APPLICATION OF MULTI-PROXY TREE-RING PARAMETERS IN THE RECONSTRUCTION OF CLIMATE VIS-À-VIS GLACIAL FLUCTUATIONS FROM THE EASTERN HIMALAYA THESIS SUBMITTED TO LUCKNOW UNIVERSITY LUCKNOW UTTAR PRADESH FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN BOTANY BY MAYANK SHEKHAR, M.Sc. M.Tech. BIRBAL SAHNI INSTITUTE OF PALAEOBOTANY LUCKNOW-226007, INDIA
152
Embed
APPLICATION OF MULTI-PROXY TREE-RING PARAMETERS IN …shodhganga.inflibnet.ac.in/bitstream/10603/43041/1/mayank_shekha… · Mrs. Sandhya Misra, Mrs. Archana Singh ir moral support
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
APPLICATION OF MULTI-PROXY TREE-RING PARAMETERS IN
THE RECONSTRUCTION OF CLIMATE VIS-À-VIS GLACIAL
FLUCTUATIONS FROM THE EASTERN HIMALAYA
THESIS SUBMITTED TO
LUCKNOW UNIVERSITY
LUCKNOW
UTTAR PRADESH
FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
IN
BOTANY
BY
MAYANK SHEKHAR, M.Sc. M.Tech.
BIRBAL SAHNI INSTITUTE OF PALAEOBOTANY
LUCKNOW-226007, INDIA
This thesis is dedicated to my beloved parents
i
Acknowledgements
It gives me immense pleasure to express my deep sense of gratitude to
my mentor and Ph.D. supervisor Dr. Amalava Bhattacharyya, Emeritus Scientist, at
Birbal Sahni Institute of Palaeobotany (BSIP), Lucknow, Uttar Pradesh (India) for his
valuable guidance, scientific discussion, encouragements and tremendous moral
support during the entire course of the completion of this dissertation.
I would like to express my gratitude to Prof. Sunil Bajpai, Director, BSIP,
Lucknow (U.P., India) for providing all necessary facilities to me and also for the
encouragements.
I express my sincere gratitude to Dr Santosh Kumar Shah (Santosh Bhiya)
Scientist C (BSIP) for his valuable advices and suggestion on all statistical aspects of
tree-ring analysis and also for moral support.
This work is financially supported by Department of Science and Technology,
New Delhi. DST Project No. ESS/91/38/2005. I extend my sincere thanks to Prof. R.
Ramesh, Director, ISO-IGBB for providing Senior Research Fellowship to me.
I wish to thank the Forest Officials of North Sikkim especially Mr. Pradeep
Kumar (IFS) for giving permission and providing necessary facilities during the
collection of tree-ring samples. I am also grateful to India Meteorological
Department, Pune for providing the climate data.
I am indebted to the entire staff of my laboratory Mrs. Nivedita Mehrotra,
Mrs. Sandhya Misra, Mrs. Archana Singh for their moral support and
encouragements. I would also like to thank to my seniors Dr. Parmindar Singh
Ranhota, Scientist C, Dr. Ruby Ghosh (Didi), Scientist B, Dr. Md. Firoze Quamar,
BSRA, Dr. Gaurav Srivastva, Scientist B, Dr Anumeha Shukla, Scientist B, Birbal
Sahni Institute of Palaeobotany, Lucknow for their moral support and
encouragements.
I would like to thank my friends Er. Devendra Kumar Maurya, Rajesh Kumar
Associate, Lucknow University) and Ms. Muskan Singh, especially for helping me
while using Remote Sensing and GIS Applications.
I would like to thank Mr. Shaktiman Singh, Research Scholar, Sharda
University (New Delhi) for the critical discussion and suggestion on hydrological
modeling.
I also acknowledge The National Snow and Ice Data Center (NSIDC) with
thanks for providing Mass Balance Data (Mark Dyurgerov, 2005) of Eastern
Himalaya and adjoining countries for analysis.
I extend my thanks to my friends Anish, Vikash, Atual, Preeti, Shikha, Siksha,
who have always supported and encouraged me to pursue this study.
I also acknowledge my sister with thanks who has morally supported me
during the course of the completion of dissertation work.
I would like to express my gratitude to my parents and my brothers Mr.
Shashank Shekhar and Mr. Shardendu Shekhar for all their encouragements and
confidence that they reposed on me.
Finally, I would like to express my deep sense of regard to Almighty for His
divine grace that He bestowed upon me throughout my life.
Mayank Shekhar
Contents Page Acknowledgements i-ii List of Figures iii-vii List of Tables viii Preface ix-x Chapter 1. Introduction
1-16
1.1. Tree-ring as proxy for Paleoenvironment Reconstruction 1
1.2. Theme of the Dissertation 2-4 1.3. Objectives 4 1.4. Overview of forests of Sikkim Himalaya 4-5
1.4.1. Forest covers Sikkim, Eastern Himalaya based on satellite data
5-6
1.4.2. North Sikkim forest cover 6
1.5. Climate of India - An Overview 6-8
1.5.1. Temperature trends 6 1.5.2. Precipitation trends 7-8
1.6. Resume of earlier tree-ring study 9-16 1.6.1. Global context 9
1.6.1.1. Paleoclimatology 9 1.6.1.2. Hydrology 10 1.6.1.3. Glaciology 10 1.6.1.4. The Palmer Drought Severity Index (PDSI) 10
1.6.2. Indian Context 11 1.6.5.1. Western Himalaya 12 1.6.5.2. Eastern Himalaya 12-13 1.6.5.3. Peninsular India 13-14
1.6.5.4. Dendrohydrology 14
1.6.5.5. Dendroglaciology 15 1.6.5.6. Tree Growth and its Relationship with El Nino 15-16 1.6.5.7. The Palmer Drought Severity Index (PDSI) 16
Chapter 2. General Principles & supporting data
17-34
2.1 Selection of tree ring site 17 2.2. Study of tree-rings 17-18 2.3. Acquisition of tree- ring width data 18-20 2.4. Standardization of tree ring data & Chronology preparation 20-22 2.5. Climate and Glacial data 22-25
2.5.1. Regional climate data 22
2.5.1.1. Details of CRU T.S2.1 data and source 24
2.5.1.2. Statistical assessment of climate data 24-25
2.5.1.3. Missing Value estimation in climate data 25
Page
2.5.1.4. Descriptive statistics of climate data 25 2.5.2. Glacial data 26
2.5.2.1. Glacier Front variation and Mass-balance data 26-27 2.6. Dendroclimatic modeling (Past climate reconstruction) 27-28
2.6.1. Principal component analysis (PCA) 28
2.6.2. Bootstrap Response Function 30
2.6.3 Bootstrap Transfer Function 30-31 2.6.4 Correlation analysis 31 2.6.5. Linear regression method for Climate reconstruction 31
2.7. Dendrohydrological modeling 32-33 2.7.1. Rivers discharge data 32
2.7.2. Climate data for discharge site 32
2.7.3. Correlation analysis for Tree-growth and discharge relationship
33
2.7.4. Discharge Reconstruction method 33 2.8.4.1. Linear regression method for discharge reconstruction
33
2.8. Correlation analysis for tree growth and its relation with PDSI/ El Nino
5.3. Physiological explanation of tree growth climate relationship 74-76
5.3.1. Positive correlation with March to April temperature 74-75 5.3.2. Negative correlation with June to September 75-76
5.4. Dendroclimatic modeling (Past Climate reconstruction) 77-92 5.4.1. Bootstrap Method for March-April Maximum temperature reconstruction
77-82
5.4.1.1 Salient features of reconstructed March April maximum temperature
83
5.4.2. Linear regression method for Average temperature of (July August) temperature reconstruction
83
5.5. Reliability of the regression model 83-92
5.5.1. Calibration verification of model for temperature
reconstruction.
83-84
5.5.2 Variability in reconstructed climate data. 84-89
5.5.3. Characteristics of reconstructed temperature 90
5.5.6. Cyclic nature of tree based reconstructed climate records
91-92
Chapter 6. Tree growth and glacier relationship 93-98
6.1 Glacier behavior and tree-ring width chronology 93-96 6.2. Reconstructed temperature and glacier fluctuation 97
Page
6.2.1. Role of Maximum March-April temperature in fluctuation of Zemu glacier.
97
6.2.2. Role of Average July-August temperature in Fluctuation of Zemu glacier.
98
Chapter 7. Dendrohydrological modeling (reconstruction of Discharge) 99-107
7.1. Study area of river discharge 99
7.2 Relationships between Tree growth and Climate 100
7.3. The stream flow reconstruction method 102 7.4. Variability in reconstructed discharge data 103-106 7.5. Cyclic nature in reconstructed January-April discharge.
106-107
Chapter 8. Tree growth and its relation with, El Nino 105-108
8.1. Tree-growth/glacier fluctuation and El Niño relation 105-106
8.2. Tree growth and its relation with PDSI
107-108
Chapter 9. Discussion and conclusions 112-115
9.1 Further work for the improvement of research work 115
Chapter 10. References 116-133
iii
List of Figures Page
Fig.1.1 Forest cover map of Sikkim, Eastern Himalaya (Adopted from Indian State
of Forest Report 2011).
7
Fig. 2.1 Map of India showing position of Zemu glacier and adjoining area in
Sikkim. 2
Fig. 2.2 Monthly variation of total precipitations (bars), mean maximum
temperature (red line), and mean temperature (green line) and mean minimum temperature (pink line) for Gangtok meteorological station.
23
Fig. 2.3 Monthly variation of total precipitations (bars), mean maximum
temperature (red line), and mean temperature (green line) and mean minimum temperature (pink line) for CRU Grid data.
24
Fig. 3.1. Location map of Zemu glacier 36 Fig. 3.2 (a) Satellite map of Zemu glacier and (b) Zemu glacier along with
Vegetation Cover in and around this region based on NDVI (Normalised
Difference Vegetation Index).
37
Fig.3.3 Sampling site Lachen showing forest of Abies densa 39 Fig.3.4. Sampling site Talem showing forest of Abies densa 41 Fig. 3.5. Sampling site Jakthang showing forest of Abies densa and Juniperus
recurva 42
Fig. 3.6: Sampling site Yabuk showing forest of Abies densa and Juniperus recurva 43 Fig. 3.7. (a)Sampling site Yabuk showing zone of Juniperus squmata scrub,(b)
Sampling site Yabuk showing forest of Juniperus recurva 44
Fig.3.8. (a) Satellite map showing location of sample collection site along with
meteorological station and (b) Sketch map route of collection of samples from ZEMA to Zemu glacier site. (the details of the abbreviations of sites and trees are given in Table 2.2)
46
Fig. 3.9. Collection of tree cores from the tree through increment borer. 47 Fig. 3.10. Processing of tree-ring cores. 47 Fig.4.1. Ring-width index chronology of Abies densa from Yabuk 49 Fig.4.2. Ring-width index chronologies of Juniperus squmata from Yabuk 49 Fig. 4.3. Ring-width index chronologies of Juniperus recurva from Yabuk site. 50 Fig. 4.4. Ring-width index chronologies of Abies densa from Zakthang site. 50 Fig. 4.5. Ring-width index chronologies of Abies densa from Talem sites. 50
iv
Fig. 4.6. Ring-width index chronology of Abies densa from Dozamkhola site 50 Fig. 4.7. Ring-width index chronologies of Abies densa Zema site. 51 Fig. 4.8. Ring-width index chronology of Abies densa from Lachen site. 51 Fig. 4.9. Ring-width index chronology of Larix griffithiana from Lachen site. 51 Fig. 4.10. Early wood width index chronology of Larix griffithiana from Lachen site. 52 Fig. 4.11. Late wood width index chronology of Larix griffithiana Lachen site. 52 Fig. 4.12. Ring width index chronology of Abies densa from Yumthang site. 52 Fig.4.13 (a,) and( b). Time-series plots of the two PCs from ring width chronologies
along with altitudinal of gradient of Zemu glacier Sikkim Himalaya. 53
Fig. 5.1. Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for ABDE_YAB. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level (p < 0.01) above and below.
58
Fig. 5.2. Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for JUSQ_YAB. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01) above and below
59
Fig. 5.3. Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for JURE_YAB, horizontal pink line indicates significance level (p < 0.05) and red line indicates significance level(p < 0.01)
60
Fig..5.4. Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for ABDE_JAK. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01)
61
Fig. 5.5. Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for ABDE_TAL. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01) above and below
62
Fig. 5.6. Plot of (a) Response function and (b) Correlation analyses based on standard
chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for ABDE_DOZ. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01) above and below
63
v
Fig. 5.7. Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for ABDE_ZEM. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01) above and below.
64
Fig. 5.8 Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for ABDE_LAC. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01) above and below.
65
Fig. 5.9 Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for LAGR_RW_LAC. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level (p < 0.01) above and below.
66
Fig. 5.10 Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for LAGR_EW_LAC. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01) above and below.
67
Fig. 5.11 Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for LAGR_RW_LW_LAC. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01) above and below
68
Fig. 5.12. Plot of (a) Response function and (b) Correlation analyses based on
standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for ABDE_YUM. Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level(p < 0.01) above and below
69
Fig. 5.13 Principle component plot in rotated space 70 Fig.5.14. Plot of (a) Response function and (b) Correlation analyses based on rotated
principal component (PC) scores [AD 1881–1994 for PC#1, PC#2,] versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation). Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level (p < 0.01) above and below.
74
Fig.5.15 Tree-ring based reconstructed Maximum March-April temperature in Zemu
Valley (dotted green line is start year of the reliable time span); Actual (green line) and estimated (red line).
79
vi
Fig.5.16 Showing Anomaly in the reconstructed March_April temperature. 80 Fig.5.17 The comparison of actual and reconstructed March April maximum
temperature from AD1966 to 2000. 81
Fig. 5.18 Scatter plot of actual and reconstructed Maximum March-April
temperature with a linear relationship highlighted during the period of 1966–2000.
81
Fig. 5.19 Tree-ring based reconstructed Mean July-August temperatures in Zemu
Valley; (dotted green line is start year of the reliable time span), actual (green line) and estimated (red line).
86
Fig. 5.20 The comparison of actual and reconstructed Maximum temperature of
March-April from 1976 to 1996.
87
Fig. 5.21 Scatter plot of actual and tree-ring reconstructed Late summer mean
temperature (July-August) temperature with a linear relationship highlighted during the period of 1966–2000.
87
Fig. 5.22 Showing Anomaly in the reconstructed July_August temperature 88 Fig. 5.23 (a) Maximum March-April temperature. (b) The wavelet power spectrum.
The power has been scaled by the global wavelet spectrum (at right). The cross-hatched region is the cone of influence, where zero padding has reduced the variance. Black contour is the 95% significance level, using a red-noise (autoregressive lag1) background spectrum. (c) The global wavelet power spectrum (black line).The dashed line is the significance for the global wavelet spectrum, assuming the same significance level and background spectrum as in.
91
Fig. 5.24 (a) July-August mean temperature (b) The wavelet power spectrum. The
power has been scaled by the global wavelet spectrum (at right). The cross-hatched region is the cone of influence, where zero padding has reduced the variance. Black contour is the 95% significance level, using a red-noise (autoregressive lag1) background spectrum. (c) The global wavelet power spectrum (black line). The dashed line is the significance for the global wavelet spectrum, assuming the same significance level and background spectrum as in.
92
Fig. 6.1 Photograph showing the Snout position of Zemu Glacier, North Sikkim
(modified after Luitel et al., 2012) 95
Fig. 6.2 Tree growth and its relation with Zemu glacier 96 Fig. 6.3 Tree growth and its relation with Changmekhangpu glacier
96
Fig. 6.3 Comparison of Fir Chronology of ABDE_YAB with available Mass
balance data of three Chinese glacier (AD 1988-1995) and Nepal glacier (AD 1996-1999).
96
vii
Fig.7.1 Map showing location of tree ring site, meteorological station and discharge gauge station at Lachen, north Sikkim, Eastern Himalaya. For the generation of the map SRTM 30 (digital terrain elevation data set was used)
99
100 Fig.7.2 Photograph showing “Zemu Chuu” at Lachen North Sikkim Fig.7.3 Mean annual variation of “Zemu Chuu” river discharge at Lachen gauging
station (1976-1996) North Sikkim.
101
Fig.7.4 Mean monthly variation of “Zemu Chuu” river discharge at Lachen
gauging station (1976-1996) North Sikkim. 101
Fig.7.5 Correlation plot of standard chronologies of ABDE_ZEM with averaged
monthly Discharge data of Lachen data (1977–1996). Horizontal pink line indicates significance level (p < 0.05) and red line indicate significance level (p < 0.01).
102
Fig.7.6 Reconstruction of January–April discharges of “Zemu Chuu” at Lachen,
North Sikkim since AD 1775. The red line represents reconstructed while green line represents actual data
104
Fig.7.7 (a) The comparison of actual and reconstructed stream flow (January-
April) from 1976 to 1996 (b) Scatter plot of actual and tree-ring reconstructed stream flow (January-April) with a linear relationship highlighted during the period of 1976–1996.
105
Fig.:7.3 (a) January-April _Discharge. (b) The wavelet power spectrum. The power
has been scaled by the global wavelet spectrum (at right). The cross-hatched region is the cone of influence, where zero padding has reduced the variance. Black contour is the 95% significance level, using a red-noise (autoregressive lag1) background spectrum. (c) The global wavelet power spectrum (black line). The dashed line is the significance for the global wavelet spectrum, assuming the same significance level and background spectrum as in
107
Fig. 8.1 Correlation values of mean monthly El Niño 3.4 with standard regional
chronologies PC#1 and PC#2. Monthly variables spanning from January to December. The pink horizontal line indicates 95% confidence limits.
106
Fig. 8.2 Plot of Zemu glacier retreat with mean data of ring-with, and El Nino 3.4 106 Fig. 8.3 Correlation values of mean monthly PDSI of the one grid points with
standard regional chronologies of all site and PC#1 and PC#2. (a) All chronologies. (b) PC#1, PC#2. Monthly variables spanning from November of the previous year to October of the current year. The pink horizontal line indicates 95% and red line indicate 99% confidence limits.
108
viii
List of Tables
Page Table 2.1 Site information and tree-ring chronologies statistics of Zemu Valley. 19
Table 2.2 Site information and tree-ring chronologies statistics Larix griffithiana (EW, LW).
19
Table 2.3 Description of IMD climate data used for temperature records. 22
Table 2.3 Description of IMD climate data used for precipitation records. 23
Table 2.5. Description of CRU T.S 2.1climate data. 23
Table 2.6 General information about the glacier and data. 29
Table.4.1. Correlation matrix for standard tree-ring chronologies. 55
Table.5.1 Summary of the PCA Statistics of tree-ring chronologies. 71
Table.5.2. Summary of rotated principal component retained in PCA. 71
Table.5.3 Monthly climatic models for Zemu Valley based on ABDE_YAB, using bootstrap method.
80
Table 5.4 Statistics of Calibration and Verification for tree-ring reconstruction of maximum March-April temperature.
80
Table. 5.5 Statistics of calibration and verification for tree-ring reconstruction of July-August mean temperature in the common period 1966–2000.
89
Table 7.1 Statistics of calibration and verification for tree-ring reconstruction of January-April Stream flow.
105
Table. 8.1 Correlation value of glacier retreat, mean data of ring-with, and El Nino
3.4.
106
ix
Preface The present dissertation entitled “Application of multi-proxy tree ring
parameters in the reconstruction of climate vis-à-vis glacial fluctuation from the
Eastern Himalaya” is based on multi-proxy tree ring parameters in record of past
climate and glacier fluctuation. Tree-ring samples analyzed were collected mostly in
the form of cores and discs from left over stumps from several conifer taxa growing in
diversified forests ranging from temperate to sub-alpine in and around adjoining area
of Zemu Glacier, Eastern Himalaya.
The research work included in this dissertation is structured into nine chapters.
The beginning with the “Introduction”, where I have introduced briefly significance
of tree-ring as proxy for paleoenvironment reconstruction emphasizing its prospects in
understanding the climatic changes, hydrological changes, glacier fluctuation and its
linkage with Palmer Drought Severity Index (PDSI) and El Nino followed by a brief
account on the overview of forests of Sikkim Himalaya and glimpse of climate of the
India emphasizing Himalayan region have been given. A brief account on resume of
earlier tree-ring study from both global and from India with emphasis to the
Himalayan region and its applications to climate, hydrology, glacier and El Nino have
also been discussed in this chapter.
In the second chapter, “General Principles & supporting data” I have
discussed dendrochronological technique for sample collection and processes
involved in retrieving climate and tree growth relationship to glacial advancement
/retreat etc.
In chapter three i.e. “Study Site selection and collection of samples and
Processing”, a brief account on description of study sites, collection and processing
of tree-ring samples have been discussed.
In chapter four i.e “Tree ring Chronology of Zemu Glacier valley”, I have
discussed chronology development which begins with dating of tree rings exactly to
the calendar years of their formation through cross dating, measurement of ring width
and details of chronology preparation and their quality assessment based on some
selected statistics and regional signal assessment of chronologies through Principal
component.
Chapter five i.e. “Dendroclimatic modeling (Climate reconstruction)”,
contains the details of dendroclimatic modeling from the Eastern Himalayan region,
x
tree growth/climate relationship with the help of Response Function Analysis using
Orthogonal Bootstrap Regression and Correlation analysis, the reconstruction of
climate using tree-ring chronologies using both Bootstrap Transfer Function and
Linear Regression method. At the end of this chapter I have discussed about the
salient features of reconstructed climate and its cyclic behaviour with the help of
wavelet analysis.
In chapter six the “Tree growth and glacier relationship” and data
comparison with adjoining glaciers have been mentioned.
In chapter seven, “Dendrohydrological modeling (Reconstruction of
Discharge)” has been discussed and reconstructed discharge data has been compared
with corresponding other data viz., drought and El Nino Years.
Chapter eight includes “Tree growth and its relation with PDSI, El Nino”
which includes correlation analysis with monthly data set of the same.
Chapter nine, “Discussions and conclusions” is the most important where
results derived from the present study are summarized and salient achievement of
these objectives have been discussed followed by an account on limitations of the
present study along with scope for improvement and further work towards better
understanding of various aspects of climate and climate related phenomena from the
Eastern Himalayan region.
Chapter 1 Introduction
1
1. Introduction In accordance to the global warming, temperature of the Eastern Himalayan
region is also increasing at the rate of 0.01oC/year or more (Shrestha et al., 2010)
when surface temperature of earth as a whole increased by about 0.74° C (+/- 0.18° C)
over the last 100 years (IPCC, 2007). One of the most important indicators of climate
change is the variation of mass of the glaciers. It is general trend that the Himalayan
Glaciers, like other glaciers of the globe, are retreating very fast as indicated by the
movement of snout position towards the higher elevations. The fast retreat of the
Himalayan glaciers is much of societal concern. We are worried about the fate of these
glaciers in respect to the global warming and associated erratic monsoon behavior in
the coming years. For better understanding of glacial movement, it is essential to
explore the variation and changes of climate for longer time span. As widespread,
direct measurements of climate variables are only available for about one to two
centuries back in time and that too are only from limited number of stations which are
inadequate for climate modeling and to established long term link with the retreat and
advancement of glaciers. It is therefore necessary to use indirect indicators or
‘‘proxy’’ to measure climate variability provided by the natural archives of
information present in our environment to reconstruct earlier changes. These natural
archives record by their biological, chemical, or physical nature, climate-related
phenomena. Additionally, information is provided by written archives from historical
documents. Some proxy indicators, including most sediment cores, low accumulation
ice cores, and preserved pollen cannot record climate changes at high temporal
resolution. These indicators generally have poor chronologies because of uncertain
radiometric dating methods or questionable ‘‘age model’’ assumptions (e.g., the
assumption of constant stratigraphic rates between marker or ‘‘dated’’ horizons).
Such proxy indicators are thus only be useful for describing to climate changes in
centennial and often longer timescales. High-resolution, annually and/or seasonally
resolved proxy climate records (historical documents, growth and density
measurements from tree rings, corals, annually resolved ice cores, laminated ocean
and lake sediment cores, and speleothems) can, however, describe year-by-year
patterns of climate in the past centuries (Folland et al., 2001; Jones et al., 2001; Mann,
2001).
Chapter 1 Introduction
2
1.1. Tree-ring as proxy for Paleoenvironment Reconstruction All proxy data are indirect measurements of climate change, and they vary
considerably in their reliability as indicators of long-term climate. Tree ring has
played a pivotal role in high-resolution paleoenvironmental analysis. This is one of
the best proxies for the understanding of climate, because tree ring are environmental
sensitive annual growth rings, which produce precise and reliable chronology of
ecology and climate (Fritts, 1976). For a reliable reconstruction of past changes from
proxy data it is essential that reconstructions based on these indirect climate indicators
be ‘calibrated’’ and independently ‘‘validated’’ against instrumental records during
common intervals of overlap. For the tree, the limiting factor may be internal i.e.
(biogenic) or external (environmental factor). Some of these factors can be a limit in
nutrient availability or growing conditions such as precipitation and temperature.
However, it is important to realize the interconnection of these two factors. Internal
processes cannot operate without supplies delivered by external forces which enhance
the environment signal in tree-ring growth (Fritts, 1976).
1.2. Theme of the Dissertation In this dissertation reconstruction of climate vis-a-vis glacial movements
(advance/retreat) based on multi proxy tree ring parameters i.e., tree ring width data as
a whole or its early wood and latewood separately, have been taken in to
consideration. Although traditional dendroglaciological techniques make it possible to
date the timing of glacier advances and the maximum glacier extent (Luckman, 1988),
but they provide little insight into the attendant climate conditions or duration of
glacier advances (Porter, 1981; Luckman, 2000). By contrast, applied
dendroglaciological research has successfully established linkages between glacier
mass balance and tree-ring-width variability. La Marche and Fritts (1971), Matthews
(1977) and others directly related tree-ring-width variability to climate conditions, and
subsequently to glacier fluctuations. They discerned that high winter snow
accumulation, and short cool and cloudy summers favored positive glacier mass
balance, but were detrimental to tree growth and resulted in narrow annual tree rings.
Conversely, warm summer air temperatures and a greater number of sunny days were
shown to result in enhanced glacial ablation, as well as increased radial growth rates
Chapter 1 Introduction
3
and the production of wider annual growth rings Since the growth rings of
climatically sensitive trees respond inversely to the same variables that drive glacier
mass balance, dendroglaciological techniques provide a means for developing
glaciological histories in remote areas without mass balance data. For stream flow reconstructions, tree-ring data are valuable proxy to
reconstruct the long-term discharge data. The records of total discharge from most of
the river gauge stations of India are available for short period which is inadequate for
modeling the long-term properties of river discharge. A short stream flow records is
not statistically reliable for assessing long term trends and shifts in flow volume. In
this context reconstruction of past discharge is very essential and useful for water
resource management.
Palmer Drought Severity Index (PDSI) was a landmark in the development of
drought indices. Tree growth is controlled by soil moisture condition, while soil
moisture balance is a result of the integrative effect of precipitation,
evapotranspiration, physical-chemical properties of soil substrate, etc. In order to
better understanding of drought condition it is necessary to take not only precipitation
but also other hydrological meteorological factors. The PDSI enables measurement of
both wetness (positive value) and dryness (negative values), based on the supply and
demand concept of the water balance equation, and thus incorporates prior
precipitation, moisture supply, runoff and evaporation demand at the surface level.
The Palmer Drought Severity Index (PDSI; Palmer, 1965) is widely used as
meteorological drought index and it was reconstructed long back based on tree-ring
records from diversified geographical region of the globe (Briffa et al., 1994; Cook et
al., 2004, 2010; Dai et al., 2004; D’Arrigo et al., 2006).
El Niño-Southern Oscillation (ENSO) is a coupled Ocean-atmosphere
phenomenon which has impact on global climate and especially on monsoons. The
negative correlation between monsoon rainfall and ENSO are shown by several
workers since long back (Webster and Palmer, 1997 references there in) in which a
weak (strong) monsoon is related to a warm (cold) event through an anomalous
Walker cell driven by tropical East Pacific SST anomalies, has weakened rapidly
since the late 1970s (Chang et al., 2001). Tree ring data has been found useful proxy
to reconstruct ENSO in the background of short instrumental data which could not
prove whether ENSO has changed earlier also or not in long time scale. Recently
Chapter 1 Introduction
4
based on tree ring chronologies from both tropics and mid-latitudes in both
hemispheres a seven century long back history of ENSO was presented (Li et al.,
2013).
In this dissertation I have attempted to reconstruct climate and its linkage with
other climatically induced phenomena like glacier movement, stream flow, PDSI, El
Niño reconstruction based on multi proxy tree ring parameters i.e., tree ring width
data as a whole or early wood and latewood separately. Thus in the course of research
work in this dissertation I tried to achieve the following objectives.
1.3. Objectives
• To prepare tree-ring database based on dating of tree ring from old trees
and shrubs growing on moraine deposits around the Zemu glacier valley
North Sikkim, Eastern Himalaya.
• To establish tree-growth/climatic relationship based on analyses of multi
proxy tree ring data (ring width, early wood width, late wood width,).
• Analysis of climatic changes based on reconstructed long climatic data
derived from tree ring.
• To understand climatic changes and their relationships with glacial
fluctuations
Besides, attempts have also been made to work in the following aspects which are
linked with climate changes.
• Reconstruction of discharge from Zemu Chuu (stream) originated from
Zemu glacier Eastern Himalaya.
• Relationships of tree growth with El Niño and PDSI.
1.4. Overview of forests of Sikkim Himalaya Eastern Himalaya especially its extreme western state, Sikkim, India is well-
known for its biodiversity and is regarded as a botanist's paradise. The steep vertical
climb from the plains of West Bengal to the high altitude areas of Sikkim represents
one of the world's steepest altitude gradients which provides diversified ecological
conditions for the great variety of flora, ranging from the sub tropical to alpine.
Chapter 1 Introduction
5
Fig. 1.1.Forest cover map of Sikkim, Eastern Himalaya (Adopted from Indian State of
Forest Report 2011).
1.4.1. Forest cover of Sikkim, Eastern Himalaya based on satellite data
The forest cover in this area, based on interpretation of satellite data of
December 2008, is 3,369 km2 which is 47.34% of the state's geographical area. In
terms of forest canopy density classes, the state has 500 km2 'area under very dense
forests, 2,161 km2 area under moderately dense forests and 698 km2 area under open
forests. (India State of Forest Report, 2011) (Fig.1.4.) Forest type mapping using
satellite data has been undertaken by Forest Survey of India with reference to ground
truth of forest survey data as per of the Classification of Champion & Seth (1968).
As per this assessment, the state has 10 forest types which belong to six groups, viz.
Tropical Moist Deciduous, Subtropical Broadleaved Hill, Montane Wet Temperate,
Himalayan Moist Temperate, Sub Alpine Forests and Moist Alpine Scrub. (India
State of Forest Report, 2011). Overall assessment of forest types of this region it
Chapter 1 Introduction
6
appears that North Sikkim seems to more potential in terms of tree ring analysis. This
area has a wide geographical and climatic variation which ensue growth of diversified
trees. Several trees and sites in its pristine high mountains close to glaciers provide
ample scope for various aspects of tree ring or dendrochronological studies.
1.4.2. North Sikkim forest cover
North Sikkim geographical area 4,226 (area in Km2) Very dense forest (VDF)
135 Km2, Moderate dense forest (MDF) 890 Km2,Open forest (OF) 292 Km2 total
1,317 Km2 and percentage of GA 31.16 sub 208 Km2. (India State of Forest Report,
2011).
1.5. Climate of India - An Overview
Before taking up the reconstruction of climate of an area it is necessary to
have knowledge of modern climate of the site of investigation and regional climate as
a whole.
1.5.1. Temperature trends
Analysis of temperature data for the period 1901-2009 suggests that annual
mean temperature for the country as a whole has risen by 0.560C over the period. It
has been recorded that annual mean temperature is generally above normal (normal
based on period, 1961-1990 since 1990. This warming is primarily due to rise in
maximum temperature across the country, over larger parts of the data set. However,
since 1990, minimum temperature is steadily rising and rate of its rise is slightly more
than that of maximum temperature (Attri et al., 2010). Warming trend over globe, of
the order of 0.740C has been reported by IPCC (2007). In general, mean maximum
and mean minimum temperature during January are 9.0 to 9.5°C and 3.0 to 3.5°C
respectively and during peak summer it ranges from 19.2 to 19.5 °C and 14.5 to
14.8°C respectively. Relative humidity in the month of July is over 60% and in
October it is over 80% (Rao, 1981). The rainfall along the Tista valley, Sikkim shows
that it is maximum during July -August in most of the sites. There is heavy rainfall at
lower ranges and it decreases sharply at higher elevations where precipitation is
Chapter 1 Introduction
7
mostly in the form of snow (Samui, 1994). The Sikkim experiences a heavy rainfall
due to its proximity with the Bay of Bengal. Pre-monsoon rain occurs in April-May
and Monsoon (South-West) operates normally from the month of May and continues
up to early October Average annual rainfall varies from 1,300 mm at valleys to 4,300
mm at the mountain ridges. The humidity remains very high during the rainy season
(85-97%) (Singh et al., 2009).
1.5.2. Precipitation trends
The all India annual and monsoon rainfall for the period 1901-2009 do not
show any significant trend (Attri et al., 2010). Similarly rainfall for the country as
whole for the same period for individual monsoon months also does not show any
significant trend. The alternating sequence of multi-decadal periods of thirty years
having frequent droughts and flood years are observed in the all India monsoon
rainfall data. The decades 1961-70, 1971-80 and 1981-90 were dry periods. The first
decade (1991-2000) in the next 30 years period already experienced wet period. The
frequency of extreme rainfall (Rainfall ≥ 124.4 mm) shows increasing trend over the
Indian monsoon region during the southwest monsoon season from June to September
(JJAS) and is significant at 98% level. It is also found that the increasing trend of
contribution from extreme rainfall events during JJAS is balanced by a decreasing
trend in category (rainfall ≤ 64.4 mm/day) rainfall events. Similarly on monthly 21
scale, the frequency of extreme rainfall events show significant (95% level) increasing
trend during June and July, whereas during August and September the increasing
trend is not significant statistically. Like the frequency of extreme rainfall events, the
contribution of extreme rainfall to the total rainfall in a season is also showing highly
significant increasing trend during the monsoon season from June to September and
during June and July on monthly scale. It is observed that the mean monthly
contribution of heavy and extreme rainfall events (rainfall > 64.4 mm in a day) during
June-July is 5 to 6% higher than that during August-September and hence contributes
significantly to the total rainfall during the first half of the season (June and July)
(Attri et al., 2010). In the Eastern Himalaya has high variability in summer rainfall is
attributed to orographic influence and variation in location, timing and intensity of
monsoon (Mani, 1981).
Chapter 1 Introduction
8
Monsoon begins earlier than other part of the Himalaya and the area is more
evenly humid than over any other part of the Himalaya because of its proximity to the
Bay of Bengal and direct exposure to the effects of the moisture laden south-west
monsoon. The average rainfall of this region is around 3,000 to 4,000 mm; of which
about more than 75% falls in the monsoon months (June to September) although the
rainy season extends from April to October. Rainfall during winter i.e., from
November to March originates from the north-east monsoon and is negligible,
sometimes nil. January is the coldest month whereas, May or early June before the
burst of monsoon are the hottest time of the year (Mani, 1981).
Chapter 1 Introduction
9
1.6. Resume of earlier tree-ring study
1.6.1. Global context
Some significant progress in the field of tree-ring analysis in global aspect has
been discussed in this section to get the knowledge of the present status of this subject
in different aspects.
1.6.1.1. Paleoclimate
Tree rings as proxy climate indicators have been extensively used for the
reconstruction of the past seasonal temperatures/precipitation/drought, and other
climatic parameters based on measurements of annual ring widths, latewood densities
and other tree parametres (Fritts, 1971, 1976, 1991; Schweingruber, 1987; Jacoby and
D’Arrigo, 1989; Briffa et al., 1992; Cook et al., 1999; Briffa and Osborn, 1999, 2002;
Briffa et al., 2001; Krepkowski et al., 2012). A good amount of tree-ring based
climate reconstruction has been made from different parts of the world. Tree ring
reconstructions offer the advantage of potentially being quite long (e.g., several
millennia) (Cook et al., 2000; Briffa and Osborn, 2002) high resolution climate.
During the most recent decades, there is evidence that the response of tree ring
indicators to climate has changed, particularly at higher latitudes and more so for
density than ring width measurements (Briffa et al., 1998). One suggested source for
this behavior is ‘‘CO2 fertilization’’, the potential enhancement of tree growth at
higher ambient CO2 concentrations. Though it is extremely difficult to establish
existence of this effect (Wigley et al., 1988), however there is evidence that it may
increase annual ring widths in high-elevation drought-stressed trees (Graybill and
Idso, 1993). Recent work making use of climate reconstructions from such trees has
typically sought to remove such influences prior to use in climate reconstruction
(Mann et al., 1998; Mann and Jones, 2003). Other factors have been suggested as
possible explanations for the apparent anomalous tree ring/climate relationships
(Briffa et al., 1998), including the changing seasonality of the climate itself (Vaganov
et al., 1999; Biondi, 2000; Druckenbrod et al., 2003).
Chapter 1 Introduction
10
1.6.1.2. Hydrology
Tree-ring record has now been recognized as a valuable proxy for stream flow
reconstruction to assess the long-term discharge behavior of a river and its
management in various water resources sectors (Meko and Graybill, 1995; Stockton
and Jacoby, 1976; Pederson et al., 2001; Woodhouse, 2001; Meko and Woodhouse,
2005; Woodhouse and Lucas, 2006; Gou et al., 2007; Lara et al., 2008; Axelson et al.,
2009; Liu et al., 2010; D’Arrigo et al., 2011; Margolis et al., 2011; Maxwell et al.,
2011; Wise, 2010; Urrutia et al., 2011). Recently Cook et al. (2013) reconstructed the
Indus River discharge based on a network of tree-ring sites from the Upper Indus
Basin covering the period AD 1452–2008.
1.6.1.3. Glaciology
Tree rings has proven promising proxy to study and date the movement of
glaciers (Luckman, 1988, 1994, 1995, 1996; Schweingruber, 1988). Most early
dendroglaciological investigations focused on determining annually-resolved moraine
ages to date the timing of maximum glacier extent (Tarr and Martin, 1914; Matthes,
1939; Mathews, 1951; Karlén, W. 1984; Braeuning, A. 2006; Gou et al., 2006;
Barclay et al., 2009; Garavaglia et al., 2010), or on documenting glacial recession
rates by dating successional trends on recently deglaciated surfaces (Cooper, 1916;
Lawrence, 1950; Sigafoos and Hendricks, 1961). The age of moraines and other
glacial deposits are commonly determined by counting the tree rings of the oldest tree
found growing on the surface (Smith and Lewis, 2007).
1.6.1.4. The Palmer Drought Severity Index (PDSI) The PDSI is widely used as meteorological drought index and tree ring records
has proven promising proxy to reconstruct PDSI long back from many parts of the
globe (Briffa et al., 1994; Cook et al., 2004; Dai et al., 2004; D’Arrigo et al., 2006; Li
et al., 2010). Positive PDSI values indicate wetter conditions, whereas negative values
indicate drier conditions. Recent development of the self-calibrated PDSI (sc-PDSI)
(Wells et al., 2004), which is spatially comparable and reports extreme wet and dry
events at frequencies expected for rare conditions. Recently, Cook et al. (2010) used
PDSI to demonstrate Asian monsoon failure and mega drought during the Last
Millennium.
Chapter 1 Introduction
11
1.6.2. In Indian Context
The history of tree-ring proxy based studies in Indian subcontinent have been
recorded since long back (Gamble 1902), but their applications are restricted mainly
to forestry aspects. But the beginning of systematic work on tree-ring analysis was by
Bhattacharyya et al. (1992) who explored potential trees for dendrochronological
analysis. Subsequently, a great deal of work; from this part of the globe especially
from the Himalayan region has been carried out by several workers.
1.6.2.1. Western Himalaya
Before 1990’s except climatic reconstruction by Hughes and Davies (1987),
most of the tree ring analyses were on the selection of sampling sites and the
evaluation of tree species suitable for dendroclimatic analysis. Bhattacharyya et al.
(1988) evaluated the potential of six conifers, in the Jammu and Kashmir region. They
demonstrated that two conifers, Cedrus deodara and Pinus gerardiana, exhibited high
age (up to 500 years) and the tree-ring chronologies were suggestive of a drought
response. Climatic reconstructions of spring and summer mean temperature and
precipitation based on well-replicated samples of Abies pindrow and Picea smithiana
were conducted in the Kashmir valley (Hughes and Davies, 1987). Bhattacharyya and
Yadav, (1989b) reported that Cedrus deodara growing in Joshimath, Uttarakhand
attains great age and its growth is inversely related to pre-monsoon temperature and
positively related to precipitation during both summer and winter. Subsequently, there
were a considerable number of studies on the reconstruction of the pre-monsoon
temperature based on tree-ring data of Cedrus deodara either individually
(Borgaonkar et al., 1996; Yadav et al. 1999; Yadav and Singh 2002a) or in
combination with Pinus wallichiana and Picea smithiana (Yadav et al., 1997, 2007).
It is salient that none of these studies produced century-scale negative temperature
anomalies which could be due to a regional impact of the Little Ice Age. Though most
of the climatic reconstructions were restricted to temperature, a few studies examined
the hydrological conditions. Reconstruction of precipitation in the non-monsoon
months (previous October to current May) back to AD 1171 revealed that the wettest
and the driest non-monsoon months occurred in the 14th and the 13th century,
respectively. Both wet and dry spring years were noted during the Little Ice Age
Chapter 1 Introduction
12
(Yadav & Park, 2000; Singh and Yadav, 2005; Singh et al., 2006). Besides ring width,
other tree-ring parameters, e.g., isotopic ratio and wood density have a promising
potential for dendroclimatology in this region. Isotopic analysis of the tree-ring
cellulose extracted from Abies pindrow growing in Gulmarg, Kashmir, revealed that
δ18O was most sensitive to precipitation and mean maximum temperature, whereas
δ13C was sensitive to temperature and δ18O to the amount of clouds and humidity
(Ramesh et al., 1985). In an exploratory analysis of Cedrus deodara at two sites in the
western Himalaya, Pant et al. (2000) suggested that density parameters, viz.,
earlywood, latewood, minimum, maximum, and mean densities, as well as total ring
width may be useful for dendroclimatic studies. Besides temperature or precipitation,
other aspects of environmental issues were also dealt with in some tree-ring studies
The applicability of tree-ring data in palaeo-seismic dating has also been explored.
Tree-ring data of Pinus wallichiana in Agora, Uttarkashi, have been studied to
evaluate the effect of the 1991-earthquake on tree growth (Yadav and Bhattacharyya,
1994).
1.6.2.2. Eastern Himalaya
In comparison to the western Himalaya, tree-ring studies in the eastern part of
the Himalaya are less. Seven conifer species, viz., Abies densa, Juniperus indica,
Larix griffithiana, Pinus roxburghii, P. wallichiana, Taxus baccata and Tsuga
dumosa, growing at diverse ecological sites have been recorded suitable for
dendroclimatic analysis (Chaudhary et al., 1999). Subsequently,a short chronology of
Larix griffithiana, a sub-alpine deciduous conifer growing in Sange, Arunachal
Pradesh, has been recorded suitable for reconstruction of May-temperature
(Chaudhary and Bhattacharyya, 2000). In an another study Pinus kesiya, growing in
and around the Shillong plateau are found not much old and these trees do not show a
common response to climate of Shillong (Chaudhary and Bhattacharyya, 2002).
Maiden report on climate reconstruction from this region is based on tree-ring data of
Abies densa combined from two sites, T-Gompa, Arunachal Pradesh and Yumthang,
Sikkim. This reconstruction (July-September temperature) extended back up to AD
1757. The warmest and coolest 10-year periods are 1978–1987 (+0.25 °C) and 1801–
1810 (-0.31 °C) respectively (Bhattacharyya and Chaudhary, 2003). Buckley et al.
(2005) showed a strong correlation of tree ring data of Pinus merkusii growing at
Chapter 1 Introduction
13
Arunachal pradesh with tropical Indian and Pacific Ocean bands in seasons preceding
the summer monsoon. Tree-ring data at this region have also been recorded suitable in
palaeo-seismic dating. The feasibility of such studies has been analyzed by
Bhattacharyya et al. (2008). In Abies densa, growing at two distantly located sites in
the north-east Himalaya, Yumthang in Sikkim and T-Gompa in Arunachal Pradesh,
the annual growth rings were narrow either in the same year of high intensity
earthquakes or in the subsequent year when an earthquake occurred in this region
during the non-growing season. Recently, a comparative analysis of tree-ring data of
both northeast and northwest Himalayan trees was pursued and evaluated the
suitability taxa and sites for tree-ring studies and climate reconstruction (Shah, 2007).
From the overall review of published data of the Himalayan region I have
recorded that the tree-ring data of the north-east Himalaya, in general were less
sensitive to climate variation in comparison to the western Himalayan sites and trees.
The most salient feature of this analysis was towards understanding climatic
variability during the Little Ice Age and towards linking monsoon with sea-surface
temperature and sea-level pressure in a global perspective. There are several
publications where tree-ring data of the Indian Himalayas have been compared with
those of the Tibetan Himalayas (Guo et al., 2009; Shao et al., 2009; Tian et al. 2009;
Zhang et al., 2009) and report that there was coherence in climate trend in both
regions.
1.6.2.3. Peninsular India
In tropical forest of south and Central India a numbers of groups have been
working to establish good quality tree-ring data network to understand monsoon
variability and related global parameters (e.g. ENSO) in the recent past. In this
context, teak (T. grandis) have been demonstrated as a potential source for high
resolution spatial reconstruction (Pant and Borgaonkar, 1983; Bhattacharyya et al.,
1992, 2007; Borgaonkar et al., 2001, 2010; Shah et al., 2007; Ram, S. 2008, 2010).
These studies indicate a great potential of T. grandis in reconstruction of monsoon
precipitation. It was also observed that the direct influence of temperature and rainfall
was not significant and they investigate the exact role of moisture and rainfall in tree
growth process and their relationship with regional summer month’s moisture index
Chapter 1 Introduction
14
and global parameters such as ENSO using wide network of teak tee-ring data over
the region of Kerala (Ram et al., 2008).
Shah et al. (2007), based on ring width data of teak, reconstructed mean
monsoon precipitation of June-September back to AD 1835 from Hoshangabad,
Central India. The reconstructed climate records show several alternating periods of
high and low monsoon episodes. Many of these low monsoon years have been shown
to coincide with most of the known principal drought years in India. Besides the ring
width, the size of vessels in dated tree-ring sequences of teak has also been found
suitable for climatic analysis. Besides ring width, Bhattacharyya et al. (2007) studied
early wood vessels of teak through image analysis of dated tree rings at
Perambikulam, Kerala, and found that rainfall during October and November
(northeast monsoon) of the previous year and April of the current year is the most
important climatic variable limiting the early wood vessel area. Based on the mean
vessel area of early wood, the northeast monsoon of this region was reconstructed for
the period AD 1743 to 1986 AD.
1.6.2.4. Dendrohydrology
So far only two studies regarding tree-ring based river discharge
reconstruction were available from the western Himalaya (Shah et al., 2013; Singh et
al., 2013). Using tree-ring data of Cedrus deodara (Deodar) growing within the Beas
river basin, Kullu valley, Himachal Pradesh, Stream flow for March-April has been
reconstructed using a simple linear regression transfer function model which goes
back to AD 1834.
In another study (Singh et al., 2013) based on combined tree-ring data of of
Pinus gerardiana (one site) and Cedrus deodara (three sites) from moisture stressed
sites in Kinnaur, the Western Himalaya developed 711 years (AD 1295-2005) long
previous year December to current year July Satluj discharge. The reconstruction
revealed 50-year low and high river discharge happened in the eighteenth and
nineteenth century, respectively. The decreasing tendency in the river discharge
noticed since the 1990s is consistent with the decreasing trend in winter precipitation
in the region.
Chapter 1 Introduction
15
1.6.2.5. Dendroglaciology
Tree-ring data of some conifers growing of the upper tree-line were used in
studying glacial behavior of the Himalayan region. Pinus wallichiana from Kinnaur
(Bhattacharyya and Yadav, 1996) and Abies pindrow growing close to the snout of the
SD, standard deviation; MS, mean sensitivity; AC1; Autocorrelation ; Rbt, mean inter-series correlation; EPS, expressed population signal; PC#1, percent variance explained by the first principal component (PC#1); NR, signal-to-noise ratio.
Chapter 2 General Principles & Supporting data
20
Cross dating is the process by which the variation in ring width characteristics is
examined to determine whether synchronity exists between series from the same tree
and different trees. If the patterns match then the series are coeval. It is the
fundamental step in any dendrochronological study as it assures that the correct
calendar date is assigned to all rings by identifying and accounting for any missing or
“absent rings”, "false rings" or unusual or indistinct ring boundaries (Stokes and Smi-
ley, 1968). Tree ring are dated through skeleton plot dating. The problem of dating
tree ring sequences in this region due to the presence of double or missing ring is not
serious as these are easily detected. It has been recorded that the presence of missing
ring is more prevalent in trees growing at lower elevations. Thus their occurrence is
more common in Juniperus and Betula .
2.4. Standardization of tree ring data and chronology preparation The computer program (Auto Regressive Standardization) ARSTAN (Cook,
1985) has been used to prepare tree-ring chronologies from the set of well cross dated
tree-ring measurement series. Program ARSTAN produces chronologies from tree-
ring measurement series by detrending and indexing (standardizing) the series, then
applying a biweight robust mean estimation of the mean value function to remove
effects of endogenous stand disturbances. Robust mean has advantage over the
arithmetic mean. The arithmetic mean is no longer minimum variance estimate of the
population mean when outliers are present, and is not guaranteed to be un-biased. In
contrast, robust mean automatically discount the influence of outliers in the
computation of mean, and thus, reduces the variance and bias caused by outliers
(Mosteller and Tukey, 1977). ARSTAN produces three types of chronologies,
"Standard", Residual" and "ARSTAN" (Holmes, 1992). In the "STANDARD"
chronology detrending of measurement series is done first by fitting a curve to model
biological growth trend to each series, and dividing out the growth model. The
chronology is then computed as a robust estimation of the mean value function to
remove effects of endogenous stand disturbances and it enhances the common signal
contained in the data. In the "RESID" chronology autoregressive modelling is
performed on the detrended ring measurement series. Robust estimation of the mean
value function produces a chronology with a strong common signal and without
persistence. The "ARSTAN reincorporates into the residual chronology the
Chapter 2 General Principles & Supporting data
21
persistence structure of a pooled model of autoregression in the entire group of ring
measurement series (Holmes, 1992). The program provides various options of
detrending methods viz., Negative Exponential, Linear Regression, Regional Curve
Standardization, Hugershoff Growth Curve, Cubic Spline smoothing etc. along with
the detailed statistics of each series and chronology.
In the present dissertation, Double-detrending approach used, first a negative
exponential curve, a linear regression or a horizontal line passing through the mean
was used to remove any age-growth trends (Fritts, 1976). Following this, the series
were detrended a second time to reduce the impact of abiotic factors on radial growth
(e.g., competition and defoliation) with a 30 years smoothing spline curve used to
remove biological trend from data because these methods are appropriate for the trees
growing in forest. Both detrending methods are believed to preserve low-frequency
climate variability (Fritts, 1976). Both standard (detrended index) and residual (index
derived from autoregressive modeling) series were included in further analyses. In
chronology development, the standard technique used in dendrochronology has been
used (Fritts, 1976; Hughes et al., 1983; Schweingruber, 1987; Cook and Kairiukstis,
1990). The first step is the removal of overall trends in tree ring sequence to enhance
climate signal. These trends result due to ageing effects of trees and exogenous
disturbances, the exogenous disturbance could be caused by several factors such as
2000 North Sikkim Grid *(http://www.cgd.ucar.edu/c
as/catalog/climind/pdsi.html)
26.25/ 88.75 1,21 1850-2000
*Indian Meteorological Department (IMD) *The CRU TS 2.1 Climate Dataset has been produced by the Climatic Research Unit (CRU). *(http://www.cgd.ucar.edu/cas/catalog/climind/pdsi.html) for Palmer Drought Severity Index data (PDSI).
0.00
100.00
200.00
300.00
400.00
500.00
600.00
700.00
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Month
Prec
ipita
tion(
mm
)
0.00
5.00
10.00
15.00
20.00
25.00Te
mpe
ratu
re(O
C)
PPT MIN MAX MEAN
Fig.2.2.Monthly variation of total precipitations (bars), mean maximum temperature
(red line), and mean temperature (green line) and mean minimum temperature (pink
4 E.Tien Shan China Gl. #1, E.Br. CN0010 622 43°05' 86°49' 4224 NA 3736 2.2 1.163 NE/NE 1988-1995
5 E.Tien Shan China Gl. #1, W.Br. CN0010 622 43°05' 86°49' 4476 NA 3795 1.95 0.677 NE/NE 1988-1995
Explanation of legends: glacier name, country, region, PSFG number (five digits identifying glacier with minimum elevations (meters), area (the latest information on total area of glacier, in km2), length (the latest information on glacier length, in km), aspect (for some accumulation area-denominator, and for ablation area- numerator) glaciers it is given separately for accumulation area- denominator, and for ablation area- numerator)
Chapter 2 General Principles & Supporting data
30
2.6.2. Bootstrap Response Function
In the present dissertation, “Bootstrap Response Function” and “Correlation
Analysis” have been used to establish tree-growth climate relationship. Bootstrap
response function in the program suite called “3Pbase” (Guiot and Goeury, 1996)
have been used to analyse tree growth climatic relationships. In tree-ring analysis this
method offers the advantage of avoiding errors caused by collinearity among
variables and providing a more realistic estimate of tree response to climate. The
bootstrapping is a very useful technique developed by Efron (1979) to estimate
statistics for unknown population distributions by techniques of Monte Carlo
simulations. The mathematical details are given in Efron (1982), more accessibly in
Efron and Gong (1983), and popularly in Diaconis and Efron (1983). This technique
involves resample of original data matrix to form a new data matrix. Afterward,
another year is randomly drawn and the corresponding data vector is stacked next in
this matrix. In this randomized combination, called ‘‘pseudo-data set’’ matrix, same
year may be drawn several times. A regression is then performed on this pseudo-data
set. This methodology is repeated 200 times, giving 200 regressions and 200
predictions. Summary statistics on the regression coefficients and multiple
correlations were obtained by calculating means and standard deviations for the 200
samples. The bootstrap regression coefficients are judged significant at the 95% level,
if they are twice in absolute value of their standard deviation (Guiot, 1991). The
statistical significance for the correlation between tree-ring and climate are judged at
95% significance level. Equivalent time spans (1966-2000) of mean monthly
temperature, maximum temperature, minimum temperature and monthly precipitation
of Gangtok, (Sikkim) and tree-ring indices and climate variables data were taken.
These climatic variables were taken for the twelve-month period i.e., from November
of the previous year to October of the current year’s.
2.6.3 Bootstrap Transfer Function
Bootstrapped orthogonal regression was also used for transfer function because
Bootstrapped orthogonal regression (Guiot, 1991) was used for estimating tree-
growth/climate relationship seems to work better than cross-validation in many cases
(Efron, 1983). In the simplest form of bootstrapping, instead of repeatedly analyzing
subsets of the data, you repeatedly analyze subsamples of the data. Each subsample is
Chapter 2 General Principles & Supporting data
31
a random sample with replacement from the full sample. Depending on what you
want to do, anywhere from 50 to 2000 subsamples might be used. There are many
more sophisticated bootstrap methods that can be used not only for estimating
generalization error but also for estimating confidence bounds for network outputs
(Efron and Tibshirani 1993).
2.6.4. Correlation Analysis
The Pearson product-moment correlation coefficient is probably the single
most widely used method for summarizing the relationship between two variables.
Under certain assumptions, the statistical significance of a correlation coefficient
depends on just the sample size, defined as the number of independent observations.
If time series are auto correlated, an “effective” sample size, lower than the actual
sample size, should be used when evaluating significance. Finally, it should be
emphasized that the Pearson correlation coefficient measures strength of linear
relationship. Scatterplots are useful for checking whether the relationship is linear.
“Correlation analysis” has also been used for Climate–growth relationships between
tree-ring indices and with climate variables, i.e. maximum temperature (MAX),
minimum temperatures (MIN) and mean or average temperature (MEAN) and
Precipitation (PPT) were taken into account as predictors.
2.6.5. Linear Regression Method for Climate reconstruction Another method, used for climate reconstruction is linear regression method.
First of all, the correlations between the chronology and Gangtok climate data of the
previous November to current year October was calculated. After identifying
significant relationship with tree growth and climate, a regression model has been
developed for reconstruction of climate.
Chapter 2 General Principles & Supporting data
32
2.7. Dendrohydrological modeling Both, Correlation (Tree growth/discharge relation) and Linear Regression
(discharge reconstruction) analyses have been performed for dendrohydrological
modeling. By using correlation method it is determined which months are best
associated with tree-ring width variability and discharge data of these selected months
could be reconstructed through regression analysis. In this dissertation, discharge
reconstruction has been done by using one conifer species viz. Abies densa (ABDE)
(56 cores from 36 trees). These samples were from two localities of this region viz.,
Zema, at the altitudinal ranges from 2,599 m a.s.l to 2,804 m a.s.l (Fig.1). The details
of the samples collected along with site characteristics from three different localities
are given in details in Chapter 3
2.7.1. Rivers discharge data Sikkim is drained by number of Perennial Rivers. However, the two main river
systems are Teesta and Rangit. The Zema Chu (Chu means river) is one of the
tributary of the Teesta river originating from Zemu glaciers (elevation of about 5200
m above m.s.l) North Sikkim. Discharge data for Zemu Chuu recorded in Lachen
gauge stations were obtained from National Hydroelectric power corporation
(NHPC). The available discharge data is extended from 1996 to 1997. The analysis of
annual river discharge of this river at Lachen station shows it has annual water yield
of 637.9 cu.mt/sec, in which the maximum and minimum discharge recorded as
1209.82 cu.mt/sec and 300.59 cu.mt/sec in the year 1984 and 1992 respectively.
Detailed analysis of this data has been discussed in chapter 7.
2.7.2. Climate data for discharge site
There is a lack of spatially disaggregated meteorological records in Sikkim.
Due to the remote location of the study site, no single climate station could be
assumed to be representative of the local climatic conditions. To study the spatial
correlations, maximum temperature (MAX), minimum temperatures (MIN) and mean
or average temperature (MEAN) and Precipitation (PPT) climate data from the
Climate Research Unit (CRU TS 2.1, 0.5°×0.5°, 27°45′N 88°45′E). A PDSI grid
(26°25′N, 88°75′E and 1950–2000) was also chosen for analysis (Fig. 8.3a).
Chapter 2 General Principles & Supporting data
33
2.7.3. Correlation analysis for Tree-growth and discharge relationship
Discharge and climate/ growth relationships were determined by examining
the correlation coefficients between tree-ring indices with climate variables, i.e.
Maximum Temperature (MAX), Minimum Temperatures (MIN), Mean temperature
(MEAN) and monthly Precipitation (PPT) (details in chapter 5) as well as with
discharge were taken into account as predictors. Correlations with Lachen discharge
were positively significant for January-April (r = 0.734., p =0.01). Analysis of the
climate-growth relationship showed a general positive correlation between the radial
growth of ABDE From November to October of the growth year.
2.7.4. Discharge Reconstruction Method
2.7.4.1. Linear regression method for discharge reconstruction Correlations between the chronology and discharge data from the previous
November to current year October has been calculated. Significant correlations were
found only for January to May discharge of the current year (P<0.01). Mean of
January to April discharge were taken into account as predictors, regression model
has been for reconstruction of discharge. No significant correlations between tree-ring
data and precipitation were found. However, the correlations between the PDSI and
chronology were positively associated for most months, but the significant
relationship was recorded only for December and January.
2.8. Correlation analysis for tree growth and its relation with PDSI/
El Nino Correlation analysis between the site specific tree-ring chronologies as well
regional chronologies (developed through PCA) were analysed for relationship with
Palmer Drought Severity Index (PDSI), E Nino 3.4. Palmer Drought Severity Index is
widely used for of meteorological drought over land regions (Dai et al., 2004). The
2.50 X2.50 gridded PDSI values were extracted from the Palmer Drought data set
(http://www.cgd.ucar.edu/cas/catalog/climind/pdsi.html) for one grid points close to
the tree ring sampling locations in the North Sikkim.
Chapter 2 General Principles & Supporting data
34
E Nino 3.4 monthly data was collected from (KNMI Climate Explorer
program (http://climexp.knmi.nl). Details analysis of this data has been discussed in
chapter 8.
2.9. Multiple tree-ring proxies (Earlywood width, Latewood width) Attempt has been made to investigate the potential of using multiple tree ring
parameters in addition to ‘traditional’ ring width data (TRW) for palaeoclimatic
studies, Here, the climatic signals of multiple tree-ring parameters, that are latewood
width (LWW), and earlywood width (EWW) are assessed and correlation of these
parameters with the regional temperature and precipitation data has been done. The
annual Larix increment consists of two, visually well distinguishable parts. Earlywood
(light coloured with large vessels) develops during spring, while latewood (dark
coloured, denser, lacks large vessels) is formed during the latter part of the growing
period. The earlywood width (EWW) and latewood width (LWW) have been
measured and analysed for their response with climate.
Chapter 3 Study area…….…. sample processing
35
3.1. Vegetation Overview of Zemu valley
Zemu glacier is located in the northern part of Sikkim in the Eastern Himalaya
(Fig.3.1). The vegetation cover around this glacier has been analysed using
September month’s Landsat-5 satellite data of 2001 (Fig.3.2) NDVI (Normalised
Difference Vegetation Index) calculated using NIR (Near Infra Red) and red bands of
ETM (Enhanced Thematic Mapper) sensor gave a picture of vegetation cover around
the glacier (Fig. 3.3). The present brief account of the distribution of trees based on
information from Smith and Cave (1913) and my personal observation provide a
general idea of the distribution of trees in the region in terms of their climatic regime.
The Zemu glacier valley area may be divided botanically into three forest types.
These are temperate forest growing from 2,438 to 3,353 m a.s.l, a subalpine shrub
region from 3,353 to 4,267 m a.s.l, and an alpine region from 4,267 to 5,182 m a.s.l.
Trees growing from lower elevation i.e., temperate forest (around 2000 m a.s.l) to
subalpine forest (3,000 m a.s.l ) close to snout of this glacier has been considered for
tree-ring analysis in the present dissertation. Frequent occurrence of rock fall, mass-
movement processes, such as debris flow or snow avalanches in the forest is a
common feature. Fieldwork for collection of samples for the present thesis was made
from seven different sites (Fig.3.8). Two of the study sites are located in close
proximity to Zemu glacier area, namely Yabuk, and Jhakthang. The other five sites
investigated are located in and around Zemu area. Sampling was done starting from
Lachen (2,753 m a.s.l) at lower elevation to Yabuk (3, 953 m a.s.l), higher elevation.
3.2. Tree-ring sampling sites of North Sikkim
One of the objectives of this dissertation is the development of well-replicated
climatically sensitive tree-ring chronologies from the Eastern Himalayan region. For
that I have chosen Zemu glacier valley at altitude gradients from lower elevation up to
the snout of the glacier so that I could extend the existing tree-ring data network of the
Sikkim Himalayan region. For that, a good number of tree-ring samples were
collected during two field trips in the year 2009 and 2010 at this region. Tree-rings
Chapter 3 Study area…………. sample processing
36
cores from the living trees and sections from left over stumps were collected for the
present study. Sampling were made through a transect (Fig.3.9) starting from Lachen
I reached to Zema covering the distance around 3 kilometers by trekking. From there I
proceed to Talem.
Fig.3.1.Map of India showing position of Zemu glacier and adjoining area in Sikkim
Chapter 3 Study area…………. sample processing
37
Fig. 3.2. (a) Satellite map of Zemu glacier and (b) Zemu glacier along with
Vegetation Cover in and around this region based on NDVI (Normalised Difference
Vegetation Index)
Zemu glacier is about 11 kilometer from Talem. Besides, samples were
collected from two other places i.e. from Zakthang and Yabuk in between Talem and
snout of the glacier. Zakthang is 2-3 Km from Talem and from Yabuk to Zemu glacier
is about 3-4 kilometer (Fig. 3.4). Tree-line is located at Yabuk which is close to the
Snout of the Zemu glacier. Juniper Squmata scrubs are flourishing close to glacier
snout. It had been quite a hard task to find sufficient aged trees as most forest are
disturbed due to human activities, and natural disasters. Some undisturbed forest sites
are in much remote areas, but those were beyond my reach with existing facilities.
Chapter 3 Study area…………. sample processing
38
Even that, after careful selection of trees a large number of tree cores were collected
from trees growing in a wide variety of ecological settings. These tree cores or
samples were collected through Increment Borers. Generally two cores in opposite
direction at the breast height of the trees were collected. In many cases it was not
possible to collect more than one core, as the other side was not approachable due to
steep slope. In some cases, to get longer samples from the active portion of the trunk,
more than one core were collected where tree had heart rot inside. Depending upon
the availability of suitable trees in the area, generally 10-20 trees were sampled. In
some cases only few trees were sampled with a view to understand the
dendrochronological potentiality of the taxa and site. In this dissertation, for the
convenient of analysis each of these tree-ring sites are abbreviated by its first three
letters. Similarly, name of taxa is abbreviated by four letters in which first two letters
are of generic and later two are from specific name. A List of the tree-ring samples
collected from different sites along with the site and species code and number of cores
collected are shown in (Table 3.1). Total 445cores from 236 trees were collected
from two conifer trees species viz. Abies densa, Juniperus recurva, and one broad
leaved taxa Betula utilis Beside 36 wood stump and 2 wood stump of Juniperus
squamata, Betula utilis were also collected (Fig.3.7). These samples were from
several localities of this region viz., Lachen, Dozom Khola, Talem, Zakthang and
Yabuk, at altitudinal ranges from 2599 to 3953 m.a.s.l. (Fig.3.1).The details of the
samples collected along with site characteristics from different localities are given
below.
3.2.1 Lachen (means ‘big pass’) is a town in North Sikkim and located at an elevation
of 2,753 m a.s.l.. This town is located about 129 km from Gangtok. This site is amidst
mixed conifers and Rhododendron forests. Abies densa growing along with Lachen
river, a tributary of the Teesta River (Fig.3.3). It is also known as Lachen-chu. Tree-
ring samples are collected from the Abies-Rhododendron forest above Lachen
Township. From this forest site, 57 tree cores from 29 trees of Abies densa and 29
trees and 50 core of Larix griffithiana were collected. It was found that most of the
Chapter 3 Study area…………. sample processing
39
trees in this locality are rotten from inside. (Alt of 2,735 m a.s.l). The town forms the
base to the Chopta Valley and Gurudongmar Lake.
Fig. 3.3. Sampling site Lachen showing forest of Abies densa
Chapter 3 Study area…………. sample processing
40
3.2.2. Zema
This place is by the side of the Zema river (Zemu chu). Crossing the Zemu
Chu to westward up the Zemu valley through the forest, the route is quite steep and
one has to walk on path which is broken here and there due to landslides and those
have to be crossed by temporarily made log bridges. Here forest is of mixed conifer
type, especially Abies densa with Rhododendron. From this forest site, 78 tree cores
from 38 trees of Abies densa are collected.
3.2.3. Dozom Khola
This site is on the way to Talem at 3125 m.a.s.l. Here trees appear to be not
much old. From the young patch of Abies densa, Only 4 cores from 3 trees were
collected.
3.2.4. Talem (TAL)
After crossing Zema II trekking through steep slope by the side of the zema
chu I reached Talem. This site is on the left bank of the Zemu chu at an altitude of
3157 m a.s.l. The forest is mostly covered by Abies and Betula with thick
undergrowth (Fig.3.4). Tree cores were collected from 47 trees, in which 56 cores are
from 33 trees of Abies densa and 28 cores are from 14 trees of Betula utilis. Most of
the huge girth trees of Abies densa in this site are also found rotten from inside.
3.2.5. Jakthang (JAK)
This site is on the left bank of the Zemu chu at an altitude of 3503 m a.s.l. The
dominant conifer taxa of this forest is Abies densa associated with few Juniperus
recurva and few Betula utlis. From this forest 66 cores from 34 trees of Abies densa
and 26 cores from 13 trees of Juniperus recurva were collected (Fig.3.5). Some of the
trees are found damaged by boulders and covered by thick moss cushion and rotten
due to fungal activities.
Chapter 3 Study area…………. sample processing
41
Chapter 3 Study area…………. sample processing
42
3.2.6. Yabuk (YAB)
This site (3, 953 m a.s.l) lies in the northern part of North Sikkim District at an
altitude of 3953 m a.s.l. It is about 18 km from Green Lake, on the Lachen- Gangtok
trek route. From Jakthang to Yabuk route there are several ascents and descends on
the route. This path is often wiped out of floods, covered by boulders and pebbles by
landsides. Being close to glacier this site experience very low temperature and high
speed wind. Tree-ring sampling: Abies- Betula –Juniperus site (Fig.3.6). A total
number of 208 tree-ring cores from 110 trees (100 cores from 52 trees of Abies densa,
25 cores from 13 trees of Betula utilis and 83 cores from 45 trees of Juniperus recurva
were collected from this site. In this site except Betula utilis and Juniperus recurva
other one taxa Abies densa, collected for tree-ring analysis, was found mostly rotten
Chapter 3 Study area…………. sample processing
43
inside. Most of the trees with huge girth were rotten from inside. New seedling of
Abies densa growing at upper line of tree, which suggest the migration of trees to
higher altitude in response to global warming.
Fig.3.6.Sampling site Yabuk showing forest of Juniperus recurva and Abies densa
Chapter 3 Study area…………. sample processing
44
3.2.7 Yumthang (YUM)
This site is 24 km. north of Lachung at an altitude of 3,880 m a.s.l. The area is
characterized by sub-alpine forest in which taxa are Abies densa and Rhododendron
conspicuous. These trees are growing on moderately steep slopes forming high stand
density. Samples were collected from ABDE growing 50 m. up slope from Yumthang
Forest Guest House, and attain 10 to 15 m. height and 1-2.5 m. in girth. 46 cores from
27 trees were collected.
3.3. Zemu Valley (Zemu glacier IN5020105032)
Zemu glacier is the largest glacier in the eastern Himalayas with a total surface
area of about 42 km2 and a liner length of about 20 km (Fig. 3.8). It is situated
between the north latitude 270 40' N and 270 41' N and 880 10' E and 880 23' E. The
glacier originates from the eastern slopes of the mount Kangchenjunga (8,536 m a.s.l),
world third highest peak. The glacier can be approached from Lachen in north
Sikkim, which in turn is connected to Gangtok, (capital of Sikkim) by an all weather
road. This glacier initially, flows towards north east changing to easterly course for
Fig.3.7(a)Sampling site Yabuk showing zone of Juniperus squmata scrub
(b) Collection of disc sample of Juniperus squmata (disc) from left over stumps at Yabuk.
Chapter 3 Study area…………. sample processing
45
the greater part of its flow downstream with an average gradient of 1 in 10 m. It is fed
by as 12 tributary glaciers from the southern side. Since last several decades the
glacier was under continuous observation of the GSI and is reported to have vacated
an area of snout 52,443 km2 along the snout front, from 1965 onwards with in
interruption of periods 1979-80 and 1984-1985 when the glacier snout has been
reported to have shown slight advancement.
a.
Chapter 3 Study area…………. sample processing
46
Fig.3.8. (a) Satellite map showing location of sample collection site along with
meteorological station and (b) Sketch map route of collection of samples from ZEMA
to Zemu glacier site. (The details of the abbreviations of sites and trees are given in
Table 2.2).
3.4. Sample Processing The cores were mounted in grooved wooden stick through glue in such a way
that the marking line (sheen caused by the shearing action of the spoon) on the sides
of core and edges of grooves remain parallel vertically (Fig 3.10). The specimen
number and such notes as name of tree and site and date of collection were written on
the mount. The surface of the cores were cut by sharp edged blade and polished using
several grades of sand papers that to make surface smooth for the rings become
visible to study under stereo zoom binocular microscope.
b.
Chapter 3 Study area…………. sample processing
47
Fig.3.9 Collection of tree cores from the tree through increment borer
Fig. 3.10. Processing of tree-ring cores.
Chapter 4 Building of Tree-Ring Chronologies
48
4.1 Building of Tree-Ring Chronologies In sum total 445 cores from 236 trees were collected from three different
conifer trees species viz. Abies densa, Juniperus recurva, and one broad leaved taxa
Betula utilis. Besides 36 and 2 wood stump of Juniperus squamata and Betula utilis
respectively were also collected from Zemu valley. These samples were from several
localities of this region viz., Lachen, Dozom Khola, Talem, Zakthang and Yabuk, at
the altitudinal ranges from 2599 to 3953 m a.s.l. (Fig. 3.8.). Due to dating problem
some samples tree ring data from all trees and sites could not be used. Thus in this
dissertation, the tree-ring data of ABDE from all sites and JURE data of one site
considered, other site of JURE from JAK were not studied. The same problem with
Betula utilis has been recorded although samples were collected from several sites
like, YAB, JAK, and TAL. These remaining works will be taken up later.
In the present dissertation total 12 tree-ring chronologies comprising ten of
ring-width and one each of late wood and early wood tree ring chronologies were
prepared. In the preparation of these chronologies, the computer programme
“ARSTAN” has been used which produces three versions of chronologies
"STANDARD", “RESIDUAL" and "ARSTAN", intend to contain a maximum
common signal and minimum amount of noise in the tree ring series. In the
"STANDARD" chronology detrending of measurement series is done first by fitting a
curve to model biological growth trend to each series, and dividing out the growth
model. The chronology is then computed as a robust estimation of the mean value
function to remove effects of endogenous stand disturbances and it enhances the
common signal contained in the data. In the "RESIDUAL" chronology autoregressive
modelling is performed on the detrended ring measurement series. Robust estimation
of the mean value function produces a chronology with a strong common signal and
without persistence. The "ARSTAN reincorporates into the residual chronology, the
persistence structure of a pooled model of autoregression in the entire group of ring
measurement series. These chronologies prepared are mostly from Abies densa, and
one each from Juniperus squmata, Larix griffithiana and Juniperus indica. These
chronologies along with samples size are shown in (Fig. 4.1). These are from sites
viz., YAB, JAK, TAL, DOZ, ZEMA, LACH, and YUM (Fig. 3.8.). Six chronologies
of Abies densa each from sites YAB, JAK, TAL, DOZ, ZEMA, LACH. One
chronology of Abies densa from YUM. One chronology of Juniperus squmata is from
Chapter 4 Building of Tree-Ring Chronologies
49
YAB. Beside, one chronology of LAGR from Lachen, two chronologies of late wood
and early wood of LAGR. Time span of chronology prepared from Abies densa of
Zema is extends from AD 1628 to 2007(380 years), From Jakthang, extending from
AD 1700 to 2010 (311 year), Talem AD 1678-2010 (333 years) and Yabuk extends
from AD 1759 to 2010 (252 years). Besides, tree rings the shortest chronology
prepared is one from stumps of JUSQ, AD 1881-2010, (140 years). The longest
chronologies in these lists are JURE which extend from AD 1556 to 2010 (456 year).
Samples of LAGR collected from LAC were less in number (only 9 cores) but these
samples exhibit good cross dating with samples of same species collected from
Lachung (approx. 40 kms apart). These two chronologies were merged together to
form one chronology (LAC) representing LAGR of northern Sikkim region. These
chronologies along with samples size are shown in (Figs 4.1).
Fig.4.1.Ring-width index chronology of Abies densa from Yabuk
Fig.4.2.Ring-width index chronology of Juniperus squmata from Yabuk
Fig. 5.1.Plot of (a) Response function and (b) Correlation analyses based on standard
chronologies of tree ring indices versus monthly climate data (mean temperature,
minimum temperature, maximum temperature and precipitation) for ABDE_YAB.
Horizontal pink line indicates significance level (p < 0.05) and red line indicate
significance level (p < 0.01) above and below.
However, in the Correlation analysis of ABDE_YAB (Fig. 5.1b) a positive
correlation was recorded with maximum and mean temperature of March-April
(p<0.01) where as negative correlation was found with minimum and mean
a.
b.
Chapter 5 Dendroclimatic modeling
59
temperature of May–September (p< 0.05) but it is highly significant only for mean
temperature of July-August (p<0.01). In case of precipitation, March was found
negative while May is positive. Thus, Correlation Analysis exhibits almost similar
results that of Response Function Analysis. Moreover, in both analyses, a positive
relationship with maximum temperature of March-April, and negative relationship
with mean temperature of July-Aug (p < 0.01) were noted.
In case of JUSQ_YAB, in RFA, a positive relationship was noted with
precipitation of May–June, where as it was negatively associated with maximum and
mean temperature of January (Fig.5.2a).
But in Correlation Analysis, only positive significant relationship was recorded with
precipitation of May but it was negative with maximum temperature of January (Fig.
5.2b).
a.
-0.400
-0.300
-0.200
-0.100
0.000
0.100
0.200
0.300
0.400
pNOV pDEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT
Corre
latio
n co
effic
ient
s
JUSQ_MEAN JUSQ_MIN JUSQ_MAX JUSQ_PPT 95%CL 95%CL
-3.000
-2.000
-1.000
0.000
1.000
2.000
3.000
pNOV pDEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT
Regr
essi
on c
oeffi
cent
/SD
JUSQ_MEAN JUSQ_MAX JUSQ_MIN JUSQ_PPT 95CL 95CLb.
a.
Fig.5.2. Plot of (a) Response function and (b) Correlation analyses based on standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for JUSQ_YAB. Horizontal pink line indicates significance level (p < 0.05) above and below.
Chapter 5 Dendroclimatic modeling
60
In case JURE_YAB; RFA revealed a negative relationship between tree
growths and mean temperature, minimum temperature of January but no significant
relationship existed with precipitation (Fig.5.3a). In case of Correlation Analysis, a
negative relationship was recorded with minimum and mean temperature of January
whereas it was negative with mean and maximum temperature of August-September
In the next site towards lower elevation, ABDE_JAK, in its RFA, a positive
and significant relationship was noted with mean and minimum temperature of March
while a negative relation was with minimum temperature of January (Fig.5.4a). On
a.
b.
Fig. 5.3. Plot of (a) Response function and (b) Correlation analyses based on standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for JURE_YAB, horizontal pink line indicates significance level (p < 0.05) and red line indicates significance level(p < 0.01) above and below
-0.400
-0.300
-0.200
-0.100
0.000
0.100
0.200
0.300
0.400
pNOV pDEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT
Corr
elat
ion
coef
ficie
nts
JURE_MEAN JURE_MIN JURE_MAX JURE_PPT 95%CL 95%CL
Chapter 5 Dendroclimatic modeling
61
the other hand in Correlation Analysis, a positive relationship with only mean and
minimum temperature of March (Fig 5.4b) was recorded.
-3.000
-2.000
-1.000
0.000
1.000
2.000
3.000
4.000
pNOV pDEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT
Regr
essi
on c
oeffi
cent
/SD
JAK_MEAN JAK_MAX JAK_MIN JAK_PPT 95CL 95CL
-0.400
-0.300
-0.200
-0.100
0.000
0.100
0.200
0.300
0.400
pNOV pDEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT
Corr
elat
ion
coef
ficie
nts
JAK_MEAN JAK_MIN JAK_MAX JAK_PPT 95%CL 95%CL
Fig.5.4. Plot of (a) Response function and (b) Correlation analyses based on standard
chronologies of tree ring indices versus monthly climate data (mean temperature,
minimum temperature, maximum temperature and precipitation) for ABDE_JAK.
Horizontal pink line indicates significance level (p < 0.05) above and below
In ABDE_TAL, RFA; a negative relationship was noted with maximum
temperature of previous year December and with precipitation, whereas a negative
and significant relationship with maximum temperature of May and September and
with minimum temperature of July-August were recorded. In case of precipitation
only January exhibits negative relationship (Fig. 5.5a). However, in case of
Correlation Analysis, a negative relationship with minimum temperature of previous
a.
b.
Chapter 5 Dendroclimatic modeling
62
November to current year February and mean temperature with previous November to
January were recorded. A negative relationship was with mean, maximum and
minimum temperature of May to September, except maximum temperature of January
though it was related but not significant (Fig.5.5b.) In case of precipitation a negative
Fig. 5.7. Plot of (a) Response function and (b) Correlation analyses based on standard chronologies of tree ring indices versus monthly climate data (mean temperature, minimum temperature, maximum temperature and precipitation) for ABDE_ZEM. Horizontal pink line indicates significance level (p < 0.05) above and below
Chapter 5 Dendroclimatic modeling
65
July-September temperature recorded (Fig.5.8a) While in Correlation Analysis, a
Negative Significant relationship for May on the other hand positive relationship with
mean temperature August at (p<0.05) level (Fig.5.8b) recorded.
Fig.5.14. Plot of (a, b) Response function and (c, d) Correlation analyses based on
rotated principal component (PC) scores [AD 1881–1994 for PC#1, PC#2,] versus
monthly climate data (mean temperature, minimum temperature, maximum
temperature and precipitation). Horizontal pink line indicates significance level (p <
0.05) and red line indicate significance level (p < 0.01) above and below.
5.3. Physiological explanation of tree growth climate relationship
The correlation analyses suggest that in this region the influence of rainfall on
tree growth is weaker than that of temperature. Precipitation at higher elevation sites
does not reach thresh hold limits to reduce tree growth as no significant correlations
were recorded between monthly precipitation and the tree-ring chronologies,
suggesting that moisture is not limiting factors for radial growth at tree line. Here
attempts have been made to find out logical explanation for tree growth climate
relationship in terms of physiolological aspects of trees.
5.3.1. Positive correlation with March to April temperature
At tree-line locations, the growth period is relatively short and most active
tracheid formation occurs at the beginning of the summer (Rossi et al., 2006).
Therefore, for production of wide tree rings is the combination of initiation of early
cambial activity with enough available resources, i.e., stored carbohydrates and other
substances produced in the previous growth period and favorable climate conditions
during the late spring and early summer months of the growth year is needed. The
d.
Chapter 5 Dendroclimatic modeling
75
effect of previous year climate is expressed by strong first-order autocorrelation often
found in tree-line trees (Fritts, 1976). A positive response to spring and summer
temperatures has also been found in a number of studies of P. cembra in the European
Alps (Carrer and Urbinati, 2004) and the Carpathian Mountains (Kern and Popa
2007).
5.3.2. Negative correlation with June to September:
Low temperatures in June and July could slow down the initiation of cambial
activity and first stages of cell division and development. In addition to this, unusually
low temperatures during this period can also slow down the formation of secondary
cell walls, and deposition of lignin, and can therefore be responsible for production of
light rings or narrow tree rings in extreme years (Gindl, 1999; Rossi et al., 2006). The
growth of conifer in the Eastern Himalaya is often controlled by late summer
temperature (Bhattacharyya et al., 2003).
Dendroclimatological studies of high-elevation conifer sites on eastern Tibetan
plateau and in Eastern Himalaya suggested that tree-ring chronologies were an
indicator of late summer temperature (Bräuning and Mantwill, 2004; Bräuning, 2006;
Bhattacharyya and Chaudhary, 2003). Similar results were reported by various studies
in nearby areas (Fan et al., 2009; Liang et al., 2008; Bräuning, 2006). Bhattacharyya
et al. (2003) Suggested that late summer temperature in Eastern Himalaya is vital for
growth of Abies densa. In the study area, sampled trees are assumed to be growth
limited by a single climate variable (usually temperature). Because of Forest
desiccation and reduced root activity due to low soil temperature that restricts tree
growth. Liang et al. (2008) suggested that mean summer minimum temperature was
the major limiting factor to tree-ring growth of P. likiangensis var. balfouriana in the
source region of the Yangtze River on the Tibetan Plateau. The growth of conifer in
the Eastern Himalaya is often controlled by late summer temperature (Bhattacharyya
et al., 2003). Abies densa sensitive to summer drought stress, as reflected by the
negative correlations with June and September temperatures. This suggests that tree
growth is limited in years with dry and hot summers. Such findings are remarkable for
tree-line sites, where tree growth mostly reacts positively to high summer
Chapter 5 Dendroclimatic modeling
76
temperatures (Korner, 1998; Buntgen et al., 2006; Carrer et al., 2004) leaves are not
frozen (Brauning, 2001; Chabot and Hicks, 1982; Havranek and Tranquillini, 1995;
Pederson et al., 2004). Highest temperature in the study area occurred from June to
August, which is earlier than the highest precipitation timing (from July to
September). That is to say, the higher the temperature is in the growth season, the
more intensive the soil evaporations and plant transpirations are, and this leads to
worse conditions for tree growth. The negative correlation between tree-ring width
and the mean temperature from June to September suggests temperature can be the
most limiting climatic factor of trees in Zemu area, by accelerating soil water
evaporation and tree’s transpiration Higher winter temperature might advance the
beginning of the growing season.
Chapter 5 Dendroclimatic modeling
77
5.4. Dendroclimatic modeling (Past Climate reconstruction) In general, in stressed environments sites, inter annual tree growth variations
according to changes in growth limiting climate variables are conspicuous. The
process of deriving a climate from tree growth and extending it beyond the
instrumentally measured climate or meteorological records is referred to as a
calibration. The outcome is a chronology of climate as old as the oldest preserved tree
in the tree ring series (Fritts, 1976). Details methodologies involved in the climate
reconstruction were discussed in the chapter 2.
To reconstruct the past climate variations, the instrumental records
(temperature and precipitation) were regressed against the regional chronology. Out
of 12 tree ring chronology made from four conifers, only two sites chronologies of
ABDE i.e. ABDE_YAB and ABDE_ZEM were recorded statistically reliable for
climate and others were statistically weak for reconstruction. In case of ABDE_YAB,
A clear and high significant positive relationship to maximum March-April and a
negative to mean temperature of July- August recorded. With precipitation, there were
no persistent patterns of correlation with the tree ring data among the all site
chronologies. For maximum March April temperature reconstruction, I applied only
Bootstrap method was applied because it qualifies the statistical significance level
successfully in both RFA and CA. For reconstruction of Mean July-Augst
temperature, I have used Linear regression method, because in this process the July-
August negative relationship are highly significant at 99% confidence level, even
though signals were also seen also in RFA but it just touched 95% level.
5.4.1. Bootstrap Method for March_April Maximum temperature reconstruction
Average March-April Maximum temperature has been observed significant
controlling growth of fir at tree line of Zemu glacier area in the both methods, i.e.
Bootstrap and Correlation. Since response function analysis using Bootstrap shows
better possibility for reconstruction (Table 5.3). Significant climatic variables limiting
tree growth and these climatic parameter have been selected for the reconstruction
through Bootstrapped transfer function (Guiot, 1991). Reliability of the reconstruction
Chapter 5 Dendroclimatic modeling
78
models is tested following the usual procedure adopted for calibration and
verification. In the present study total climatic data has been divided in to two sub-
periods, AD 1966 to 1983 and 1984 to 2000. Initially, the calibration was made on the
first half of the data- set and the independent verification was done on the other half
and vice-versa. Finally, for the final reconstruction of Maximum March_April
temperature, regression coefficients were calculated for the whole period of 1966–
2000 and various calibration and verification statistics were calculated (Table 2). The
advantage of Bootstrapping seems to perform better than cross-validation in many
cases (Efron, 1983). In the simplest form of bootstrapping, instead of repeatedly
analyzing subsets of the data, subsamples of the data repeatedly analyze. Each
subsample is a random sample with replacement from the full sample. The reliability
of the Bootstrap orthogonal regression model was evaluated by statistics on
calibration and verification periods. Evaluative statistics provided for the calibration
period are the Pearson correlation coefficient (r), the coefficient of determination (R2),
R2 adjusted where as for the verification period, it is t-test, reduction of error (RE),
and sign test (ST). Values of RE was positive, indicating a significant skill present in
the estimates of late summer temperature. All calculations were made by using DPL
programme.
The reconstructed data of the Maximum temperature for March -April extends
back to 1759 to 2000 but the reliable reconstruction is only from 1821 to 2000 (Fig.
5.9.1). Because before A.D. 1821 the EPS value of the tree-ring chronology of YAB
site was below 0.85, so the statistically reliable reconstruction may be considered for
1821 to 2000 AD. Descriptive statistics reveals that in this series values are for mean
(19.822oC); standard deviation (0.412) maximum (20.972oC) and minimum
(18.821oC). The reconstructed MA temperature series indicated annual to decadal
scale variations (Fig. 5.3). A 5-year of moving average was calculated to identify
variations in pre- summer temperature. The value greater than that of maximum and
minimum represents hot and cold months respectively observed in the reconstruction.
The reconstructed temperature series for the last 180 years showed annual to
multiyear fluctuations punctuated with cool and warm periods.
Chapter 5 Dendroclimatic modeling
17.000
18.000
19.000
20.000
21.000
22.000
23.000
1750
1760
1770
1780
1790
1800
1810
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
Tem
pera
ture
o C
18.000
18.500
19.000
19.500
20.000
20.500
21.000
21.500
1820
1824
1828
1832
1836
1840
1844
1848
1852
1856
1860
1864
1868
1872
1876
1880
1884
1888
1892
1896
1900
1904
1908
1912
1916
1920
1924
1928
1932
1936
1940
1944
1948
1952
1956
1960
1964
1968
1972
1976
1980
1984
1988
1992
1996
2000
Year
Max
imum
Tem
pera
ture
o C
(Rec
onst
ruct
ed)
12
14
16
18
20
22
24
Max
imum
Tem
pera
ture
o C
(Act
ual)
Fig.5.15 Tree-ring based
reconstructed Maximum
March-April temperature in Zemu Valley (dotted green
line is start year of the reliable
time span); Actual (green line)
and estimated (red line).
Chapter 5 Dendroclimatic modeling
80
Site _Name Model Name Calibration Verification YABUK ABDE_YAB/ Max temperature_(March-April) 0.273± 0.1267 a 0.204± 0.105 b
Calibration Verification Calibration Verification
(1966-1983) (1984-2000) (1984-2000) (1966-1983)
r 0.723* 0.477* 0.477* 0.7233* R2 0.523* 0.231 0.231 0.5232* RE
0.4453*
0.7119*
t-value
4.0409*
5.0492* Sign test
16/1
15/3
-3
-2
-1
0
1
2
3
4
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
Year
Max
imum
Tem
pera
ture
o C
( Mar
ch-A
pril)
R2 is the square of the correlation coefficient calculated between actual and estimated data. t-value is derived using the product mean test. The departures from the actual and reconstructed series are multiplied in each year; the means of negative and positive products are calculated, and the difference between the means is tested with student’s t-test. Sign test is number of agreement/total values with the correct deviation (first number) from the mean. Reduction of error statistics in any positive value demonstrates skill in reconstruction (varies between -1.0 and +1.0).
a The value is averaged multiple correlation coefficients from 200 replications and their standard deviations
b Indicates the independent verification are significant at 95% level
Table.5.3.Monthly climatic models for Zemu Valley based on ABDE_YAB using bootstrap method.
Table 5.4 Statistics of calibration and verification for tree-ring reconstruction of maximum March-April temperature
Fig.5.16. Showing Anomaly in the reconstructed March_April temperature.
Chapter 5 Dendroclimatic modeling
81
.
Fig5.18. Scatter plot of actual and tree-ring reconstructed Maximum (March-April) temperature with a linear relationship highlighted during the period of 1966–2000.
Fig.5.17 The comparison of actual and reconstructed Maximum March-April temperature from 1966 to 2000
16.00
18.00
20.00
22.00
24.00
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
Year
Tem
pera
ture
(0 C)
ReconstuctionActual
R2 = 0.3757
19.000
19.500
20.000
20.500
21.000
17.00 18.00 19.00 20.00 21.00 22.00 23.00
Actual temperature (0C)
Reco
nstu
ctue
d te
mpe
ratu
re (
0 C)
Chapter 5 Dendroclimatic modeling
5.4.1.1 Salient features of reconstructed March April maximum temperature
The reconstruction indicates that the observed value is higher than mean value i.e.
warmer years includes:
low temperatures than mean value i.e. represents cold years are:
5.4.2. Linear regression method for Average temperature of (July August)
temperature reconstruction
Correlation analysis of the climate-growth relationship showed high
significant negative correlation between the radial growth of ABDE from YABUK
site and mean temperature of July-August (Fig. 5.10) and positive correlation with
average (March-April) temperature. The highest correlation between tree rings and
July to August) temperature (r = - 0.693; p < 0.01). Therefore, average July-August
temperature used as the climate variable for the reconstruction. A simple linear
regression model was obtained to reconstruct (July-August) temperature history of the
study area (Cook and Kairiukstis, 1990). Linear regression model used between tree-
ring indices and temperature for the 1966–2000 calibration periods.
The model was designed as:
Average temperature for (July August) temperature = 21.086-1.8424*YAB CRN
In this equation, average temperature for (July August) (in oC) represents the
reconstructed average temperature and YAB_CRN represents the tree-ring width
index of ABDE_YAB. The ability of this equation has been tested to build climate
reconstruction by applying it to the validation dataset. The correlation coefficient is
−0.694 (n=35, P<0.01), and variance explanation (R2) is 23.6% (R2 adj=19.0%) in the
calibration period. The function F test value is 8.439. Split calibration verification
results (Table 1) show that it has a high correlation coefficient both in the calibration
and verification periods. It also passed the Sign test (P<0.05) and Reduction of error
(RE), statistical tests for paleoclimate reconstruction. The model was applied to the
pre-instrument ring widths to reconstruct past climate.
5.5. Reliability of the regression model
5.5.1. Calibration verification of model for temperature reconstruction
Past climate was reconstructed from tree rings by a linear regression model. The
statistics from splitting samples into calibration-verification intervals showed a
Chapter 5 Dendroclimatic modeling
84
significant correlation between the reconstructed and the actual temperatures during
the calibration and verification periods (Table 1).The reliability of the regression
model was evaluated by statistics on calibration and verification periods. Evaluative
statistics provided for the calibration period are the Pearson correlation coefficient
(R), the coefficient of determination (R2), and F test (F) and for the verification period
are reduction of error (RE), Values of RE was positive, indicating a significant skill in
the estimates of late summer temperature, sign test (ST), and Durbin-Watson test
(DW) (Fritts 1976; Cook et al. 1994), DW value was about 1.5, indicating
insignificant autocorrelation in the model residuals and target climate data. These
results demonstrated the reliability of our regression model. R2, RE, are all measures
of shared variance between climate and tree rings, and a positive RE is evidence for a
valid regression model. The results of the sign test and product mean test also
exceeded the 95% confidence level
5.5.2 Variability in reconstructed climate data.
The reconstructed data of the Late summer temperature (July-August) history
since extends back to 1759 to 2000, but the reliable reconstruction is only from 1821
to 2000 (Fig. 5.19). Because before A.D. 1821 the EPS value of the tree-ring
chronology of YAB site was below 0.85, so the statistically reliable reconstruction
may be considered for AD 1821 to 2000. Descriptive statistics showed that, Mean
(19.299 OC); standard deviation (0.305) maximum (20.07 OC) and minimum (18.327
OC). The reconstructed summer temperature series indicated annual to decadal scale
variations (Fig. 5.15). A 5-year of moving average was calculated to identify
variations in late summer temperature. The Value greater than that of Maximum and
minimum represents Hot and cold month in the reconstruction. The reconstructed
temperature series for the last 180 years showed annual to multiyear fluctuations
punctuated with cool and warm periods.
The reconstructed temperature showed that warm episodes occurred in the
intervals A.D. 1821–2000, with regular interval of cooler summer (Fig. 5.b).
Although continues low temperature then mean value i.e. a cold interval in 1822 to
Chapter 5 Dendroclimatic modeling
85
1828, 1836 to 1841, 1857 to 1860, 1875 to 1877, 1879 to 1881, 1883 to 1886, 1888 to
1891, 1895 to 1904, 1910 to 1914, 1916 to 1920 1939 to 1942, 1944 1945, 1947 to
1958 1960 to 1967, 1970, 1973 to 1979, 1984, 1985, 1987 1988, 1993, 1994, 2000.
Notable warm anomaly of the reconstruction are from 1831 to 1835, 1842 to 1847,
1850 to 1856, 1861 to 1874, 1905 to 1909 etc. In the last warm period from A.D.
1991 to present, the temperatures of all the years were above the long-term mean.
Chapter 5 Dendroclimatic modeling
17.00
17.50
18.00
18.50
19.00
19.50
20.00
20.50
21.00
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
Year
Tem
pera
ture
o C
Reconstuctucted(Jul-Aug) Actual temperature
17.0
17.5
18.0
18.5
19.0
19.5
20.0
20.5
21.0
1750
1760
1770
1780
1790
1800
1810
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
Year
Tem
pera
ture
oC
Fig. 5.19. Tree-ring based reconstructed Mean July-August temperatures in Zemu Valley; (dotted green line is start year of the reliable time span), actual (green line) and estimated (red line).
Chapter 5 Dendroclimatic modeling
87
Fig. 5.20. The comparison of actual and reconstructed Mean temperature of July-August from 1976 to 1996.
Fig. 5.21. Scatter plot of actual and tree-ring reconstructed Mean July-August
temperature with a linear relationship highlighted during the period of 1966–2000.
5.5.3. Characteristics of reconstructed temperature
In order to identify whether this climatic reconstruction (July- August)
represents features that are coherent over a large spatial scale, I compared the present
reconstruction data with other tree-ring based temperature reconstructions. The
present reconstruction shows a good coherence with the climatic reconstruction made
from the nearby other site of the Eastern Himalaya. In that region cool late-summer
temperature during 1782-1786, 1830-1831,1899,1993 and 1975 and warmer summers
during 1777-1779, 1817, 1843, 1904-1906, 1926-1927, and 1980-1982 reported by
Bhattachryya et al. (2003) have also matched with the present reconstruction. This
similarity of data also supports the authenticity of the present reconstruction. In order
to identify whether our reconstruction represents features that are coherent over a
large spatial scale, I also compared present data with Tibet and other higher elevation
sites of China. Fan et al. (2009) reconstructed summer temperatures in the central
Hengduan Mountains China, and reported warm episodes during A.D. 1780s, 1810s–
1820s, 1840s–1850s, 1920s–1940s,1990s-present, were are in agreement with warm
periods occurred at 1780s, 1820s–1850s, 1930s and 1990s in the Zemu region. The
most salient finding in the reconstruction of temperature from Zemu glacier valley is
the cooling during the major part of the nineteenth century in comparison to warming
in the 20th century, particularly 1830 to 1870 and 1900 to 2000. These finding are also
well corroborative with study of China. Based on instrumental data and other proxies
(ice cores, tree rings), Wang et al. (2004) showed that temperature anomalies during
the period 1920–1950 are noticeable positive over China throughout the last century,
especially in southwestern China and on the Tibetan Plateau. Warm conditions around
1950, and the cool period around 1970 were also reported in West Sichuan (Shao and
Fan, 1999) and Tibet (Briffa et al., 2001; Brauning and Mantwill, 2004). The
pronounced negative summer temperature trends from 1970 to 1990 on the Tibetan
Chapter 5 Dendroclimatic modeling
91
Plateau are probably the consequence of enhanced cloudiness and rainfall at the upper
treeline and thus of increasing monsoon intensity (Brauning and Mantwill, 2004).
5.5.6. Cyclic nature of tree based reconstructed climate records
Wavelet analyses provide insights into recurrent variability within a time
series that has changed in strength and frequency (Gedalof and Smith, 2001b; Rigozo
et al., 2001). These analyses has been undertaken at the interactive website
(www.atoc.colorado.edu/research/wavelets/) developed by Torrence and Compo (1998).A
visual examination of reconstructed series of Maximum March-April temperature and
mean July-August temperature suggests that it has cyclic trends (Fig. 5.23) and Fig
5.24) respectively. For wavelet analysis has confirmed a dominant mode of variability
of less than 16 years and 8 years (Fig. 5.23 and 5.24).
Chapter 5 Dendroclimatic modeling
92
Fig. 5.23 (a) Maximum March-April temperature. (b) The wavelet power spectrum.
The power has been scaled by the global wavelet spectrum (at right). The cross-
hatched region is the cone of influence, where zero padding has reduced the variance.
Black contour is the 95% significance level, using a red-noise (autoregressive lag1)
background spectrum. (c) The global wavelet power spectrum (black line).The dashed
line is the significance for the global wavelet spectrum, assuming the same
significance level and background spectrum as in.
Fig. 5.24 (a) July-August mean temperature (b) The wavelet power spectrum. The
power has been scaled by the global wavelet spectrum (at right). The cross-hatched
region is the cone of influence, where zero padding has reduced the variance. Black
contour is the 95% significance level, using a red-noise (autoregressive lag1)
background spectrum. (c) The global wavelet power spectrum (black line). The
dashed line is the significance for the global wavelet spectrum, assuming the same
significance level and background spectrum as in.
Chapter 6 Tree growth and glacier relationship
93
Among several factors controlling glacier dynamics and tree growth, climate
have pivotal role in both cases. Thus there might be relationship between
advancement/retreat with variation of tree growth low/high at least in trees growing
close to the glacier. Based on this assumption in this chapter of dissertation, an
attempt has been made to analyze tree growth/ glacier fluctuations relationship, with
special reference to the Zemu Glacier area of the Eastern Himalaya. Glaciers develop
where mass gain (by snowfall and avalanches) exceeds mass loss (by melting and
calving). Generally Lower temperatures and greater snowfall favor mass gain
(accumulation) and conversely, higher temperatures favor mass loss (ablation). The
sum of accumulation and ablation over any time period is the mass budget. If ablation
dominates over several years, the mass flux is reduced and the glacier starts to retreat
this event could be reflected as increased radial growth rates and the production of
wider annual growth rings. Conversely, if net annual accumulation (positive balance)
dominates for a long time, the glacier flow speed increases and eventually advances.
As mentioned earlier in chapter 1 positive glacier mass balance are detrimental to tree
growth and resulted in narrow annual tree rings.
6.1. Glacier behavior and tree-ring width chronology
The correlation analysis or simple matching of the data was made between the
ring-width chronology of fir (Abies densa ) from Yabuk site close to snout of the
Zemu glacier (Fig. 6.1) and available glacier fluctuation data (Fig.6.2). A reverse
Relationship between Zemu glacier fluctuation and tree-ring width of fir is observed.
In initial phase of correlation diagram (Fig.6.2) reveals average tree-growth is high
during 1976-78(fastest retreat) in second phase when glacier shows advancement in
1988-2000 tree-growth get suppressed and in last phase (2001-2005) again tree
growth increased when again retreat starts. As mentioned earlier in the chapter 2 that
during the last 100 years, the Zemu glacier advanced 1988- 2000 is in bracket years of
the longest period of low tree-growth as indicated in our data also during this period.
The other prominent high growth periods recorded in tree ring data were around
1776-1804 and 2007-2010. The high growth indices during this time span possibly
reflect the retreat of glacier. Since glacier behaviour data are very limited in Zemu
area (Eastern Himalaya), and mass balance records from Himalayan glaciers are
Chapter 6 Tree growth and glacier relationship
94
extremely rare and of short duration (Zemp et al. 2009) thus to understand that any
relationship existed in longer time scale and considering synchrony in regional glacier
behavior, I simply correlated tree-ring data with regional glacier fluctuation history of
Eastern Himalaya, Central Himalaya (Nepal) and China, which are taken form
published record (Dyurgerov, 2005). It shows that during the twentieth century, most
of the glaciers in the China E.Tien Shan CN0010 (Fig.6.4) have undergone period of
general retreat. Although three pronounced intervals of negative Mass balance (bn)
1988, 1991, and 1994 observed, whereas a positive mass balance has been observed in
1989, 1990, 1992, as a result, glaciers continue to maintain terminal positions. In
matching with the ring-width of fir for these periods shows a reverse relationship, i.e.
higher tree growth observed in 1988, 1991 but after 1992 a continuous increasing
trend of tree growth observed till 1994 and then decreases when glacier exhibit
increasing trend in negative mass balance. In respect to Central Himalayan (Nepal)
glacier, (Lat. 27°42’N, Long. 86°34'E) where data is available for four years from
1996 to 1999 showing continuous negative mass balance (Dyurgerov ,2002), tree-
growth shows increase trend till 1998 and since then a sharp downfall have been
observed both in bn as well tree growth. Changmekhangpu glacier India (Fig.6.3).
having little older records of glacier retreat to that of Tibetan glaciers also shows
similar pattern, where pronounced intervals of negative mass balance are recorded
from 1981 to 1986. Tree-growth shows increasing trend with respect to highest
negative mass balance 1981, 1983, and 1985. These preliminary results show
potential to established relationship between climate, glacier and tree-growth. This
ongoing rapid global warming has a much effect on the Himalayan environment and
this is clearly visible in the rapid retreat of Himalayan glaciers (Dyurgerov and Meier
2005). Data from the higher elevation Himalayan glaciers indicate consistently
negative mass balance values, but the extent to which they can be considered
regionally as representative is not known (Armstrong, R. L. 2010). The 2005 IPCC
statement about the possible disappearance of Himalayan glaciers by 2035. No
evidence was presented that Himalayan glaciers are receding faster than those in other
parts of the world, as in this model only rates of retreat for the Himalayan glacier
were presented. Although simple correlation analysis revels negative relationship with
Chapter 6 Tree growth and glacier relationship
95
retreat of Zemu glacier and mass balance of chinese and Nepal glaciers i.e.
Significant value were bn1, -0.285, bn2, -0.258, bn3, -0.384 but not at high significant
level, because of short time span of glacier record data. But definitely it gives a clear
indication of trends of negative relationship with the glacier mass balance data that
may further be explained if long term data for common time period may be available.
Fig. 6.1. Photograph showing the Snout position of Zemu Glacier, North Sikkim
(modified after Luitel et al., 2012).
Chapter 6 Tree growth and glacier relationship
96
-50-40-30-20-10
01020
1976-1978 1988- 2000 2001-2005
Year
Ret
reat
rat
e(m
)
0.850
0.900
0.950
1.000
1.050
Ave
rage
Rin
g-w
idth
inde
x
Average Retreat rate Average Ring-width
Fig.6.2.Tree growth and its relation with Zemu glacier
-450-400-350-300-250-200-150-100-50
0
1981 1982 1983 1984 1985 1986
Net
mas
s B
lanc
e
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
Tree
-rin
g w
idth
Indi
ces
Net mass Blance ABDE_YAB_
Fig.6.3. Tree growth and its relation with Changmekhangpu glacier
Fig.6.4. Comparison of Fir Chronology of ABDE_YAB with available Mass balance data of three Chinese glacier (AD 1988-1995) and Nepal glacier (AD 1996-1999).
Himalaya Nepal AX010 NP00005 E.Tien Shan China Gl. #1, E.Br. CN0010 E.Tien Shan China Gl. #1, W.Br. CN0010
E.Tien Shan China Urumqihe S.#1 CN0010 Tree-ring Chronology
Chapter 6 Tree growth and glacier relationship
97
6.2. Reconstructed temperature and glacier fluctuation
Warm and cold periods in our reconstruction were also generally in phase with
the periods of retreat and advance of the glaciers of this region. The notable
characteristic of the reconstruction was the increase temperature in two time span ie.
from 1830 to 1870 and 1940 to present. It coincided with a growing mean annual
temperature of the Eastern Himalaya region (Shrestha et al., 2010). The higher retreat
of glacier observed during 1976-1978, (-41.25 m), and in recent years 2001-2005, (-
3.17 m) (Raina et al., 2009) are under the bracket year of increased temperature in our
study. In accordance with the decreasing trend of summer temperature, an
advancement of Zemu glacier has been reported in 1988-2000 (7.67m). The cold
period around 1870 to 1900 correspond to Zemu glacier advancement. Hence, tree
rings seem to be a good indicator of past climate and glacier fluctuations in the Zemu
area.
6.2.1. Role of Maximum March-April temperature in fluctuation of Zemu glacier
Geomorphological records reveal that the retreat was very fast for the period
of 1976-1978 at the Zemu glacier (Raina et. al., 2009) which also coincide with the
period of increased maximum March–April temperature, both in reconstructed as well
as in the observed data. During this time span temperatures were higher than the
average value. In contrast, for the period of advancement of glacier i.e., 1988 to 2000;
both the reconstructed as well observed Maximum temperature of March- April was
below to mean i.e. (19.5130 C).
6.2.2. Role of Average July-August temperature in Fluctuation of Zemu glacier
It is of general observation that the increased temperature enhances the rate of
retreat of a glacier and it is opposite for the low temperature. But here, I have
recorded contradictory response among July-August temperature/ tree-growth/glacier
behaviour. As per published record (Raina et. al., 2009) and was mentioned earlier
also in section 2.5.2. that the retreat of Zemu glacier was very fast during 1976-1978
when temperature of July-August at Zemu Valley are below the mean value for the
years 1976, 1977, 1978 (18.7370 C, 18.5730 C, and 19.2970 C respectively). In
Chapter 6 Tree growth and glacier relationship
98
contrast, for the period of 1988 to 2000, the recorded period of Zemu glacier
advancement (Raina et. al., 2009) temperature of July-August are found high in most
of these years. This relationship i.e., increased temperature here related to glacier
advancement may be due to combined effect of temperature and precipitation during
these months. As per yearly meteorological record of this region, both temperature
and precipitation remain high during July–August. Thus precipitation in the higher
reaches during these months in the form of snow enhances accumulation and increase
snow depth. Moreover increased temperature during these months causes higher
evaporation which also makes the surface temperature low thus weakening the effect
of temperature. As mentioned earlier in the Introduction of this dissertation that the
rapid retreat of this glacier is linked with the global warming but this inverse
relationship recorded with temperature suggest that the July-August temperature does
not play much role in glacier retreat. Thus it appears that the temperature of March-
April, other significant climatic variables limiting the tree growth of this region may
have significant role in glacier advancement/ retreat in comparison to the variation of
summer temperature of this region. In support of this view it has been observed that
increased July-August temperature of prior years are found correlated with the low
March April temperature of following year i.e., current year.
Chapter 7 Dendrohydrological modeling
99
7. 1. Study area of river discharge The river Teesta is one of the main Himalayan Rivers, which originates from the
glaciers of Sikkim at an elevation of over 8,500 m a.s.l. It is being snow fed by the glaciers
Zemu, Changame Khanpu, Talung etc. flows through the states of Sikkim and West Bengal
in Indian Territory and then to Bangladesh (NHPC, 2009). It is a major source of irrigation
and hydropower generation (NHPC, 2009). But an obvious question comes here, is about the
need of enough discharge to support both irrigation and hydropower generation? this is
important because little is known about the long-term properties of the Zemu river at Lachen
hydrologic regime and its principal contributors, snow and glacier melt water. Zemu chu a
first stage of the river originating from Zemu glaciers and generally flows in north-south
direction, at a steeper gradient.
Fig.7.1. Map showing location of tree ring site, meteorological station and discharge gauge station at Lachen, north Sikkim, Eastern Himalaya. For the generation of the map SRTM 30 (digital terrain elevation data set was used)
Chapter 7 Dendrohydrological modeling
100
Fig.7.2. Photograph showing “Zemu Chuu” at Lachen North Sikkim.
7.2 Relationships between Tree growth and Climate The correlations between the chronology and climate data have calculated from the
previous November to current year October. Significant correlations were found only of
corresponding period when tree is predictable in the correlation analysis previous year
November to current year October year (P<0.01; Figure 7.5) details has been discussed in
chapter 5. Meanwhile, the correlations between the PDSI and chronology were significant
(P<0.05) only for the month of January, reaching the 95% confidence level (details in chapter
Chapter 7 Dendrohydrological modeling
101
8). High elevations snow and ice would not increase soil moisture until the next warm period
arrives.
12.00
17.00
22.00
27.00
32.00
37.00
42.00
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Year
Stre
amflo
w(m
3/s)
Fig.7.3. Mean annual variation of “Zemu Chuu” river discharge at Lachen gauging station
(1976-1996) North Sikkim.
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Stre
amflo
w(m
3 /s)
Fig.7.4. Mean monthly variation of “Zemu Chuu” river discharge at Lachen gauging station
(1976-1996) North Sikkim.
Chapter 7 Dendrohydrological modeling
102
0.000
0.100
0.200
0.300
0.400
0.500
0.600
0.700
0.800
pNOV pDEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT
Corr
elat
ion
coef
ficie
nts
Discharge 95%CL 99%CL
Fig.7.5. Correlation plot of standard chronologies of ABDE_ZEM with averaged monthly
Discharge data of Lachen data (1977–1996). Horizontal pink line indicates significance level
(p < 0.05) and red line indicate significance level (p < 0.01).
7. 3. The stream flow reconstruction method
Correlation analysis of the discharge-growth relationship showed high significant
positive correlation between the radial growth of ABDE from ZEM site and mean discharge
of January May (Fig.7.5). The highest correlation is between tree rings and mean January-
April disharge (r = 0.708; p < 0.01). Therefore, average January –April discharge is used as
the variable for reconstruction. A simple linear regression model was obtained to reconstruct
Jaunary-April discharge history of the study area (Cook and Kairiukstis, 1990). For
reconstruction of discharge at Lachen, a Linerar Regression Model approach is used. A
transform function model between the tree-ring width index and the January- April discharge
1,1912,1914,1915,1916,1928,1929,1930,1931,1944,1945,1946,1950, 1958 1975, 1990 were
observed in the reconstruction.
Chapter 7 Dendrohydrological modeling
104
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.017
70
1780
1790
1800
1810
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
Year
Reco
nstru
cted
Stre
amflo
w (m
3 /s) Reconstructed Actual
Fig7.6. Reconstruction of January–April discharges of “Zemu Chuu” at Lachen, North Sikkim since AD 1775. The red line represents
reconstructed while green line represents actual data.
Chapter 7 Dendrohydrological modeling
105
R2 = 0.5017
7
8
9
10
11
12
13
14
3 4 5 6 7 8 9 10 11 12 13 14
Actual Streamflow (m3/s)
Reco
nstru
cted
Stre
amflo
w (m
3 /s)
Fig.7.7. (a) The comparison of actual and reconstructed stream flow (January-April) from 1976 to 1996. (b) Scatter plot of actual and tree-ring reconstructed stream flow (January-April) with a linear relationship highlighted during the period of 1976–1996.
R correlation coefficient, R2 explained variance, F F-test, Sign-test sign of paired observed and estimated departures from their mean on the basis of the number of agreements/disagreements, Pmt product mean test, RE reduction of error, RMSE is a frequently used measure of the differences between values predicted by a model or an estimator and the values actually observed, DW Durbin–Watson test * p < 0.05.
3.000
5.000
7.000
9.000
11.000
13.000
15.000
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Year
Stre
amflo
w (m
3 /s)
ReconstructedActual
Table 7.1.Statistics of calibration and verification for tree-ring reconstruction of January-April Stream flow
7.5. Cyclic nature in reconstructed January-April discharge.
Wavelet analyses provide insights into recurrent variability within a time series that
has changed in strength and frequency (Gedalof and Smith, 2001b; Rigozo et al., 2001). ).A
visual examination of reconstructed series of mean January-April discharge suggests that it
has cyclic trends (Fig. 7.8). Wavelet analysis has confirmed a dominant mode of variability
of less than 16 years and 8 years (Fig. 7.8).
Chapter 7 Dendrohydrological modeling
107
Fig.7.8. (a) January-April _Discharge. (b) The wavelet power spectrum. The power has been
scaled by the global wavelet spectrum (at right). The cross-hatched region is the cone of
influence, where zero padding has reduced the variance. Black contour is the 95%
significance level, using a red-noise (autoregressive lag1) background spectrum. (c) The
global wavelet power spectrum (black line). The dashed line is the significance for the global
wavelet spectrum, assuming the same significance level and background spectrum as in.
Chapter 8 Tree growth and its relation with PDSI and El Nino
105
8.1. Tree-growth/glacier fluctuation and El Niño relation To precisely attribute glacier retreat to a particular climate forcing requires
detailed knowledge and understanding of the climatic changes that have taken place in
the 20th century; unfortunately the details of Zemu glacier fluctuation history is
available only for limited time scale. As far as El Niño 3.4 concern with in general,
negative mass balances are correlated with El Niño conditions, while positive mass
balances predominate during La Niña conditions (Vincent et al., 2009). Earlier studies
also supported this, that during negative Southern Oscillation Index values (El Niño,
warm tropical Pacific SSTA in the eastern equatorial Pacific) accumulation is reduced
and ablation is enhanced at a monthly timescale and over longer periods (Francou et
al., 1995a, b, 2000; Ribstein et al., 1995; Wagnon et al., 2001; Francou et al., 2003,
2004; Favier et al., 2004a, b). In this dissertation, the relationship of retreat (retreat is
result of prolonged negative mass balance) and El Niño for Zemu glacier has been
analysed. In this Chapter correlation analysis has been performed between averaged
tree ring data, and average data of El Niño 3.4, and all India rainfall for the period of
Zemu glacier history. Correlation analysis between ABDE_YAB chronology, glacier
fluctuation data and average El Niño 3.4 data shows a negative correlation between
retreat and El Niño (r = -0.942, 90% significance level) and between tree ring width
and El Niño 3.4 also negative correlation has been observed. Relation with average
data of respective period of Zemu glacier history and all India rainfall is positive i.e.
(r = 0.953, 95% Significance level). The more frequent El Niño years between AD
1909-1965, 1965-1975 and 1975-1986 and between AD 1986-2005 may be
responsible for Zemu glacial fluctuation (Fig.8.1, and Table.8.1). However, this
conclusion is not supported by solid evidence because the data set is very limited.
Although correlation between regional chronology of principle component scores
PC#1, PC#2, were negatively correlated but not at significance level. For the use of
glacier fluctuation as a climate proxy it essential to identify the significant climatic
factor controlling their behaviour, i.e. establishing a transfer function between current
climate and glacier variations. However, the accelerated glacier retreat at the end of
the 19th century is in agreement with the results of Torrence and Webster (1999), who
Chapter 8 Tree growth and its relation with PDSI and El Nino
106
used SST in the ELNiño-3 zone (5°S–5°N, 90°–150°W) and monthly precipitation
measurements in India (both available since 1871), and concluded that the period
from 1875 to 1920 was characterized by a high frequency of El Niño events.
-0.400
-0.350
-0.300
-0.250
-0.200
-0.150
-0.100
-0.050
0.000
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
corr
elat
ion
coef
ficen
ts
PC#1 PC#2 95CL
Fig. 8.1 Correlation values of mean monthly El Niño 3.4 with standard regional
chronologies PC#1 and PC#2. Monthly variables spanning from January to
December. The pink horizontal line indicates 95% confidence limits.
0
5
10
15
20
25
30
35
1909-1965 1965-1975 1975-1986 1986-2005
Gla
cier
retre
at (m
)
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
Mea
n da
ta
annual retreat(m) YAB_CRN EL Nino 3.4
Fig. 8.2. Plot of Zemu glacier retreat with mean data of ring-with, and El Nino 3.4 Table 8.1 Correlation value of glacier retreat, mean data of ring-with, and El Nino 3.4
Retreat(m) ABDE_YAB ABDE_YAB 0.402 1.000
El Niño 3.4 -0.942* -0.684*
All India rainfall 0.530 0.953*
* Indicate 95% confidence limit,
Chapter 8 Tree growth and its relation with PDSI and El Nino
107
8.2. Tree growth and its relation with PDSI
Tree growth is also controlled by soil moisture condition, while soil moisture
balance is a result of the integrative effect of precipitation, evapotranspiration,
physical-chemical properties of soil substrate, etc. In order to better understanding of
drought condition it is necessary to take not only precipitation but also other
hydrological meteorological factors. Based on water balance model, Palmer Drought
severity index was developed, considering monthly air temperature, precipitation, and
local available soil water content. In this context to find out the relationship of tree
growth with the Palmer Drought Severity Index (PDSI) of the site (88.75'E 26.25'N).
For that, average tree-ring chronologies of all sites have also been compared with 2.50
x 2.50 gridded Palmer Drought Severity Index (PDSI) datasets (Dai et al., 2004). For
this analysis monthly dataset from AD 1967 to 2000 has been used. The analysed,
relationship of regional chronology i.e PC#1, PC#2 with PDSI has been analysed. The
monthly variables were taken for the twelve-month period i.e., from November of the
previous year to October of the current year’s. Correlation analysis of all ABDE
,JURE, JUSQ chronologies with PDSI, a positive correlation were recorded for almost
all site except ABDE_DOZ and JUSQ_YAB sites where negative correlation was
found, only month of April of JUSQ_YAB (p<0.05). In case of ABDE_ZEM,
previous year December and January were found Positive and significant (p<0.05). In
case of regional chronologies i.e PC#1, PC#2, with the monthly PDSI values of grid
points, showed a strong positive relationship with the respective site chronologies and
PC#1 PC#2. A positive relation correlation was recorded for all months with
significance of January month ((p<0.05) where as with PC#2 a negative correlation
recorded with February –April (Fig.8.3).
Chapter 8 Tree growth and its relation with PDSI and El Nino
108
(a)
-0.300
-0.200
-0.100
0.000
0.100
0.200
0.300
pNOV pDEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT
Corr
elat
ion
coef
ficie
nts
PC#1 PC#2 95CL 95CL 99 CL 99 CL
(b) Fig.8.3. Correlation values of mean monthly PDSI of the one grid points with standard regional chronologies of all site and PC#1 and PC#2. (a) All chronologies. (b) PC#1, PC#2. Monthly variables spanning from November of the previous year to October of the current year. The pink horizontal line indicates 95% and red line indicate 99% confidence limits.
1958, 1975, 1990 were observed based on the reconstruction. The relation with tree-
ring index and the monthly PDSI data were developed for this region the tree-ring
chronologies of fir and monthly or seasonal PDSI were positively correlated for all
month starting from previous November to current year October, but was statistically
significant for only the month of January.
Correlation analysis between ABDE_YAB chronology and with the glacier
fluctuation (advance), positive mass balance and average El Niño 3.4 data show a
negative correlation. Correlation between retreat and El Niño is -0.942 (90%
significance level) where as between tree ring width and El Niño 3.4 also negative
correlation. It has been observed that period of Zemu glacier history (advance) and all
India rainfall is positive i.e. r = 0.953 (95% Significance level). The more frequent El
Niño years between AD 1909-1965, 1965-1975 and 1975-1986 and between AD 1986
Chapter 9 Discussion & conclusion
115
to 2005 may be responsible for Zemu glacial fluctuation (retreat). However, this
conclusion is based on limited data.
9.1 Further work for the improvement of research work A comparison of the tree-ring and glacier fluctuations (advancement /retreat or
mass balance positive/ negative) records indicates that the trees in the area are much
sensitive and respond to glacier fluctuations. In the absence of long term instrumental
records, they permit exploration of the relative contribution of changes in temperature
to net mass balance. An overall assessment of analyses of climate, stream flow and El
Niño signatures in Zemu glacier area presented in the present research work appears
to have great promise and should look forward to a more comprehensive analysis
based on development of tree-ring data in both spatial and temporal coverage close to
the snout of the glaciers of the Eastern Himalaya. Because of the complex landscape
in this mountainous region, more tree-ring data are required for better understanding
of the regional climate variability. Though careful interpretation and selection is
required, tree-ring based proxy records of climate would allow the reconstruction of
continuous records of glacier. The substantial glacier recession of this century has
continued to expand the opportunities for advancing dendroglaciological studies by
exposing the remains of tree/shrubs buried by glacier advances at various times
during Little Ice Age or other period of advances during the Holocene. These findings
would help for the development of millennia-long tree-ring chronologies in glaciated
sites. The development of long tree-ring chronologies also allows for comparison of
accurately dated deposits over large areas, essential for assessing the synchroneity of
global glacier activity to climate forcing mechanisms at different timescales. Due to
increased human pressure in the Himalayas it is hard to find wood pieces in the
exposed sediments as are these used as fuel or for construction purposes. Though it is
difficult but with the extensive field survey there may be chances to get buried wood
or old trees. A detailed analysis through multidisciplinary approach would bring out
valuable information regarding glacier advancement and retreat in greater details from
the Himalyan region.
Chapter 10 Bibliography
116
Attri, S. D., Tyagi, A., 2010. Climate Profile of India. Contribution to the Indian Network of Climate Change Assessment (National Communication-II), Ministry of Environment and Forests 1501, 1-129.
Barclay, D.J., Wiles, G.C., Calkin, P.E., 2009. Tree-ring cross dates for a First Millennium AD advance of Tebenkof Glacier, southern Alaska. Quaternary Research, 71, 22–26.
Bhattacharyya, A., LaMarche Jr, V. C., Telewski, F. W., 1988. Dendrochronological reconnaissance of the conifers of northwest India. Tree-Ring Bulletin 48, 21-30.
Bhattacharyya, A., Yadav, R. R., Borgaonkar, H. P., Pant, G. B., 1992.Growth-ring analysis of Indian tropical trees; dendroclimatic potential. Current Science 62, 736–741.
Bhattacharyya, A., Yadav, R.R., 1996. Dendrochronological reconnaissance of Pinus wallichiana to study glacial behaviour in the western Himalaya. Current Science70, 739–744.
Bhattacharyya, A., Yadav, R. R., 1999. Climatic reconstructions using tree-ring data from tropical and temperate regions of India—a review. IAWA Journal 20, 311-316.
Bhattacharyya A., Chaudhary., V., 2003. Late-summer temperature reconstruction of the Eastern Himalayan region based on tree-ring data of Abies densa Arctic. Antarctic Alpine research 35, 196 – 202.
Bhattacharyya, A., Shah, S. K., Chaudhary, V., 2006. Would tree-ring data of Betula utilis have potential for the analysis of Himalayan glacial fluctuations? Current Science 91, 754–761.
Bhattacharyya, A., Eckstein, D., Shah, S.K., Chaudhary, V., 2007.Analyses of climatic changes around Parambikulam, south India, based on earlywood mean vessel area of teak. Current Science 93, 1159–1164.
Biondi, F., Lange, C. B., Hughes, M. K., Berger, W. H., 1997. Inter-decadal signals during the last millennium (AD 1117–1992) in the Varve record of Santa Barbara Basin, California. Geophysical Research Letters, 24, 193-196. Biondi, F., 2000. Are climate-tree growth relationships changing in north-central idaho, USA?. Arctic, Antarctic, and Alpine Research, 111-116. Borgaonkar, H.P., Pant, G.B, Rupa Kumar. K., 1994. Dendroclimatic reconstruction of summer precipitation at Srinagar Kashmir India since the late 18th century. The Holocene 4, 299–306.
Chapter 10 Bibliography
117
Borgaonkar, H. P., Pant, G. B., Rupa Kumar, K., 1996. Ring‐width variations in Cedrus deodara and its climatic response over the Western Himalaya. International Journal of climatology 16, 1409-1422.
Borgaonkar, H. P., Pant, G. B., Kumar, K. R., 1999. Tree-ring chronologies from Western Himalaya and their dendroclimatic potential IAWA J 20, 295–309.
Borgaonkar, H. P., Ram, S., Sikder, A. B., 2009. Tree-ring analysis of high elevation Cedrus deodara D. Don from Western Himalaya in relation to climate and glacier fluctuations. Dendrochronologia 27, 59-69.
Borgaonkar, H. P., Sikder, A. B., Ram, S., Pant, G. B., 2010. El Niño and related monsoon drought signals in 523-year-long ring width records of teak (Tectona grandis L.F.) trees from south India. Palaeogeography, Palaeoclimatology, Palaeoecology 285, 74-84.
Borgaonkar, H. P., 2011. Dendroclimatology and climate change: Indian perspective. Journal of the Indian Academy of Wood Science 8, 52-61.
Brauning, A., 2001. Combined view of various tree ring parameters from different forest habitats in Tibet for the reconstruction of seasonal aspects of Asian Monsoon variability, Palaeobotanist 50, 1–12.
Bräuning, A., Mantwill, B., 2004. Summer temperature and summer monsoon history on the Tibetan plateau during the last 400 years recorded by tree rings. Geophysical Research Letters 31, 24.
Braeuning, A., 2006. Tree-ring evidence of "Little Ice Age" glacier advances in southern Tibet. The Holocene 16, 369–380.
Briffa, K. R., Jones, P. D., Schweingruber, F. H., 1992. Tree-ring density reconstructions of summer temperature patterns across western North America since 1600. Journal of Climate 5, 735-754. Briffa, K. R., Jones, P. D., 1993. Global surface air temperature variations during the twentieth century: Part 2, implications for large-scale high-frequency palaeoclimatic studies. The Holocene 3, 77-88. Briffa, K. R., Jones, P. D., Hulme, M., 1994. Summer moisture variability across Europe, 1892–1991: an analysis based on the Palmer drought severity index. International Journal of Climatology 14, 475-506.
Chapter 10 Bibliography
118
Briffa, K. R., 1995, Interpreting high-resolution proxy climate data the example of dendrochronology, in Analysis of Climate Variability: Applications of Statistical Techniques, edited by H. von Storch and A. Navarra, 77– 94, Springer-Verlag, New York.
Briffa, K. R., Osborn, T. J., 1999, Seeing the wood from the trees, Science 284, 926–927.
Briffa, K. R., Osborn, T. J., Schweingruber, F. H., Harris, I. C., Jones, P. D., Shiyatov, S. G., Vaganov, E. A., 2001. Low-frequency temperature variations from a northern tree ring density network. Journal of Geophysical Research, 106, 2929–2941.
Briffa, K. R., Osborn, T. J., 2002, Blowing hot and cold, Science 295, 2227–2228.
Briffa, K. R., Schweingruber, F. H., Jones, P. D., Osborn, T. J., Shiyatov, S. G., Vaganov, E. A., 1998. Reduced sensitivity of recent tree-growth to temperature at high northern latitudes. Nature 391, 678–682. Buckley, B. M., Palakit, K., Duangsathaporn, K., Sanguantham, P., Prasomsin, P., 2007. Decadal scale droughts over northwestern Thailand over the past 448 years: links to the tropical Pacific and Indian Ocean sectors. Climate Dynamics 29, 63–71.
Büntgen, U., Frank, D. C., Schmidhalter, M., Neuwirth, B., Seifert, M., Esper, J. 2006. Growth/climate response shift in a long subalpine spruce chronology. Trees, 20, 99–110.
Carrer, M., Urbinati, C., 2004. Age-dependent tree-ring growth responses to climate in Larix decidua and Pinus cembra. Ecology 85, 730–740.
Chabot, B. F., Hicks, D. J., 1982. The ecology of leaf life spans. Annual Review of Ecology and Systematics, 13, 229–259.
Chalise, S. R., Kansakar, S. R., Rees, G., Croker, K., Zaidman, M. (2003). Management of water resources and low flow estimation for the Himalayan basins of Nepal. Journal of Hydrology 282, 25-35. Chang, C. P., Harr, P., Ju, J., 2001. Possible roles of Atlantic circulations on the weakening Indian monsoon rainfall-ENSO relationship. Journal of Climate 14, 2376-2380. Chaudhary, V., Bhattacharyya, A., Yadav, R.R., 1999. Tree-ring studies in the Eastern Himalaya. IAWA J 20, 317–324.
Chapter 10 Bibliography
119
Cook, E.R, Kairiukstis, L.A., 1990. Methods of dendrochronology: applications in the environmental sciences. Kluwer Academic Publishers, Dordrecht 394 pp.
Cook, E.R., Peters, K., 1981. The smoothing spline: a new approach to standardizing forest interior tree-ring series for dendroclimatic studies. Tree-ring Bulletin 41, 45-53 Cook, E.R., 1985. A Time Series Approach to Tree-ring Standardisation. Ph.D. thesis. University of Arizona, Tucson, Arizona USA. Cook, E. R., Meko, D. M., Stahle, D. W., Cleaveland, M. K., 1999. Drought reconstructions for the continental United States. Journal of Climate, 12, 1145-1162.
Cook, E. R., Buckley, B. M., D'Arrigo, R. D., Peterson, M. J., 2000. Warm-season temperatures since 1600 BC reconstructed from Tasmanian tree rings and their relationship to large-scale sea surface temperature anomalies. Climate Dynamics, 16, 79-91.
Cook, E. R., Krusic, P. J., Jones, P. D., 2003. Dendroclimatic signals in long tree-ring chronologies from the Himalayas of Nepal. International Journal of Climatology, 23, 707-732. Cook, E. R., Woodhouse, C. A., Eakin, C. M., Meko, D. M., Stahle, D. W., 2004. Long-term aridity changes in the western United States. Science 306, 1015-1018. Cook, E. R., Anchukaitis, K. J., Buckley, B. M., D’Arrigo, R. D., Jacoby, G. C., Wright, W. E., 2010. Asian monsoon failure and megadrought during the last millennium. Science, 328, 486-489. . Cooper, W. S., 1916. Plant successions in the Mount Robson Region, British Columbia. The Plant World 19, 211–238.
Cropper, J.P., 1985. Tree-Ring Response Functions: An Evaluation by Means of Simulations. Ph.D dissertation, The University of Arizona, Tucson. University Microfilms International, Ann Arbor. Cruz, R., 2007. Asia. Pages 469–506 in M. Parry, et al. editors. Climate change 2007: impacts, adaptation and vulnerability. Contribution of Working Group II to the Fourth assessment report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom.
Chapter 10 Bibliography
120
Dai, A., Trenberth, K. E., Qian, T., 2004. A global dataset of Palmer Drought Severity Index for 1870-2002: Relationship with soil moisture and effects of surface warming. Journal of Hydrometeorology 5, 1117-1130. Dai, A., 2011. Characteristics and trends in various forms of the Palmer Drought Severity Index during 1900–2008. Journal of Geophysical Research: Atmospheres (1984–2012), 116, D12115, doi:10.1029/2010JD015541. D'Arrigo, R. D., Jacoby, G. C., Krusic, P. J., 1994. Progress in dendroclimatic studies in Indonesia. Terrestrial, Atmospheric and Oceanic Sciences, 5, 349-363. D'Arrigo, R., Wilson, R., Jacoby, G., 2006. On the long-term context for late twentieth century warming. Journal of Geophysical Research, 111, D03103. Druckenbrod, D. L., Mann, M. E., Stahle, D. W., Cleaveland, M. K., Therrell, M. D., Shugart, H. H., 2003. Late-eighteenth-century precipitation reconstructions from James Madison's Montpelier plantation. Bulletin of the American Meteorological Society 84, 57-71. Dyurgerov, Mark., 2002. "Glacier Mass Balance and Regime: Data of Measurements and Analysis." INSTAAR Occasional Paper No. 55, ed. M. Meier and R. Armstrong. Boulder, CO: Institute of Arctic and Alpine Research, University of Colorado. Distributed by National Snow and Ice Data Center, Boulder, CO. Efron, B and Gong, G., 1983. A leisurely look at the bootstrap, the jackknife, and cross-validation. The American Statistician 37, 36 – 48. Efron, B., 1979. Bootstrap methods: another look at jackknife. The Annals of Statistics 7, 1 – 26. Efron, B., 1982. The Jackknief, the bootstrap and other resampling plans. Society for Industrial and Applied Mathematics, Providence, RI, pp. 92. Efron, B., 1983, Estimating the Error Rate of a Prediction Rule: Improvement of Cross-Valuation. Journal of the American Statistical Association 78, 316–331. Efron, B., & Tibshirani, R. J. (1993). Confidence intervals based on bootstrap “tables,”. An Introduction to the Bootstrap. London, UK, Chapman and Hall/CRC, 160-162.
Chapter 10 Bibliography
121
Efron, B. and Tibshirani, R.J.1997 "Improvements on cross-validation: The .632+ bootstrap method," Journal of the American Statistical Association, 92, 548-560. Esper J, Cook, E.R., Schweingruber. F.H., 2002. Low-frequency signals in long tree-ring chronologies for reconstructing past temperature variability. Science 295, 2250–2253 Fan, Z.X, Bräuning, A, Bao Y, Cao K.F., 2009. Tree ring density-based summer temperature reconstruction for the central Hengduan Mountains in southern China. Glob Planet Change 65, 1–11 Favier, V., Wagnon, P., Chazarin, J-P., Maisincho, L., Coudrain, A., 2004. One-year measurements of surface heat budget on the ablation zone of Antizana Glacier 15, Ecuadorian Andes. Journal of Geophysical Research 109, D18105. doi:10.1029/ 2003JD004359.
Folland, C. K., Karl,T. R., Christy, J. R., Clarke, R. A., Gruza, G. V., Jouzel, J., Mann, M. E., Oerlemans, J., Salinger, M. J., Wang, S. W., 2001, Observed climate variability and change, in Climate Change 2001: The Scientific Basis, edited by J. T. Houghton et al., pp. 99–181, Cambridge Univ. Press, New York.
Francou, B., Ramírez, E., Cáceres, B., Mendoza, J., 2000. Glacier evolution in the Tropical Andes during the last decades of the 20th century: Chacaltaya, Bolivia, and Antizana, Ecuador. Ambio 29, 416–422.
Francou, B., Vuille, M., Wagnon, P., Mendoza, J., Sicart, J. E., 2003. Tropical climate change recorded by a glacier in the central Andes during the last decades of the twentieth century: Chacaltaya, Bolivia, 16 S. Journal of Geophysical Research 108, 4059 DOI:10.129/2002JD002473.
Fritts HC (1976) Tree-rings and climate. Academic Press, London pp 567
Fritts, H.C, Guiot, J., Gordon, G.A., 1990.Verification. In: Cook ER, Kairiukstis LA (eds) Methods of dendrochronology: applications in the environmental sciences. Kluwer, Dordrecht, pp 178–185.
Fritts, H. C., 1991, Reconstructing Large-scale Climatic Patterns From Tree-Ring Data, 286 pp., Univ. of Ariz. Press, Tucson. Fritts, H. C. 1976 Tree Rings and Climate. Academic Press, London.
Chapter 10 Bibliography
122
Fritts, H. C., Blasing, T. J., Hayden, B. P., Kutzbach, J. E., 1971. Multivariate techniques for specifying tree-growth and climate relationships and for reconstructing anomalies in paleoclimate. Journal of applied meteorology, 10, 845-864. Garavaglia, V., Pelfini, M., Bollati, I., 2010. The influence of climate change on glacier geomorphosites: the case of two Italian glaciers (Miage Glacier, Forni Glacier) investigated through dendrochronology. Geomorphologie- Relief Processus Environnement 153–164.
Gedalof, Z. E., Smith, D. J., 2001. Interdecadal climate variability and regime-scale shifts in Pacific North America. Geophysical Research Letters 28, 1515-1518. Gindl, W., 1999. Climatic significance of light rings in timberline spruce, Picea abies, Austrian Alps. Arctic, Antarctic, and Alpine Research 242-246. Gou X., Chen F., Yang M., Jacoby, G., Peng J., Zhang Y., 2006. A comparison of tree-ring records and glacier variations over the past 700 years, northeastern Tibetan Plateau. Annals of Glaciology 43, 86-90. Gou, X., Chen, F., Cook, E., Jacoby, G., Yang, M., Li, J., 2007. Streamflow variations of the Yellow River over the past 593 years in western China reconstructed from tree rings. Water Resources Research 43, W06434. Graybill, D. A., Idso, S. B., 1993. Detecting the aerial fertilization effect of atmospheric CO2 enrichment in tree-ring chronologies. Global Biogeochemical Cycles, 7(1), 81–95. Guiot, J., 1985. The extrapolation of recent climatological series with spectral canonical regression. Journal of Climatology 5,325–335.
Guiot, J., 1990. Methods of calibration. In: Methods of Dendrochronology: Application to Environmental Sciences, Cook, E., Kairiukstis, L., (Eds.) pp. 165–178. Kluwer Academic Press and IlASA, Dordrecht.
Guiot, J., 1991. The bootstrapped response function. Tree-ring bulletin, 51, 39–41. Guiot, J., Goeury, C., 1996. PPPBASE, a software for statistical analysis of paleoecological and paleoclimatological data. Dendrochronologia, 14, 295–300.
Chapter 10 Bibliography
123
Hardman, G., Reil, O.E., 1936. The relationship between tree-growth and stream runoff in the Truckee River basin, California-Nevada. The University of Nevada Agricultural Research Station, Bulletin No.141, pp. 1–38. Havranek, M., Tranquillini, W., 1995. Physiological processes during their winter dormancy and their ecological significance, in: Ecophysiology of coniferous forest, edited by: Smith, W. K. and Hinkley, T. M., Academic Press, New York, USA, 95–124
Hawley, F.M., 1937. Relationship of southern cedar growth to precipitation and runoff. Ecology 8,398-405.
Holmes, R.L., 1983.Computer-assisted quality control in tree-ring dating and measurement. Tree-Ring Bull 43, 69–78.
Hughes, M. K., Graumlich, L. J., 1996. Multimillennial dendroclimatic studies from the western United States. Nato Asi Series I, Global Environmental Change 41, 109-124.
Hughes, M.K., Kelly, P.M., Pilcher, J.R., LaMarche, V.C. Jr. (Eds.) 1982. Climate from Tree Rings. Cambridge University Press, Cambridge. 233 pp.
India State of Forest Report, 2011. Forest Survey of India, (Ministry of Environment and Forests) Government of India, Dehradun. 214-218 pp.
Metz, B., Davidson, O., De Coninck, H. C., Loos, M., Meyer, L. A. 2005. IPCC, 2005. IPCC special report on carbon dioxide capture and storage. Prepared by Working Group III of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA, 442 pp.
IPCC, 2007: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S.D., Qin, M., Manning, Z., Chen, M., Marquis, K.B., Averyt, M., Tignor, H.L., Miller (Eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 996 pp.
Isdale, P. J., Stewart, B. J., Tickle, K. S., Lough J. M., 1998. Paleohydrological variation in a tropical river catchment: A reconstruction using flourescent bands in corals of the Great Barrier Reef, Australia, Holocene 8, 1–8.
Chapter 10 Bibliography
124
Jacoby Jr, G. C., D'Arrigo, R., 1989. Reconstructed Northern Hemisphere annual temperature since 1671 based on high-latitude tree-ring data from North America. Climatic Change, 14, 39-59. Jacoby, G. C., D’Arrigo. R. D., 1990. Teak (Tectona grandis L. F.), a tropical species of large scale dendroclimatic potential, Dendrocronologia 8, 83–98. Jones, P. D., Mann, M. E., 2004. Climate over past millennia. Reviews of Geophysics 42, RG2002.
Karl, T. R., Knight, R. W., 1998. Secular trends of precipitation amount, frequency, and intensity in the United States. Bulletin of the American Meteorological society, 79, 231-241. Kaspari S, Mayewski P, Kang S, Sneed S, Hou S, Hooke R, Kreutz K, Introne D, Handley M, Maasch K, Qin D, Ren J., 2007. Reduction in northward incursions of the South Asian monsoon since 1400 A.D. inferred from a Mt. Everest ice core. Geophys Res Lett 34:L16701. doi:10.1029/2007GL030440
Kern, Z., Popa, I., 2007. Climate–growth relationship of tree species from a mixed stand of Apuseni Mts., Romania. Dendrochronologia, 24,109-115.
Körner, C., 1998. A re-assessment of high elevation treeline positions and their explanation. Oecologia 115, 445–459.
LaMarche V.C., Mooney, H.A., 1967. Altithermal timberline advance in western United States. Nature 213, 980–982
LaMarche, V. C., Fritts, H. C., 1971. Tree rings, glacial advance, and climate in the Alps. Zeitschrift fur Gletscherkunde und Glazialgeologie 7, 125–131.
LaMarche, V. C., 1973. Holocene climatic variations inferred from treeline fluctuations in the White Mountains, California. Quaternary Research, 3, 632–660
LaMarche, V. C., 1974. Paleoclimatic Inferences from Long Tree-Ring Records Intersite comparison shows climatic anomalies that may be linked to features of the general circulation. Science 183, 1043–1048.
Li, J., Xie, S. P., Cook, E. R., Huang, G., D'Arrigo, R., Liu, F., Zheng, X. T., 2011. Interdecadal modulation of El Niño amplitude during the past millennium. Nature Climate Change, 1, 114–118.
Li, J., Xie, S. P., Cook, E. R., Morales, M. S., Christie, D. A., Johnson, N. C., Chen, F., D’Arrigo, R., Fowler, A.M., Gou, X ., Fang, K., 2013. El Nino modulations over the past seven centuries. Nature Climate Change 2, 1–5
Liang, E., Shao, X., Eckstein, D., Huang, L., Liu, X. 2006. Topography- and species-dependent growth responses of Sabina przewalskii and Picea crassifolia to climate on the northeast Tibetan Plateau. Forest Ecology and Management, 236, 268–277.
Liang, E., Shao, X., Qin, N., 2008. Tree-ring based summer temperature reconstruction for the source region of the Yangtze River on the Tibetan Plateau. Global and Planetary Change 61, 313–320.
Luckman, B. H., 1988. Dating the moraines and recession of Athabasca and Dome Glaciers, Alberta, Canada. Arctic and Alpine Research 20, 40–54
Luckman, B.H., 1994. Glacier fluctuation and tree-ring records for the last millennium in the Canadian Rockies. Quaternary Science Reviews 12, 441-450.
Luckman, B.H., 1995. Calendar-dated, early 'Little Ice Age' glacier advance at Robson Glacier, British Columbia, Canada. The Holocene, 5,149–159.
Luckman, B.H., 1996. Dendroglaciology at Peyto Glacier, Alberta, Canada. In: J.S. Dean, D.M. Meko, and T.W. Swetnam, eds., Tree Rings, Environment, and Humanity. Radiocarbon 1996, 679–688.
Luckman, B. H., Briffa, K. R., Jones, P. D., Schweingruber, F. H., 1997. Tree-ring based reconstruction of summer temperatures at the Columbia Icefield, Alberta, Canada, AD 1073-1983. The Holocene, 7, 375-389.
Luckman, B. H., 2000. The Little Ice Age in the Canadian Rockies. Geomorphology 32, 357–384.
Luitel, K. K., Shrestha, D. G., Sharma, N. P., Sharma, R. K., 2012. Impact of Climate Change on East-Rathong Glacier In Rangit Basin, West Sikkim. Climate Change in Sikkim Patterns, Impacts and Initiatives. Information and Public Relations Department, Government of Sikkim, Gangtok.
Managave, S. R., Sheshshayee, M. S., Borgaonkar, H. P., Ramesh, R., 2010. Past break‐monsoon conditions detectable by high resolution intra‐annual δ18O analysis of teak rings. Geophysical Research Letters 37, 5.
Chapter 10 Bibliography
126
Managave, S. R., Sheshshayee, M. S., Borgaonkar, H. P., Ramesh, R., 2010. Intra-annual oxygen isotope variations in central Indian teak cellulose: possibility of improved resolution for past monsoon reconstruction. Current Science, 98, 930-937.
Mani, A., 1981., The climate of the Himalayas. In: Lall, J.S., Moddie, A.D. (Eds). The Himalayas–Aspects of Change, Oxford University Press. 3-15 pp.
Mann, M. E., 2001. Climate during the past millennium. Weather, 56, 91-102.
Mann, M. E., Park, J., 1994. Global‐scale modes of surface temperature variability on interannual to century timescales. Journal of Geophysical Research: Atmospheres (1984–2012), 99(D12), 25819-25833.
Mann, M. E., Bradley, R. S., Hughes, M. K., 1998. Global-scale temperature patterns and climate forcing over the past six centuries. Nature, 392, 779-787.
Masters, T., 1995. Advanced Algorithms for Neural Networks: A C++ Sourcebook, NY: John Wiley and Sons, ISBN 0-471-10588-0
Matthews, F., 1939. Report of Committee on Glaciers, April 1939, Eos Trans. AGU, 20, 518–523. Matthews, J. A., 1977. Glacier and climatic fluctuations inferred from tree-growth variations over the last 250 years, central southern Norway. Boreas 6, 1–24.
Mathews, W. H., 1951. Historic and prehistoric fluctuations of alpine glaciers in the Mount Garibaldi map-area, southwestern British Columbia. Journal of Geology 59, 357–380.
Mayewski, P.A., Jeschke, P.A., 1979. Himalayan and Trans- Himalayan glacier fluctuations since A.D. 1812. Arctic and Alpine Research 11, 267 – 287.
Mayewski, P. A., Pregent, G. P., Jeschke, P. A., Ahmad, N., 1980. Himalayan and Trans-Himalayan glacier fluctuations and the south Asian monsoon record. Arctic and Alpine Research, 171-182.
Meko, D. M., 1981. Applications of Box-Jenkins methods of time series analysis to the reconstruction of drought from tree rings, Ph.D. thesis, 149 pp., Univ. of Ariz., Tucson.
Chapter 10 Bibliography
127
Meko, D., Cook, E. R., Stahle, D. W., Stockton, C. W., Hughes, M. K., 1993. Spatial patterns of tree-growth anomalies in the United States and southeastern Canada. Journal of Climate, 6, 1773-1786. Meko, D.M., Graybill, D.A., 1995. Tree-ring reconstruction of upper Gila River discharge. Water Resour. Bull. 314, 605–616. Meko, D., Graybill, D. A., 1995. Tree‐ring reconstruction of upper gila rwer discharge. JAWRA Journal of the American Water Resources Association 31, 605-616. Meko, D. M., Therrell, M. D., Baisan, C. H., Hughes, M. K., 2001. Sacramento River flow reconstructed to AD 869 From Tree Rings. JAWRA Journal of the American Water Resources Association 37, 1029-1039. Pant, G. B., Kumar, K. R., Borgaonkar, H. P., Okada, N., Fujiwara, T., & Yamashita, K., 2000. Climatic response of Cedrus deodara tree-ring parameters from two sites in the western Himalaya. Canadian Journal of Forest Research 30, 1127-1135. Payette, S., Fillion, L., Delwaide, A., Begin, C., 1989. Reconstruction of tree-line vegetation response to long-term climate change. Nature 341, 429–432
Pederson, N., Jacoby, G., D’Arrigo, R., Cook, E., Buckley, B., Dugarjav, C., Mijiddorj, R., 2001. Hydrometeorological reconstructions for northeastern Mongolia derived from tree rings: 1651-1995. Journal of Climate 14, 872-881. Porter, S. C., 1981. Glaciological evidence of Holocene climatic change. In Climate and history–studies in past climates and their impact on man (T. M. L. Wigley, M. J. Ingram and G. Farmer, Eds.), 82–110 pp. Cambridge University Press,Cambridge. Pumijumnong, N., Eckstein, D., Sass, U., 1995. Tree-ring research on Tectona grandis in Northern Thailand. IAWA J 16, 385–392 Raina, V. K., 2009. Himalayan glaciers: a state-of-art review of glacial studies, glacial retreat and climate change. Himalayan glaciers: a state-of-art review of glacial studies, glacial retreat and climate change. Somaru, R., Borgaonkar, H. P., Sikder, A. B., 2011. Growth and climate relationship in teak trees from Conolly’s plot, South India. Current Science 100, 630-633.
Chapter 10 Bibliography
128
Ram, S., Borgaonkar, H.P., Sikder, A.B., 2008.Tree-ring analysis of teak (Tectona grandis LF) in Central India and its probable linkage with moisture fluctuation: a case study. Journal of Earth System Science 117, 637–645
Ram, S., Borgaonkar, H. P., Sikder, A. B., 2010. Varying strength of the relationship between tree-rings and summer month moisture index (April–September) over central India: a case study. Quaternary International 212, 70-75.
Ram, S., 2012. Tree growth–climate relationships of conifer trees and reconstruction of summer season Palmer Drought Severity Index (PDSI) at Pahalgam in Srinagar, India. Quaternary International 254, 152-158.
Ram, S., 2012. On the recent strengthening of the relationship between Palmer Drought Severity Index and teak Tectona grandis tree-ring width chronology from Maharashtra, India: A case study. Quaternary International 248, 92-97.
Ramesh, R., Bhattacharyya, S.K., Pant, G.B., 1989. Climatic significance of dD variation in a tropical tree species from India. Nature 337,149–150
Rao, R.K, Juneja, K.B.S., 1971. A handbook for field identification of fifty important timbers of India. Publication of Forest Research Institute and colleges, Dehra Dun .
Ribstein, P., Tiriau, E., Francou, B., Saravia, R.,1995. Tropical climate and glacier hydrology: acase study in Bolivia. Journal of Hydrology 165, 221–234.
Rossi, S., Deslauriers, A., Anfodillo, T., Morin, H., Saracino, A., Motta, R., Borghetti, M., 2006. Conifers in cold environments synchronize maximum growth rate of tree-ring formation with day length. New Phytologist 170, 301-310.
Sano, M., Buckley, B. M., & Sweda, T., 2009. Tree-ring based hydroclimate reconstruction over northern Vietnam from Fokienia hodginsii: eighteenth century mega-drought and tropical Pacific influence. Climate dynamics 33, 331-340. Schulman, E., 1945. Runoff histories in tree rings of the Pacific slope. The Geographical Review 5, 59-73. Schulman, E., 1951. Tree-ring indices of rainfall,temperature, and river flow. In T.F. Malone, ed.,Compendium of Meteorology,.American Meteorological Society, 1024-1029 pp. Schulman, E., 1945. Runoff histories in tree rings of the Pacific slope. Geographical Review 35, 59-73.
Chapter 10 Bibliography
129
Schweingruber, F.H., 1987. Tree Rings: basics and applications of Dendrochronology.” Kluwer Academic Publishers, Dordrecht, Netherlands. 276 pp.
Shah, S.K, Bhattacharyya, A., Chaudhry, 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.
Shao, J., 1993, "Linear model selection by cross-validation," J. of the American Statistical Association 88, 486-494.
Shrestha, A. B., Devkota, L. P., 2010. Climate change in the Eastern Himalayas: observed trends and model projections. International Centre for Integrated Mountain Development (ICIMOD).pp 13.
Sigafoos, R.S., Hendricks, E.L., 1961. Botanical evidence of the modern history of Nisqually Glacier, Washington. U.S. Geological Survey Professional Paper 387-A: 1–20.
Singh, J., Yadav, R. R., 2000. Tree-ring indications of recent glacier fluctuations in Gangotri, Western Himalaya, India. Current Science 79, 1598-1601.
Singh, J., Yadav, R.R., Dubey, B., Chaturvedi, R., 2004. Millennium-long ring-width chronology of Himalayan cedar from Garhwal Himalaya and its potential in climate change studies. Current Science 86, 590-593. Singh, J., Yadav, R.R., 2005. Spring precipitation variations over the western Himalaya, India since AD 1731 as deduced from tree rings. Journal of Geophysical Research 110, D01110. http://dx.doi.org/10.1029/2004JD004855. Singh, J., Park, W.-K., Yadav, R.R., 2006. Tree-ring-based hydrological records for western Himalaya, India, since AD 1560. Climate Dynamics 26, 295-303. Singh, J., Yadav, R.R., 2007. Dendroclimatic potential of millennium-long ring-width chronology of Pinus gerardiana from Himachal Pradesh, India. Current Science 93, 833-836. Singh, J., Yadav, R.R., Wilmking, M., 2009. A 694-year tree-ring based rainfall reconstruction from Himachal Pradesh, India. Climate Dynamics 33, 1149-1158. Singh, J., Yadav, R.R., 2012. Application of tree-ring data in development of long term discharge of River Satluj. Current Science 103, 1452-1454.
Chapter 10 Bibliography
130
Singh, K. K., Gaira, K. S., Rai, L. K., 2011. Agricultural scenario vis-a-vis the pollinator elements of the Sikkim Himalayan region. Biodiversity of Sikkim-Exploring and Conserving a Global Hotspot. Information and Public Relations Department, Government of Sikkim, 427-441.
Smith, D., Lewis, D., 2007.Glacial landforms, tree-rins/Dendrochronlogiology, pp 988-199.
Smith, W.W., Cave, G.H., 1913.The vegetation of Zemu and Llonakh valley of Sikkim. Records of the Botanical Survey of India 4 (5).
Somaru, R., Borgaonkar, H. P., Sikder, A. B., 2008. Tree-ring analysis of teak (Tectona grandis LF) in Central India and its probable linkage with moisture fluctuation: a case study. Journal of Earth System Science117, 637-645.
Stockton, C.W., Jacoby, G.C., 1976. Long-term surface water supply and streamflow trends in the upper Colorado River basin based on tree-ring analyses. Lake Powell Res. Proj. Bull. 18, 1–70. Stokes, M.A., Smiley, T.L., 1968. An Introduction to Tree-ring Dating. The University of Chicago Press, Chicago. Su, Z., Y. Shi., 2002. Response of monsoonal temperate glaciers to global warming since the Little Ice Age. Quaternary International 97-98, 123-131. Sumi., R.P., 1994.Rainfall variation along testa valley in Mountainous slope of Sikkik Mausam 45, 165-170.
Tarr, R. S., Martin, L., 1914. Alaskan glacier studies of the National Geographic Society in the Yakutat Bay, Prince William Sound and lower Copper River regions. National Geographic Society. Tibshirani, R., 1996. "A comparison of some error estimates for neural network models," Neural Computation, 8, 152-163. Torrence, C., Compo, G. P., 1998. A practical guide to wavelet analysis. Bulletin of the American Meteorological society, 79(1), 61-78.
Chapter 10 Bibliography
131
UNEP and WGMS. 2008. Global glacier changes: facts and fi gures. United Nations Environment Programme and World Glacier Monitoring Service. 88 p. http://www.grid.unep.ch/glaciers/.
UNEP. 2008. Global Glacier Changes: facts and figures. WGMS.
Urban, F., Cole, J., Overpeck, J., 2000. Influence of mean climate change on climate variability from a 155-year tropical Pacific coral record. Nature 407,989-993. Urrutia, R.B., Lara, A., Villalba, R., Christie, D.A., Le Quesne, C., Cuq, A., 2011. Multicentury tree ring reconstruction of annual streamflow for the Maule River watershed in south central Chile. Water Resources Research 47,W06527. http://dx.doi.org/10.1029/2010WR009562. Vaganov, E. A., Hughes, M. K., Kirdyanov, A. V., Schweingruber, F. H., Silkin, P. P. 1999. Influence of snowfall and melt timing on tree growth in subarctic Eurasia. Nature 400(6740), 149-151.
Webster, P. J., Palmer, T. N., 1997. The past and the future of El Niño. Nature, 390, 562-564.
Wagnon, P., Ribtein, P., Francou, B., Sicart, J.E., 2001. Anomalous heat and mass budget of glacier Zongo, Bolivia, during the 1997/1998 El Niño year. Journal of Glaciology 47, 21–28.
Wigley, T. M. L., Jones, P. D., Briffa., K. R., 1988. Detecting the effects of acidic deposition and CO2-fertilization on tree growth, in Methods of Dendrochronology Applications in the Environmental Science, Kairiukstis, L., et al. (Eds.),. 239–253 pp, Polish Acad. of Sci., Warsaw.
Woodhouse, C.A., 2000. Extending hydrologic records with tree rings. Water Resour: Impact 2, 25–27. Woodhouse, C.A., Meko, D.M., 2002. Introduction to tree-ring based streamflow reconstructions. Southwest Hydro 1, 14–15. Wu, X, Shao, X., 1995. Status and prospects of dendrochronological study in Tibetan Plateau. Dendrochronologia 13, 89–98.
Chapter 10 Bibliography
132
Yadav, R.R., Park, W.K., Bhattacharyya, A., 1997. Dendroclimatic reconstruction of April-May temperature fluctuations in the western Himalaya of India since A.D. 1698. Quaternary Research 48, 187-191.
Yadav R.R., Park W.K., Bhattacharya, A., 1999. Spring temperature fluctuations in the western Himalayan region as reconstructed from tree-rings; AD 1390–1987. The Holocene 9, 85–90
Yadav, R.R., Park, W.K., 2000. Precipitation reconstruction using ring-width chronology of Himalayan cedar from western Himalaya: preliminary results. Proceedings, Indian Academy of Sciences (Earth andPlanetary Science) 109, 339-345. Yadav, R.R., Singh, J., 2002. Tree-ring-based spring temperature patterns over the past four centuries in western Himalaya. Quaternary Research 57, 299-305. Yadav, R. R., Park, W. K., Singh, J., Dubey, B., 2004. Do the western Himalayas defy global warming? Geophysical Research Letters 31, L17201. Yadav, R.R., 2011a. Tree-ring evidence of 20th century precipitation surge in monsoon shadow zone of western Himalaya, India. Journal of Geophysical Research 116. http://dx.doi.org/10.1029/2010JD014647. Yadav, R.R., 2011b. Long-term hydroclimatic variability in monsoon shadow zone of western Himalaya, India. Climate Dynamics. http://dx.doi.org/10.1007/s00382-010-0800-8. Yadav, R.R., Braeuning, A., Singh, J., 2011. Tree ring inferred summer temperaturevariations over the last millennium in western Himalaya, India. Climate Dynamics 36, 1545-1554. Yadav, R.R., 2013. Tree ring-based seven-century drought records for the Western Himalaya, India. Journal of Geophysical Research 118. http://dx.doi.org/10.1029/ 2012JD018661. Yao, T., Wang, Y., Liu, S., Pu, J., Shen, Y., Lu, A., 2004. Recent glacial retreat in High Asia in China and its impact on water resource in Northwest China. Science in China Series D: Earth Sciences,47, 1065-1075.
Zemp, M., Hoelzle, M., Haeberli, W., 2009. ‘Six decades of glacier mass balance observations: a review of the worldwide monitoring network.’ Annals of Glaciology 50,101-111
Summary
1
Title of this dissertation “Application of multi-proxy tree- ring parameters in
the reconstruction of past climate vis-à-vis glacier fluctuations from the Eastern
Himalaya”. The research work included this dissertation was carried out using multi-
proxy tree ring parameters i.e. width of tree ring as a whole, early wood and latewood
separately provided high resolution temporal climate reconstruction for longer time-
span (1759 to 2010 AD). This reconstructed climate data was analyzed to seek a long-
term linkage with El Niño and advancement and retreat of Zemu glacier. This work is
also supplemented with reconstruction discharge of Zemu Chu originated from Zemu
glacier. In this pursuit, a good amount of tree ring data has been generated from Zemu
glacier. These data are from several conifer trees viz., Abies densa, Juniperus recurva,
Larix griffithiana (Ring width, late wood, and early wood) growing sub alpine forest
and Juniperus squamata scrub growing at the tree-line zone close to snout of the
Zemu glacier. I reconstructed summer temperature in Zemu area and also explored
glacier and tree growth relationship of North Sikkim Eastern Himalaya. In this
context, I developed 12 chronologies from a transect of lower (m) to higher elevation
close to snout of the Zemu glacier and established site-specific climate growth
response. After that I made regional chronology using PCA to find out common
variance to enhance tree growth relation at regional level i.e. First PC#1 and the
second PC (PC#2) of the chronology PCA were significant, representing 37.21% and
18.18% of the total variance, respectively. Regional Response Function Analysis
suggested that average February-March temperature and July-September temperature
were limiting factors for tree-growth. I reconstructed maximum temperature of
March-April using Bootstrap Regression Method, maximum March-April temperature
was successfully reconstructed for the Zemu glacier valley regions. Secondly, I
reconstructed mean temperature of July-August using Linear Regression Method, viz.,
Correlated tree-ring chronologies (TRCs) were used as predictor variables in linear
regression (LR) to determine the optimal regression model with the highest skill.
Average July-August temperature was successfully reconstructed for the various
regions with R2 values ranging from 23.60%. And thirdly, I reconstructed January-
April discharge of Zemu Chu (1775 to 1996 AD) a tributary of Tista river using linear
regression (LR) and R2values ranging from 50.2%. Established relationships with
tree growth and glacier fluctuation history with available glacier fluctuation data.
Beside this, analysis of the relation of tree growth/ glacier retreat and El Nino 3.4 has
also been made. The major contributions of this dissertation are first, maximum
Summary
2
March-April temperature reconstruction. Secondly, mean July-August temperature
reconstruction in Zemu glacier valley North Sikkim (Eastern Himalaya) India, and
thirdly the first successful hydrological reconstruction, and conclusive evidence of
climate change signals and 3rd is glacier history with relation to tree growth and El
Nino 3.4. Generated data will allow for a better understanding of climate,
hydrological variability in the region and glacier history of this region. Based on tree-
ring, the data generated can be used to enhance the understanding of
paleoenvironments. A comparison of the tree-ring and glacier fluctuations
(advancement/retreat or mass balance positive/negative) records indicate that the trees
in the area are much sensitive and respond to glacier fluctuations. An overall
assessment of analyses of climate, stream flow and El Niño signatures in Zemu
glacier area presented in the present research work appears to have great promise and
should look forward to a more comprehensive analysis based on development of tree-
ring data in both spatial and temporal coverage close to snout of the glaciers of the
Eastern Himalaya. Because of the complex landscape in this mountainous region,
more tree-ring data are required for better understanding of the regional climate
variability. In the absence of long-term instrumental records, they permit exploration
of the relative contribution of changes in temperature to net mass balance. A detailed
analysis through multidisciplinary approach would bring out valuable information
regarding glacier advancement and retreat in greater details from the Himalyan