant. Recently, D’Arrigo et al. (2011) compiled data from 29 samples, including 20 living trees, spanning from 1613 to 2009. They showed that Myanmar teak growth positively correlates with rainfall around April to August and with the Palmer Drought Severity Index from May to June. Myanmar still contains abundant natural teak, which has been incorporated into the country’s architecture. This resource could potentially be used to further elucidate the region’s dendrochronology. Cook et al. (2010) compiled the Monsoon Asia Drought Atlas (MADA) using tree rings from more than 300 sites across the forested areas of Monsoon Asia (Monsoon Asia regional emphasis of the monsoon over Africa, India, East Asia and South Asia into northern Australia) to reconstruct the seasonalized Palmer Drought Severity Index (PDSI). These results represent outstanding dendroclimatology research in Southeast Asia. The lack of monsoons and the massive drought in Asia during the last millennium was demonstrated using a tree network that included 327 tree-ring chronologies (the list of contribution of tree-ring network; www.sciencemag.org/cgi/content/full/328/5977/ 486/DC1). These authors also reported a drought from 1638 to 1641 during the reign of the Ming dynasty, which interestingly parallels a drought that occurred during 1756–1768, when the severe monsoon failure coincided with the collapse of many Southeast Asian countries, namely Vietnam and Myanmar. The authors found no corresponding information in historical records; however, the drought was first discovered using teak from Northwest Thailand (Buckley et al. 2007a), and it was confirmed using tree-ring width data from Vietnamese cypress (Sano et al. 2009). In addition, Cook et al. demonstrated evidence for the East Indian Drought of 1792–1796 and the Great Drought of 1876–1878. These data represent an accurately built network of tree-ring chronologies and verify the usefulness of global climate modelling to counter the lack of long-term instrumental records from this region. To support these findings, Wahl and Morrill (2010) focused on palaoeclimate studies in a region that depends on monsoon rainfall for water and agricultural production. Developing additional tree-ring records is necessary for continued improvement of MADA. Tree-ring scientists, however, must expand spatial coverage by exploring new sources of relict wood, rare species that assure record climate. Bell et al. (2011) demonstrated ‘‘probabilistic tools’’ derived

123

Trees (2013) 27:343–358

from the MADA by Cook et al. (2010), summarizing that climate reconstructed from tree-ring data from more than 300 sites across Monsoon Asia can forecast climate signals. This has the potential to improve inter-annual decisions on Southeast Asia agriculture. Rainfall is the environmental factor that has the most influence on teak tree-ring width. This variable indeed differs between rainy and dry season, and it has been shown that the transition between dry season’s end and the beginning of the rainy season most affects teak growth. Difficulties working with teak Although teak works best of the Southeast Asian species for dendrochronology, caveats in working with teak exist. Teak grows quickly for 25–30 years, keeping growth controlled genetically with relatively little influence of external factors. Characteristically, teak stands upright and tall and its ring formations are often wavy. Working with a teak disk is easier than working with a teak core. Growth occurs in an 11-year cycle; therefore, matching the living teak ring with and without the remainder of the last ring is imperative. In general, two cores per tree are appropriate for dendrochronology, although more core samples are preferred. Ring wedging is explained by light saturation changes that occur during tree life when competition pressure from neighbouring trees changes its growth direction. This phenomenon leads to differences in the local supply of carbohydrates, water, nutrients, and phytohormones (Du¨nisch et al. 1999). Teak rings often exhibit ring wedging, sometimes interfering with sample measurements (Fig. 1). We show complicated rings with less latewood and distinct boundaries (Fig. 1d). Pine Two common pine species (Pinus merkusii and Pinus kesiya) in Southeast Asia are naturally distributed throughout India (South Asia), Myanmar, Thailand, Lao (PDR), and Kampuchea. The pine species used for dendrochronological studies in Thailand include the two-needle tree Pinus merkusii and the three-needle tree P. kesiya. The results of pine tree rings and climate responses in Thailand, however, have been inconsistent due to site-specific and non-climactic external factors affecting pine growth. Buckley et al. (1995) established four pine chronologies for P. kesiya and P. merkusii from Northern and Northwestern Thailand: three dendrochronology studies were performed on P. kesiya, and one study focused on P. merkusii. Results revealed that the tree-ring index of P. kesiya at the Nam Nao site (Phechabun province) correlated positively with rainfall and negatively with November temperature. Interestingly, the tree-ring index of P. merkusii at SLM (Thung Salaeng Luang, Phitsanulok province) exhibited a strong negative

Trees (2013) 27:343–358

347

Fig. 1 Photographs of teak features. a A teak disk, b teak tree-ring width, c teak surface with wavy rings, and d the complicated ring. Source: Pumijumnong (1995)

correlation with precipitation throughout much of the year. D’Arrigo et al. (1997) performed dendroclimatic studies on mountain pines in northern Thailand and reported that the oldest Thai pine (P. merkusii) from Phu Kradung existed from 1647 to 1993. The lowest growth rate observed for P. kesiya occurred in 1979 due to severe drought that year. Buckley et al. (2005) used pine chronologies from Thailand, India, and Bhutan to examine the relationship between pine tree-ring widths and Asian monsoons. Grid data for the tropical Indian and Pacific Oceans showed ring widths for Thai and Indian pine (both P. merkusii) that strongly correlated with the tropical climate of the Indian and Pacific Oceans, whereas the ring widths in the Bhutan pine, P. wallichiana, showed strongest association with climates of the northern Pacific and Asian landmasses. Buckley et al. (2005) found that this difference in findings arose from meteorological data used in their calculations. The Thai meteorological data were obtained from a weather station located at lower latitude than the area in which the pine samples were collected. Pumijumnong and Eckstein (2010) reconstructed the pre-monsoon weather conditions in Northwest Thailand using the pine tree-ring network of P. merkusii and

P. kesiya. These pine tree chronologies were closely associated with temperatures from March through May, and the pine tree-ring network accurately illustrates the warm/dry and cool/wet periods from 1834. Buckley et al. (2007b) investigated the growth rings of P. merkusii from Laos (Lao Peoples Democratic Republic) using three parameters: tree-ring width, early wood width, and late wood width, correlating these parameters with climate data from 13 nearby weather stations in Thailand. All three parameters exhibited significant negative correlation with prior June rainfall and positive correlation with August maximum temperature from the previous year. These three indices also correlated with gridded sea surface temperature in the central and eastern tropical Pacific. Buckley et al. suggested that further studies are necessary to determine the relationship between the pine tree-ring parameters and local climate. Ideally, these future studies should use principal component analysis of pine tree-ring networks including larger regions. With limited dendroecology research that has been conducted in Southeast Asia, Zimmer and Baker (2009) used reconstructed pine recruitment histories to examine forest dynamics of northern Thailand. They found forest

123

348

recruitment data synchronized significantly with the stand, site, and region, but the correlation between recruitment and climate was statistically insignificant. Comparing recruitment and climate data indicated a complex intercorrelation among local disturbances and environment. Pine tree-ring width exhibits great potential for dendroclimatology and climate reconstruction in Southeast Asia. Pine wood produced much resin, so much that resin penetrated the growth ring boundary when working with pine samples. Moreover, local people in remote areas have burned pine tree for resin, which affects ring width formation and occasionally causes false rings. Natural fire occurs often in Southeast Asia’s dry season and remains an additional influence on pine growth, reflecting in tree-ring width. Results show that the correlation between pine ring width and climate data is heterogenic between sites, and phenology and pine growth analysis are both needed in the future. Cypress The most recent species used in dendrochronology research in Southeast Asia is the cypress Fokienia hodginsii. This cypress has a long life span, and is naturally distributed in humid, montane forests at 600–2,200 m asl in Southern China, Northern Lao PDR, and Vietnam. Currently, this tree’s existence is threatened globally (Truyen and Osborn 2006). Tree ring-based hydroclimate reconstructions for Southeast Asia were recently published for Vietnam. Sano et al. (2009) studied Fokienia hodginsii, a rare conifer from northern Vietnam. Tree rings from this study spanned 535 years, and the authors determined that the annual growth of Fokienia was controlled primarily by soil moisture content in the pre-monsoon season. This reconstruction also revealed two prominent periods of drought in the mid-eighteenth and late-nineteenth centuries. Buckley et al. (2010) used ring width records for F. hodginsii growing at two sites in the highlands of Vietnam’s Bidoup Nui Ba national park to reconstruct a 759-year history (1250–2008 AD) of the early monsoon (March through May) (Palmer Drought Severity Index). This study revealed weaker monsoons in the mid to late fourteenth century, and a shorter, more severe drought in the early fifteenth century coincided with Angkor’s eventual demise. Xu et al. (2011) developed tree-ring cellulose d18O chronology of Fokienia hodginsii in Northern Laos from 1951 to 2002. Their results indicate that tree-ring cellulose d18O correlated negatively with May–October regional PDSI, but matched the Multivariate ENSO Index (MEI). Cypress is limited to high elevation areas, but its ring is well captured with PDSI; therefore, forest logging should be prohibited and preservation initiatives for this species should be instated.

123

Trees (2013) 27:343–358

Wood anatomical features and monitoring of cambial activity Teak The study of tropical trees is complex because these trees lack the normal appearance of annual rings. Eckstein et al. (1981) summarized the first international conference on the age and growth of tropical trees as ‘‘a new direction for research.’’ Several authors have suggested that observing the phenology and periodicity of cambial activity in tree species is useful for dating tropical species (i.e., Waisel and Fahn 1965; Mariaux 1976; Borchert 1980; Fahn et al. 1981). Wood anatomy investigations have also determined that observing and monitoring cambium activity is a useful method for studying the factors that govern cell differentiation. Rao and Dave (1981) investigated 12-year-old teak trees growing in a botanical garden in India, determining that cambial activity begins at the beginning of the rainy season in June. Venugopal and Krishnamurthy (1987) established that cambial activity in Indian teak begins 1 month before new leaves bud; however, this research was conducted on branches and not on stems. The initiation of cambial activity in teak trees in northern Thailand was investigated using five teak trees from six sites in a 12-month observational study from October 1993 to October 1994. It was discerned that cambial growth begins during the first half of May and ends in October. The phenological observations showed that the cambial activity of teak trees in northern Thailand is initiated shortly before the burst of new leaves. Leaf falling occurs December to March depending on location (Pumijumnong 1995). In addition to tree-ring widths, vessel diameters are used to identify climate data correlations of teak trees (Pumijumnong and Park 1999). Early results revealed vessel parameters in total rings in early wood correlated negatively with rainfall during the transition period between the dry and wet seasons. Further, vessel parameters were used to reconstruct pre-monsoon temperatures, and latewood vessel parameters exhibited a negative correlation with June temperatures (Pumijumnong and Park 2001). The climatic effects on vessel parameters and tree-ring widths are different. Bhattacharyya et al. (2007) studied teak vessels in India and found that the October and November climate of the previous year and April climate of the current year is especially important for early wood vessel formation. Pine Linasmita (2004) studied the cambium activity of P. kesiya in the Khun Tan National Park, Lampun province, Thailand in 2002 and 2003. Cambium activity started in the beginning of rainy season (May), reaching its peak in

Trees (2013) 27:343–358

August, declining slightly in the following year, and ceasing around December to January (beginning of dry season). This study showed that the cambium layer of P. kesiya displayed a significant positive correlation with humidity (0.475) and rainfall (0.676), whereas temperature had no effect on activity (R = 0.289). Pumijumnong and Wanyaphet (2006) investigated the cambial activity in two species of Thai pines (P. merkusii and P. kesiya). The onset of rain in May reactivated cambial divisions until peaks in September (P. merkusii) and October (P. kesiya). Cambial cell layer numbers decreased from November to February, and all trees were still dormant in March and April. Thus, soil moisture influenced both tree species; however, rainfall and temperature did not have a significant direct effect on the cambial activity in either species.

349

These were not clearly related to the rainy and dry season change. No dormancy of this cambium was observed. The coastal zone in Thailand is approximately 2,600 km long, and the mangrove tree species is abundant. There remains little information regarding their growth and age. Buajan (2010) investigated the relationship between the cambial activity of Rhizophora mucronata, Avicennia alba, and Avicennia marina and climatic factors (e.g., rainfall and temperature) in the Samut Sakhon province of Thailand. The results revealed that correlations between the cambial activity of these three species and the rainfall, temperature, and salinity were weak and statistically insignificant. It can be concluded that the cambial activity in these three species is not driven by the seasonal climatic factors. For A. alba and A. marina, more than one cambium was found to be simultaneously active.

Other species Relatively few species of trees growing in the tropical– subtropical regions of Southeast Asia have been selected for cambium dynamics examination. For example, Suwanpatra (2006) investigated the cambial dynamics of Podocarpus neriifolius, an evergreen species from the family Podocapaceae. Seasonal variations in the cambial activity were observed to ultimately investigate the relationship between cambial activity and environmental factors of precipitation, temperature, and soil moisture. The study area was located in Khao Keaw Peak, Khao Yai National Park, where cambial zones of 10 Podocarpus trees were sampled once per month from February 2005 to January 2006. Research revealed that the cambial zone of P. neriifolius remained active throughout the year, including four growth flushes with three mid-season growth pauses between each growth period. Correlation analysis demonstrated that cambial activity did not associate with climatic factors examined. Regression analysis showed only 13.80 % of activity was influenced by factors mentioned. Moreover, macroscopic investigation of wood defined three types of growth boundaries: narrow, wedging, and wide. The cambial activity was only monitored for a year (Figs. 2, 3). Buckley et al. (1995) suggested that Podocarpus has potential for dendrochronological research in Thailand. Poussart et al. (2004) examined the rhythms of Podocarpus neriifolius growth bands in Doi Inthanon National Park (northern Thailand). Stable isotopic methods revealed that these growth bands were annual, which counters the findings of Suwanpatra (2006). Fujii et al. (1999) demonstrated seasonal changes affecting the zone of differing xylem of three Shorea species (Shorea patoiensis Ash., S. pinanga Scheff. and S. dasyphylla Foxw. in Family Dipteracarpaceae) growing in Sarawak using thin-sectioning methods. They explained that seasonal changes of xylem development were visible in the width of secondary wall zone rather than of the primary.

Study of the age and growth of ringless species in subtropical forests Killman and Thong (1995) reviewed the periodicity of growth in tropical trees in Malaysia, with special reference to Dipterocarpacea. Most trees in seasonal tropical climates show periodically recurring intervals of rest, demonstrating a rhythmic growth that is possibly a function of varying amounts of precipitation. Sass et al. (1995) investigated wood formation in Dryobalanops sumatrensis and Shorea leprosular, which are naturally distributed in western Malaysia. Results revealed that wood formation appears to be a continuous process and is not related to seasonality in rainfall and phenology. The Huai Kha Khaeng permanent forest research plot located in Western Thailand is part of a network of longterm, large-scale study plots in tropical and subtropical seasonal evergreen forests. It was established to stimulate silviculture research for the purpose of conserving biodiversity. The respective research in this area has focused on gap modelling and forest dynamics (Bunyavejchewin et al. 2009). Baker et al. (2005) combined the direct estimates of long-term establishment and growth patterns of 12 tree species forming annual rings with indirect estimates of growth patterns using stand structure analysis, individual trees architecture, and age-modelling approaching. The authors summarized the complex history that took place in this area over 250 years through to the mid-1800s, and this approach uncovered widespread disturbances that occurred at least three times in the 1910s, 1940s, and 1960s. Williams and Bunyavejchewin (2008) monitored the phenology of hundreds of tree species in Huai Kha Khaeng, revealing that foliar phenology is controlled by the intraannual patterns of dry–wet conditions. However, for a better understanding of tree phenology, the complex interactions between species-specific physiologies, local

123

350

Trees (2013) 27:343–358

Fig. 2 Photographs of the Podocarpus disk. a A whole-trunk cross section of P. neriifolius, 54.5 cm in girth. The darker area on the left is the underside of the felled stem. b The inner part of the section

shows ordinary ring boundaries. c The transitional area from ordinary growth (bottom of the picture) to anomalous growth (top of the picture). Scale bar: a 20 mm; b, c 5 mm. Source: Suwanpatra (2006)

edaphic conditions, and intra-annual variations rainfall timing must be discerned. Further studies targeting specific species for detailed observations of physiology and microclimate are required. Additional species in Southeast Asia were investigated as potential subjects for continuing dendrochronological research. The final report for the Asia–Pacific-Network for Global Change Research, consisting of researchers from the Philippines, Malaysia, Thailand, Sri Lanka, India, and Vietnam, has listed the promising species in each of these countries that have potential for dendrochronological

research. The report suggests that if these countries undertake future research, a greater understanding of tree growth in Southeast Asia will be achieved, ultimately benefiting sustainable forest management and biodiversity (Asia–Pacific Network for Global Change Research 2010).

123

Application of isotopic techniques for Southeast Asian trees Measurements of stable oxygen and carbon isotope ratios in tree rings have been widely used as a valuable tool to

Trees (2013) 27:343–358

351

Fig. 3 Photographs of all three types of defined growth boundaries of Podocarpus neriifolius. b Microscopic view of the narrow boundaries showing persistent ring boundaries. In radial files, latewood-like cells were few but prominent. c The convergent or wedging type of growth

boundaries showing progressively narrower rings until two latewoodlike cell rows fused together. d The wide type of growth boundaries (rotated). Scale bar: a 5 mm; b–d 200 lm. Source: Suwanpatra (2006)

study climate variability, specifically for estimating the growth and age of tropical trees. Poussart et al. (2004) used these techniques to prove that seasonality occurs in tropical trees from Indonesia and Thailand. Teak (from a site in Saradan, central-eastern Java), Podocarpus neriiferius (from Doi Inthanon National Park, Chiang Mai province, north Thailand), and Samanea saman (from a lumberyard in Bali) were chosen and d18O, d13C, and bomb radiocarbon levels in cellulose were extracted from the wood. Results for Java teak trees revealed that the d13C cycle followed a more consistent annual rhythm than d18O. The authors also observed that variation in d18O throughout the seasonal cycle was two to three times greater than the variation of d13C, indicating that d18O levels are noisy. The Thai tree, P. neriifolius, showed six cycles of d13C and d18O incorporation, corresponding to when the tree was

wounded and cut. For this tree, it was concluded that most growth occurs during the wet season. These trees are excellent candidates for isotope dendrochronology. Using isotopic techniques, Poussart and Schrag (2005) examined 11 trees that grow naturally in northern Thailand including Quercus kerrii, Miliusa velutina, Dipterocarpus tuberculatus, Dipterocarpus obtusifolius, Pterocarpus macrocarpus (2 trees), Lagerstoemia cochinchinensis, Chukrasia velutina (2 trees), Lannea coromandelica, and Shorea obtusa. They concluded that Quercus kerrii and Miliusa velutina are two excellent candidates for establishing isotope chronologies because they have the most regular d13C and d18O cycles. Additional species including Pterocarpus macrocarpus, Dipterocarpus tuberculatus, Lagerstoemia cochinchinensis and Shorea obtuse exhibited cycles only in their d18O records. From d18O and d13C

123

352

records, the authors concluded that Miliusa velutina, Quercus kerrii, and Dipterocarpus tuberculatus produced ring boundaries at the end of the wet season. Only Lannea coromandelica, Dipterocarpus obtusifolius, and Chukrasia velutina lacked a seasonal isotope signal. Poussart et al. (2006) used the first X-ray microprobe synchrotron record of calcium (Ca) from a ringless Miliusa velutina tree from Thailand to estimate the tree’s age and growth pattern. The results revealed that measured calcium cycle yield annual growth estimates agree with the previous study (Poussart and Schrag 2005). Using vessel anatomy and carbon isotope measurements, Ohashi et al. (2009) investigated Dipterocarpaceae species in Thailand (Shorea, Dipterocarpus, Hopea) as well as teak, concluding that the vessel anatomy assessment for tropical dendrochronological research is simpler and more cost-effective than isotope measurements. A combination of tree-ring measurements and carbon and oxygen stable isotope determinations was used to demonstrate the intrinsic water-use efficiency in three species (Chukrasia tabularis, Melia azedarach and Toona ciliata) (Nock et al. 2010). Interestingly, CO2 levels usually result in more efficient water use, associated with increased growth; this research, however, revealed a decreased growth in the studied tree species. The authors hypothesized that changes in growth reflected water stress, warming, and more frequent and severe El Nin˜o effects (Nock et al. 2010). Recent research investigated the levels of d18O in tree-ring cellulose samples from the Fokienia hodginsii species in northern Laos, and these results correlate with the PDSI. This study represents the first attempt to examine cellulose preserved in tree rings, which has been found to have advantage over dendrochronology for research in Laos PDR (Xu et al. 2011). Managave et al. (2010, 2011) measured the oxygen isotopes in teak wood cellulose from southern India and determined that higher oxygen isotope values are found at the ring boundary and lower values are found in intermediate parts. Managave et al. (2011) discovered a negative correlation between the amount of rainfall and the cellulose d18O levels. This relationship was used to reconstruct local precipitation data since 1743. In Malaysia, Loader et al. (2011) used stable carbon isotopes to analyse the sequential radial wood increments of three tree species (Eusideroxylon zwageri, Shorea johorensis and Shorea superba) from 1850 to 2009. Because of the seasonal equatorial climate, conventional dendrochronology cannot be used to date these ringless trees. By introducing isotopic techniques, this group established the ages of pith/central wood (±1 sigma) of the Eusideroxylon zwageri to be 670 ± 40 years old, whereas the less dense Shorea johorensis and the Shorea superba yielded ages of 240 ± 40 and 330 ± 40 years, respectively. This demonstrates the

123

Trees (2013) 27:343–358

potential of this technique for assessing the impact of climate change on ringless trees. The few tree species in subtropical and tropical forests have shown clear annual rings. Dendrochronological research thus begins with trees with a proven annual ring. By observing and monitoring the physiology and cambial activity of trees, coupled with the most sophisticated technology, understanding the relationship between tropical and subtropical trees and environmental factors is supported. The most recent species used in dendrochronology is the Vietnamese and Laos cypress. Table 1 shows the summary of the tree-ring index and response.

Future perspectives The emphasis on dendrochronology in Southeast Asia has concentrated on teak trees, which hold high potential for dendrochronological research. The largest standing living teak tree is found in the Uttaradite province northeast of Thailand, with other single trees found in other areas of Thailand. Throughout the country, many village residences, old palaces, and buildings were constructed using teak wood. Problems arise when collecting samples from these buildings, as owners do not want their buildings drilled. Samples from houses that granted access to their wood were even problematic, as remaining ring widths are too short (the outer part of the wood was removed during the construction of the building) (Fig. 4a–c). If core samples are small or short, the risk of miscalculating the crossdate is large. To further study these samples, isotope or wiggle techniques are recommended. Myanmar is the only country in Southeast Asia that still exports teak logs to other countries. Naturally growing teak trees and plantation teak trees are still plentiful in Myanmar, and teak can be found in modern and old wooden buildings, such as the Golden Palace in Bago (Fig. 4d, e). These samples have not yet been taken for dendrochronological research. During the past two decades, Southeast Asian dendrochronology has been in intensive development. Initially, studies were dedicated to unearthing the age and growth of tropical trees. Foreign scholars like Coster (1927, 1928) and Berlage (1931) introduced the basic concepts of dendrochronology into Southeast Asia. Hughes (2002) published a review of dendrochronological research, and his ideas and comments are valuable and beneficial for further research. Rozendaal and Zuidema (2011) recently published a general review of tropical dendroecology and provide intensive critiques concerning dendrochronology in climatology. They also expressed the weakness and strengths of this field, ultimately reinforcing the findings we obtained in our study.

Trees (2013) 27:343–358

353

Table 1 Summary of the tree-ring index and response Tree species

Location

Time span

Parameter

Response

Sources

Fokienia hodginsii

Vietnam

1250–2008

Ring width

March–May palmer drought severity index

Buckley et al. (2010)

Fokienia hodginsii

Vietnam

1470–2004

Ring width

March–May Palmer Drought severity index

Sano et al. (2009)

Fokienia hodginsii

Lao PDR

1951–2002

Cellulose d18O

May–October palmer drought severity index

Xu et al. (2011)

Tectona grandis

Indonesia, C. Java

1514–1929

Ring width

West Java rainfall (rainfall during the dry season), June–September, sea level pressure

De Boer (1951)

Tectona grandis

Indonesia, Java, Muna

1564–1995, 1673–2005

Ring width

Palmer drought severity index for JAVA, ENSO, stream flow

D’Arrigo et al. (2006, 2011), Bijaksana et al. (2007)

Tectona grandis

Thailand, Maehongson

1558–2005

Ring width

Rainfall, Palmer drought severity index (PDSI)

Palakit (2004), Buckley et al. (2007a, b)

Tectona grandis

Thailand, Phrae

1880–1990

Ring width

April–June rainfall

Pumijumnong et al. (1995a, b)

Tectona grandis

Thailand, Maehongson

1947–1996

Vessel parameters

Tem. April–May

Pumijumnon and Park (1999)

Tectona grandis

C. India, S. India

1654–2001, 1481–2003

Ring width

Annual rainfall (C. India), S. India

Borgaonkar et al. (2007, 2010)

Tectona grandis

C. India

1835–1997

Ring width

June–September rainfall

Shah et al. (2007)

Tectona grandis

S. India

1743–1986

Early wood vessel area

Oct–Nov rainfall of previous year and April rainfall of current year

Bhattacharyya et al. (2007)

Tectona grandis

C. India

1827–2001

Ring width

Rainfall and moisture index

Ram et al. (2008)

Tectona grandis

W. India, S. India

1743–1988

Cellulose oxygen isotope

Rainfall

Managave et al. (2011)

Tectona grandis

Myanmar, Mandalay

1834–1998

Ring width

April rainfall

Pumijumnong et al. (2001)

Tectona grandis

Taungoo, Myanmar

1880–2000

Ring width

May–October rainfall

Kyaw (2003)

Tectona grandis

Mabein, Myanmar

1875–2000

Ring–width

May–October rainfall

Tectona grandis

Kanbalu, Myanmar

1865–2000

Ring width

May–October rainfall

Tectona grandis

Maingtha Forest Reserve, Myanmar

1613–2009

Ring width

April–August rainfall, May–June PDSI

D’Arrigo et al. (2011)

Pinus merkusii

NW Thailand

1820–2002

Ring width

March–May temperature; May rainfall

Pumijumnong and Eckstein (2010), Pumijumnong and Wanyaphet (2006)

Pinus kesiya

NW Thailand

1690–2003

Ring width

March–May temperature; May rainfall

Pumijumnong and Eckstein (2010), Pumijumnong and Wanyaphet (2006)

Pinus kesiya

Namnao, Thailand

1789–1994

Ring width

Nov. rainfall

Buckley et al. (1995)

Pinus merkusii

Thung-SalaengLung, Thailand

1835–1992

Ring width

Neg. rainfall through much of year

Buckley et al. (1995)

Pinus kesiya

Shilong, Meghalaya, NE India

1859–2000

Ring width

Prior December rainfall, March rainfall

Chaudhary and Bhattacharyya (2002)

Pinus merkusii

Lao PDR

1743–2005

Ring width, earlywood width, latewood-width

Prior June rainfall, August– September temperature

Buckley et al. (2007a, b)

123

354

Trees (2013) 27:343–358

Fig. 4 Photographs of the potential extended teak chronology in Southeast Asia. a Teak log coffin, Thailand. b Log coffin at Chiang Dao, Thailand. c Teak house at Saiem, Phare, Thailand.

d Bago archaeological site, Myanmar. e Storage teak house at Bago archaeological site, Myanmar. Source: by Author

We speculate that a connection exists between ring width and climate data. Using tree-ring width to reconstruct past climates is complicated by monsoon periods. Some tree-ring width data indicate short periods of growth, which could indicate varying climate conditions. Further error is introduced due to the location of the meteorological station, which is at lower elevation than the tree-ring study sites. Due to continuous logging, most living trees in Southeast Asia are not old enough to be included in a statistically

sound study. Due to insufficient availability of raw materials (complete samples/mature trees), material from old buildings or archaeological sites has been used to compensate and extend climate data chronology. Because different types of trees have their own unique growth signatures, the source/species must be identified when age is known. Teak growth shows a rhythmic pattern of approximately 11 years; therefore, if teak is used for extending a chronology, an overlap period of at least 60 years is recommended.

123

Trees (2013) 27:343–358

Anatomical wood measurements and cambial dynamics observations are essential. These requirements can be reduced during a study to alleviate costs, and due to time involved, can only be applied for a short period. It is recommended that a precise, detailed anatomical observation and understanding of tree physiology techniques specific to this region be applied. This analysis should be combined with research of proxy records, which is recommended to confirm the results of our study. This review of Southeast Asia and the future challenges of dendrochronological research emphasizes the following: 1.

2.

3.

4.

The intensive investigation of long-term tree physiology, and the anatomy of trees should be applied to individual trees in the same region or to the same species across the region. Climate change impact on trees is needed. Trees adapt to survive, but with abnormal climate patterns, trees are subjected to stress that manifests years later. Recruitment or regeneration, and abrupt or gradual change may affect tree seedlings. Failure to understand the interaction of climate change and tree physiology may result in the extinction of many tree species, and the tropical forests in Southeast Asia would suffer from this dearth of understanding, a phenomenon already occurring in many places around the world. Understanding the correlation between the warm pool in the Indian Ocean and ENSO, which influences the monsoon variation, is urgently needed. The development of the extended teak index in Southeast Asia by cooperative research is promising for the future of Southeast Asian forests. In addition to conventional dendrochronological techniques, incorporating other methods is required to continue this research.

Conclusion Dendrochronological research is under promising development in Southeast Asia, where investigators have used dendrochronological techniques to explore the paleoclimate and tree growth patterns. Most Southeast Asian countries are influenced by a monsoon climate, conditions that are conducive to dendrochronological research in only a few countries. In these areas, trees grow year-round, and clearly demarcated annual rings are not observed. Dendrochronology, in these cases, is only effective when used in the subtropical zone, where the hydroclimate is seasonal. Despite difficulty, there exists an urgent need to understand climate change and variability. Various methods have been employed, from traditional tree-ring dating techniques to complicated and highly

355

technical applications. Because these endeavours are nascent, academic conferences dedicated to dendrochronology will be necessary to transfer knowledge, connect likeminded researchers, and facilitate further research. Dendrochronology has the potential to influence related fields such as archaeology or dendroecology (e.g., in the case of an insect outbreak in a teak forest). Teak is of particular interest, and other tree species may become informative with novel technique development, such as isotope analysis and cambium activity. In fact, climatology will require this type of technological innovation. It remains difficult to extend analysis to the very distant past with currently available technologies. Wood anatomy measurements represent a relatively easy and uncomplicated method that can be adapted by Southeast Asian countries to explore dendrochronology; invisible annual growth boundaries, however, may require additional expensive methods. An obstacle in these studies is the dearth of financial support from external and internal sources. There is a lack of interest in sustainable ecosystems, particularly for the role of the forest service. With climate change-impact issues and environmental problems gaining popularity and recognition worldwide, officials may be inspired to take greater interest in the restoration of forest ecosystems and in understanding of their role in the global carbon cycle. Acknowledgments The authors wish to express their thanks for the support received from Mahidol University. We are deeply thankful to Prof. Dr. Achim Bra¨uning, Friedrich-Alexander Universita¨t ErlangenNu¨rnberg, Germany, for the critique of our review. We also thank the anonymous referees for their valuable input and criticism regarding the manuscript.

References Asia-Pacific network for global change research (APN) (2010) Collaborative studies in tropical Asian dendrochronology: addressing challenges in climatology and forest ecology. Final report for APN project—ARCP2008-03CMY-Baguinon. AsiaPacific network for global change research, 48 p Baker PJ, Bunyavejchewin S, Oliver CD, Ashton PS (2005) Disturbance history and historical stand dynamics of a seasonal tropical forest in western Thailand. Ecol Monogr 75(3):317–343 Bass P, Vetter RE (eds) (1989) Growth rings in tropical woods. IAWA Bull 10:95–174 Bell AR, Cook BI, Anchukaitis KJ, Buckley BM, Cook ER (2011) Repurposing climate reconstructions for drought prediction in Southeast Asia. Climatic Change 106:691–698 Berlage HP (1931) Over het verband tusschen de dikte der jaarringgen van Djatiboomen (Tectona grandis L.f.) en den regenval op Java. About the relationship between annual ring width of Djati trees (Tectona grandis L.f.) and rainfall on Java. Tectona 24:939–953 Bhattacharyya A, Yadav RR, Borgaonkar HP, Pant GB (1992) Growth-ring analysis of Indian tropical trees: dendroclimatic potential. Curr Sci 62:736–741 Bhattacharyya A, Eckstein D, Shah SK, Chaudhary V (2007) Analyses of climatic changes around Perambikulum, South

123

356 India, based on early wood mean vessel area of teak. Curr Sci 93(8):1159–1164 Bijaksana S, Ngkoimani LO, D’Arrigo R, Krusic P, Palmer J, Sakulich J, Zulaikah S (2007) Status of tree-ring research from teak (Tectona grandis) for climate studies. J Geofisika 2:1–7 Borchert R (1980) Phenology and ecophysiology of tropical trees: Erythrina poeppigiana O.F. COOK. Ecology 61(5):1065–1074 Borgaonkar HP, Sikder AB, Ram S, Kumar KR, Pant GB (2007) Dendroclimatological investigations of high altitude Himalayan conifers and tropical teak in India. Korean J Quart Res 21(1):15–25 Borgaonkar HP, Sikder AB, Ram S, Pant GB (2010) El Nin˜o and related monsoon drought signals in 523-year-long ring width records of teak (Tectona grandis L.F.) trees from south India. Palaeogeogr Palaeoclimatol Palaeoecol 285:74–84 Bormann FH, Berlyn G (eds) (1980) Age and growth rate of tropical trees. New directions for research. School of Forest and Environmental studies. Bull 94. Yale University, New Haven Buajan S (2010) Ecological system of mangrove forest: cambial activity, soil property and distribution of heavy metals in mangrove forest at Samut Sakhon province. Unpublished master’s thesis, Mahidol University, Nakhon Pathom, Thailand Buckley BM, Barbetti M, Watanasak M, D’Arrigo RD, Boonchirdchoo S, Sarutanon S (1995) Dendrochronological investigations in Thailand. IAWA J 16(4):393–409 Buckley BM, Cook ER, Bhattacharyya A, Dukpa D, Chaudhary V (2005) Global surface temperature signals in pine ring-width chronologies from southern monsoon Asia. Geophys Res Lett 32(1–4):L20704. doi:10.1029/2005GL023745 Buckley BM, Palakit K, Duangsathaporn K, Sanguantham P, Prasomsin P (2007a) Decadal scale droughts over northwestern Thailand over the past 448 years: links to the tropical Pacific and Indian Ocean sectors. Clim Dyn 29(1):63–71 Buckley BM, Duangsathaporn K, Palakit K, Butler S, Syhapanya V, Xaybouangeun N (2007b) Analyses of growth rings of Pinus merkusii from Lao P.D.R. For Ecol Manage 253:120–127 Buckley BM, Anchukaitis KJ, Penny D, Fletcher R, Cook ER, Sano M, Nam LC, Wichienkeeo A, Minh TT, Hong TM (2010) Climate as a contributing factor in the demise of Angkor, Cambodia. PNAS 107:6748–6752 Bunyavejchewin S, LaFrankie JV, Baker PJ, Davies SJ, Ashton PS (2009) Forest Trees of Huai Kha Khaeng Wildlife Sanctuary, Thailand. National Parks, Wildlife and Plant Conservation Department, Thailand Chaudhary V, Bhattacharyya A (2002) Suitability of Pinus kesiya in Shillong, Meghalaya for tree-ring analyses. Curr Sci 83:1010–1015 Chowdhury KA (1940) The formation of growth rings in Indian trees. Part III, vol II, No. 3, Indian Forest Records Cook ER, Anchukaitis KJ, Buckley BM, D’Arrigo RD, Jacoby GC, Wright WE (2010) Asian Monsoon failure and megadrought during the last millennium. Science 328:486–489 Coster C (1927) Zur Anatomie und Physiologie der Zuwachszonenund Jahresringbildung in den Tropen. Ann Jard Bot Buitenzorg 37:49–161 Coster C (1928) Einiges u¨ber das Dickenwachstum und die Inhaltsstoffe des Djatistammes, Tectona grandis L.f. Tectona 17:1056–1057 D’Arrigo R, Jacoby GC, Krusic PJ (1994) Progress in Dendroclimatic studies in Indonesia. TAO 5(3):349–363 D’Arrigo R, Barbetti M, Wattanasak M, Buckley B, Krusic P, Boonchirdchoo S (1997) Progress in dendroclimatic studies of mountain pine in northern Thailand. IAWA J 18(4):433–444 D’Arrigo RD, Wilson R, Palmer J, Krusic PJ, Curtis A, Sakulich J, Bijaksana S, Zulaikah S, Ngkoimani LO (2006) Monsoon drought over Java, Indonesia, during the past two centuries. Geophys Res Lett 33, L04709. doi:10.1029/2005GL025465

123

Trees (2013) 27:343–358 D’Arrigo R, Allan R, Wilson R, Palmer J, Sakulich J, Smerdon JE, Bijaksana S, Ngkoimani LO (2008) Pacific and Indian Ocean climate signals in a tree-ring record of Java monsoon drought. Int J Climatol 28:1889–1901. doi:10.1002/joc.1679 D’Arrigo R, Abram N, Ummenhofer C, Palmer J, Mudelsee M (2009) Reconstructed stream flow for Citarum river, Java, Indonesia: linkages to tropical climate dynamics. Clim Dyn. doi:10.1007/ s00382-009-0717-2 D’Arrigo R, Palmer J, Ummenhofer CC, Kyaw NN, Krusic P (2011) Three centuries of Myanmar monsoon climate variability inferred from teak tree rings. Geophys Res Lett 38:L24705. doi: 10.1029/2011GL049927 Dahms KG (1989) Das Holzportrait Teak (Tectona grandis L.f., Familie Verbenaceen). Holz als Roh-und Werkstoff 47:81–85 De Boer HJ (1951) Tree-ring measurements and weather fluctuations in Java from A.D. 1514. Kon Ned Akad wetensch 54:194–209 Douglass AE (1914) A method of estimating rainfall by the growth of trees. In: The climatic factors. Carnegie Institution, Washington, DC Du¨nisch O, Bauch J, Sack M, Mu¨ller M (1999) Growth dynamics in wood formation of plantation-grown Swietenia macrophylla King and Carapa guianensis. Aubl Mitteilungen der Bundesforschungsanstalt fu¨r Forst- und Holzwirtschaft, Hamburg 193:79–96 Eckstein D, Ogden J, Jacoby GC, Ash J (1981) Age and growth rate determination in tropical trees: the application of dendrochronological methods, In: Bormann FH, Berlyn G (eds) Age and growth rate of tropical trees. School of Forestry and Environmental Studies, Bull. 94, Yale Univ., New Haven, pp 83–106 Eckstein D, Sass U, Baas P (eds) (1995) Growth periodicity in tropical trees. IAWA J 16:323–442 Fahn A, Burley J, Longman KA, Mariaux A, Tomlingson PB (1981) Possible contributions of wood anatomy to the determination of the age of tropical trees. In: Bormann FH, Berlyn G (eds) Age and growth rate of tropical trees. School of Forestry and Environmental Studies, Bull. 94, Yale University, New Haven, pp 31–54 Fritts HC (1976) Tree rings and climate. Academic Press, London, p 567 Fujii T, Salang AT, Fujiwara T (1999) Growth periodicity in relation to the xylem development in three Shorea spp. (Dipterocarpaceae) growing in Sarawak. In: Wimmer R, Vetter RE (eds) Treering analysis: biological, methodological, and environmental aspects. CABI Publishing, Oxon, UK, pp 169–183 Geiger F (1915) Anatomische Untersuchungen u¨ber die Jahresringbildung von Tectona grandis. Jahrb f Wiss Botanik 55:522–607 Gorman CF (1970) Excavation at spirit Cave, north Thailand: some interim interpretation. Asian Perspect 13:79–107 Gyi K, Tint K (1998) Natural teak forest management practiced under the Myanmar selection system. http://www.fao.org. Accessed 1 Nov 2011 Gyi K, Tint K (1985) Management of natural teak forests in teak for future. In: Proceeding of the second regional seminar on Teak, 249 p. http://www.fao.org/DOCREP/005/AC773E/AC773E00. HTM Huber B (1941) Aufbau einer mitteleuropa¨ischen Jahrringchronologie. Mitt Akad Dtsch Forstwiss 1:110–125 Hughes MK (2002) Dendrochronology in climatology-the state of the art. Dendrochronologia 20(1–2):95–116 Jacoby GC, D’Arrigo R (1990) Teak (Tectona grandis L.f.), a tropical species of large-scale dendroclimatic potential. Dendrochronologia 8:83–98 Killman W, Thong HL (1995) The periodicity of growth in tropical trees with special reference to Dipterocarpaceae—a review. IWAW J 16(4):329–335 Kyaw NN (2003) Site influence on growth and phenotype of teak (Tectona grandis Linn.f.) in natural forests of Myanmar. Dissertation, University of Go¨ttingen, 163 p

Trees (2013) 27:343–358 Lamprecht H (1986) Waldbau in den Tropen. Parey-Verl, HamburgBerlin, pp 287–290 Liese W (1986) To the memory of Sir Dietrich Brandis. Indian For 112:639–644 Linasmita V (2004) Seasonal variation on cambial activity of Pinus kesiya in Doi Khun Tan National Park, Thailand. Unpublished master’s thesis, Mahidol University. Nakhon Pathom, Thailand Loader NJ, Walsh RPD, Robertson I, Bidin K, Ong RC, Reynolds G, McCarroll D, Gegan M, Young GHF (2011) Recent trend in the intrinsic water-use efficiency of ringless rainforest trees in Borneo. Phi Trans R Soc B 366(1582):3330–3339 Managave SR, Sheshshayee MS, Borgaonka HP, Ramesh R (2010) Intra-annual oxygen isotope variations in central Indian teak cellulose: possibility of improved resolution for past monsoon reconstruction. Curr Sci 98(7):930–937 Managave SR, Sheshshayee MS, Ramesh R, Borgaonkar HP, Shah SK, Bhattacharyya A (2011) Response of cellulose oxygen isotope values of teak trees in differing monsoon environments to monsoon rainfall. Dendrochronologia 29:89–97 Mariaux A (1967) Tree-rings in tropical woods. Bois et Foreˆts des Tropiques 113:3–14 Mariaux A (1976) The nature and periodicity of tree-rings in Samba wood. Bois et Foreˆts des Tropiques 169:29–35 Mariaux A (1995) Growth periodicity in tropical trees. IAWA J 16(4):327–328 Murphy JO, Whetton PH (1989) A re-analysis of tree ring chronology from Java. Proc Kon Ned Akad Wetensch 92:241–257 Nobuchi T, Ogata Y, Siripatanadilok S (1995) Seasonal characteristics of wood formation in Hopea odorata and Shorea henryana. IAWA J 16:361–369 Nock CA, Baker P, Wanek W, Leis A, Grabner M, Bunyavejchewin S, Hietz P (2010) Long-term increases stem growth in a tropical monsoon forest in western Thailand. Global Change Biol 1–14. doi:10.1111/j.1365-2486.2010.02222.x Ohashi S, Okada N, Nobochi T, Siripatanadilok S, Veenin T (2009) Detecting invisible growth rings of trees in seasonally dry forests in Thailand: isotopic and wood anatomical approaches. Trees 23:813–822 Palakit K (2004) Dendroclimatological studies of Teak (Tectona grandis L.): a case study in Mae Hong Son province, Thailand. Unpublished master’s thesis, Mahidol University, Nakorn Pathom, Thailand Palmer JG, Murphy JO (1993) An extended tree ring chronology (teak) from Java. Proc Kon Ned Akad Wetensch 96:27–41 Pant GB, Borgaonkar HP (1983) Growth rings of teak trees and regional climatology. In: Singh LR, Singh S, Tiwari RC, Srivastava RP (eds) Environmental management. The Allahabad Geographical Society, Department of Geography, University of Allahabad, India, pp 154–158 Poussart PM, Schrag DP (2005) Seasonally resolved stable isotope chronologies from northern Thailand deciduous trees. Earth Planet Sci Lett 235:752–765 Poussart PM, Evans MN, Schrag DP (2004) Resolving seasonality in tropical trees: multi-decade, high-resolution oxygen and carbon isotope records from Indonesia and Thailand. Earth Planet Sci Lett 218:301–316 Poussart PM, Myneni SCB, Lanzirotti A (2006) Tropical dendrochemistry: a novel approach to estimate age and growth from ringless trees. Geophys Res Lett 33:L17711. doi:10.1029/2006GL026929 Pumijumnong N (1995) Dendrochrnologie mit Teak (Tectona grandis L.) in Nord-Thailand. Dissertation, Universita¨t Hamburg, 109 p Pumijumnong N, Eckstein D (2010) Reconstruction of pre-monsoon weather conditions in northwest Thailand from the tree-ring widths of Pinus merkusii and Pinus kesiya. Trees 25(1):125–132 Pumijumnong P, Park W-K (1999) Vessel Chronologies from Teak in Northern Thailand and their Climate Signal. IAWA J 20(3):285–294

357 Pumijumnong N, Park W-K (2001) Teak vessel chronologies as an indicator of Southeast Asian Pre-monsoon temperature. Palaobotanist 50:21–26 Pumijumnong N, Wanyaphet T (2006) Seasonal cambial activity and tree-ring formation of Pinus merkusii and Pinus kesiya in Northern Thailand in dependence on climate. For Ecol Manage 226:279–289 Pumijumnong N, Eckstein D, Sass U (1995a) Tree-Ring research on Tectona grandis in Northern Thailand. IAWA J 16(4):385–392 Pumijumnong N, Eckstein D, Sass U (1995b) Reconstruction of rainfall in northern Thailand from tree-ring series of teak. IGBPPAGES/PEP-II Symposium on Paleoclimate and environment variability in Austral-Asian transect during the past 2000 years, Nov 28–Dec 1, 1995, Nagoya Japan, pp 186–191 Pumijumnong N, Eckstein D, Park W-K (2001) Teak tree-ring chronologies in Myanmar—a first attempt. Palaeobotanist 50: 35–40 Ram S, Borgaonkar HP, Sikder AB (2008) Tree-ring analysis of Teak (Tectona grandis L.F.) in central India and its relationship with rainfall. J Earth Syst Sci 117(5):637–645 Rao KS, Dave YS (1981) Seasonal variations in the cambial anatomy of Tectona grandis (Verbenaceae). Nord J Bot 1:535–542 Roberts C (2011) Regional and identity: the many faces of Southeast Asia. Asian Politics Policy 3(3):365–382 Rozendaal DMA, Zuidema PA (2011) Dendroecology in the tropics: a review. Trees 25:3–16 Sano M, Buckley BM, Sweda T (2009) Tree-ring based hydroclimate reconstruction over northern Vietnam from Fokienia hodginsii: eighteenth century mega-drought and tropical Pacific influence. Clim Dyn 33:331–340 Sass U, Killman W, Eckstein D (1995) Wood formation in two species of Dipterocarpaceae in Peninsular Malaysia. IAWA J 16(4):371–384 Shah SK, Bhattacharyya A, Chaudhary V (2007) Reconstruction of June-September precipitation based on tree-ring data of teak (Tectona grandis L.) from Hoshangabad, Madhya Pradesh, India. Dendrochronologia 25:57–64 Shiokura T (1989) A method to measure radial increment in tropical trees. IAWA Bull 10:147–154 Sørensen P (1973) Prehistoric iron implements from Thailand. Asian perspective XVII (2). The university press, Hong Kong Suwanpatra K (2006) Seasonal variation in cambial activity of Podocarpus neriifolius at Khao Yai National Park. Unpublished master’s thesis, Mahidol University, Nakhon Pathom, Thailand Tiwary M (2003) Fluid boundaries of India’s forests: concessions and conflicts in joint forest management in Bihar and west Bengal, India. UNU/IAS working paper no 104: 48 p Troup RS (1921) Silviculture of Indian trees, vol 2. Clarendon Press, Oxford, pp 337–783 Truyen NP, Osborn T (eds) (2006) Report on the trade and utilization of Fokienia hodginsii in Lao Cai and son La provinces, north Vietnam. Fauna and Flora International. Vietnam conservation Support Programme, Vietnam, p 117 Venugopal N, Krishnamurthy KV (1987) Seasonal production of secondary xylem in the twigs of certain tropical trees. IAWA Bull 8:31–40 Wahl ER, Morrill C (2010) Toward understanding and predicting monsoon patterns. Science 328:437–438 Waisel Y, Fahn A (1965) The effect of environment on wood formation and cambial activity in Robinia pseudoacacia L. New Phytol 64:436–442 Wannasri S, Pumijumnong N, Shoocongdej R (2007) Teak log coffin head styles in Northern Thailand: time sequencing with dendrochronology. Sci Asia 33:47–56 Williams LJ, Bunyavejchewin S (2008) Deciduousness in a seasonal tropical forest in western Thailand: interannual and intraspecific

123

358 variation in timing, duration and environmental cues. Oecologia 155:571–582 Worbes M (1983) Vegetationskundliche Untersuchungen zweier ¨ berschwemmungswa¨lder in Zentralamazonien. Amazoniana U 8:47–65 Worbes M (1984) Periodische Zuwachszonen an Ba¨umen zentral¨ berschwemmungswa¨lder. Naturewissenschaften amazonischer U 71:157–158 Worbes M (1986) Lebenbedingungen und Holzwachstum in zentral¨ berschwemmungswa¨lder. Scripta Geobotanica amazonischen U 17:1–112

123

Trees (2013) 27:343–358 Worbes M (1989) Growth rings and age of trees in inundation forest, savannas and a mountain forest in the Neotropic. IAWA Bull 10:109–122 Xu C, Sano M, Nakatsuka T (2011) Tree-ring cellulose d18O of Fokienia hodginsii in Northern Laos: a promising proxy to reconstruct ENSO? J Geophys Res 116:D 24109. doi:10.1029/ 2011JD016694 Zimmer H, Baker P (2009) Climate and historical stand dynamics in the tropical pine forests of northern Thailand. For Ecol Manage 257:190–198