Figure 1.1. A record of volcanic sulphate concentrations in ppb derived from H2O4 measurements on the GISP2 ice core, Greenland (Zielinski et al.,… [309810]

LIST OF FIGURES

Figure 1.1. A record of volcanic sulphate concentrations in ppb derived from H2O4 measurements on the GISP2 [anonimizat] (Zielinski et al., 1994) 5

Figure 1.2. Sea-level curve for the western margin of the East Sea during the past 20 kyr (Tanabe et al., 2003). 12

Figure 2.1. [anonimizat] (modified from Tanabe et al., (2006)) 15

Figure 2.2. Simplified onshore Quaternary stratigraphic column of the RRD Basin (Tran & Nguyen, 1991) 17

Figure 2.3. [anonimizat], Vietnam 19

Figure 2.4. [anonimizat], [anonimizat] 20

Figure 2.5. Ba Be Lake and surrounding topography 21

Figure 2.6. [anonimizat] 21

Figure 3.1. Illustrating flow chart sampling and methodological procedures in RRD 23

Figure 3.2. [anonimizat] 24

Figure 3.3. Preparation sample for grain size analysis 24

Figure 3.4. Preparation and measurement of the grain size analyses were completed at the Key Lab ([anonimizat]) 26

Figure 3.5. [anonimizat] 27

Figure 3.6. Sample processing flow chart to prepare clay films and analysis clay minerals (Moore & Reynolds Jr, 1997) 28

Figure 3.7. Preparation sample for analysis 30

Figure 3.8. Vibration mixer with 12 tubes in the laboratory 30

Figure 3.9. Stable Isotope Mass Spectrometry (ISMS) – Nu Instruments systems (Vietnam National University) 31

Figure 3.10. Gas chromatography system (Vietnam National University) 31

Figure 3.11. EuroVector EA3000 system (Vietnam National University) 32

Figure 4.1. [anonimizat] 33

Figure 4.2. Tri-plot for textural analysis of all sediments in core sample RRD 34

Figure 4.3. Grain-size distribution of different kinds of sediment: (A), (B), (C) and (D) [anonimizat] 34

Figure 4.4. Sector plot showing the bivariate relationship between (a) [anonimizat] (µm) and sorting, (b) skewness and sorting, (c) skewness and kurtorsis 35

Figure 4.5. Grain-[anonimizat] (Tue et al., 2019) 36

Figure 4.6. [anonimizat] (m) 37

Figure 4.7. Sedimentary columns of the core sample in RRD. (A) Lithological characteristic of sediment core; (B) Mud content in sediment core; (C) Sedimentation rates were calculated by linear interpolation between 14C ages (Tanabe et al., 2006, Li, Saito, Matsumoto, Wang, Tanabe, et al., 2006) 39

Figure 4.8. The variation of LOI (%), C/N ratios, TOC (%) and δ13C (‰) [anonimizat] (m) 41

Figure 4.9. Variation in clay mineral (smectite, kaolinite and illite) proportion (%) [anonimizat] 42

Figure 4.10. [anonimizat] 43

Figure 4.11. Sedimentary parameter in grain size distribution from sediment core showing with depth (cm) 44

Figure 4.12. The variation of LOI (%), OC (%), C/N ratios, δ15N and δ13C in the sediment core with depth (cm) 45

Figure 4.13. Relationship between LOI and density of sediment (a), LOI and OC content (b), LOI and C/N ratios (c), LOI and δ13C values 46

Figure 4.14. Relationship between OC content and Norg (a), OC content and C/N ratios (b), OC content and δ13C values 47

LIST OF TABLES

Table 1.1. Climate stations in the North of Vietnam ((Sterling et al., 2008)) 13

Table 2.1. Features of the Red River (Milliman, 1995) 18

Table 2.2. Hydrological features in the Ba Be National Park 21

Table 3.1. Size scale adopted in the GRADISTAT program, a modified Udden (1914), Wentworth (1922) and Friedman and Sanders (1978) (Blott & Pye, 2001) 25

Table 3.2. The identification of clay minerals base on peak of X-ray diffraction 29

Table 4.1. Sediment parameters in grain size distribution 38

. GENERAL INTRODUCTION

1.1. The climate of the Holocene

Although the climate of Holocene has been important role in the growth and development of modern society, the knowledge about climate change during this time was limited. Decades of paleoclimate studies shows the frequency of extreme climatic event during the last glacial maximum (LGM). Reconstruction of paleoclimate changes is important not only for understanding the characteristic of the climatic features and human activities are impacting the climate system, but also providing a data systems support to run global climate models which in great detail and high accuracy (Berger, Mesinger and Sijacki 2012). The studies show the change in the spatial and seasonal incoming solar radiation due to the quasi-cyclic variations in the Earth’s orbit with the major cause of global climate variability is solar radiation.

Holocene climate research has acquired a key role in providing wealthy information about natural climate variability since the LGM. Pioneering Holocene climate research based on investigation about peat stratigraphy, megafossils and macrofossils of pine trees preserved in peat bogs in Scandinavia:

– Heinrich Dau (1829) became the first scientist to study of peat and pine stumps. He recognized and described several different types of peat bog, the occurrence of pine trunks in peat, and the stratigraphic differences in peat color and peat type. Moreover, He interpreted the occurrence of pine megafossils as reflecting a phase in his hypothetical forest history of Denmark. However, he did not have a chance to test his forest-history hypothesis.

– Japetus Steenstrup (1841) proposed four periods in Danish forest history and emphasized the importance of plant and animal remains preserved in peat bogs as the best available means of investigating past environmental changes, including climate. In doing so, he know as one of the fathers of Holocene paleoecology and climate research.

– Continuous investigation on pine megafossils in Denmark, Christian Vaupell (1857) investigated that there had been changes not only in moisture but also in temperature (to explain the changes in tree composition and growth) during the Holocene.

– Alfred Gabriel Nathorst (1870) investigated plant macrofossils in the clays underlying the peat in southern Sweden and subsequently in Europe. Temperature change immediately became recognized as the major factor influencing the history of plants and animals since the last Ice Age as a result of Nathorst’s macrofossil studies. He investigated the boreal and sub-boreal elements had immigrated during periods of continental climate, whereas the atlantic and sub-atlantic elements had immigrated during periods of oceanic climate.

– Axel Blytt (1876) proposed an elaborate theory for the immigration of the Norwegian flora and its various floristic elements during these oceanic and continental periods.

– Rutger Sernander (1881) combined the Swedish ideas of summer-temperature changes with Blytt’s moisture changes to propose the famous Blytt–Sernander four periods of post-glacial time. The Blytt-Sernander paradigm of Holocene climate change established in his doctoral thesis at the University of Uppsala (1894). After 1910, the Blytt-Sernander scheme was widely used in Scandinavia after accepting by International Geologic Congress in Stockholm.

– Fægri (1940) introduced the term Pre-Boreal as a unit between the Boreal and the Younger Dryas of the late-glacial.

– A major conflict and acrimonious debate ensued for about 20 years between the Blytt-Sernander scheme and Andersson’s approach was based exclusively on macrofossils. Lennart von Post (1946) proposed post-glacial climate history involved both broad-scale temperature changes (Andersson’ approach) and finer-scale precipitation changes (Blytt-Sernander scheme).

– Samuelsson (1916) analyzed the northern limit of hazel in Sweden in considerable detail and showed that summer temperature was far from uniform along the limit today. He investigated that the growing season could be longer during the Holocene. This idea was followed up in detail by Iversen (1944), Hintikka (1963) and other scientists.

– Granlund (1932) investigated several shifts in moisture during the late Holocene. Thus, He proposed that at least two “Sub-Atlantic” climate phases during Sernander’s Sub-Boreal period and after Sernander’s Fimbulwinter the climate changed twice over to the Sub-Boreal type and then again to the Sub-Atlantic.

– Thomas F. Jamieson (1865) proposed the paradigm of a Holocene climate optimum or thermal maximum. This is similar with the result of Robert Lloyd Praeger, Waldemar Christopher Brøgger (Brøgger 1901), etc.

Although the limitations of the various techniques available, the results of study by Dau, Steenstrup, Vaupell, Andersson, Jamieson, Praeger, Blytt, Sernander, Brøgger, and Granlund were very remarkable. However, they was largely ignored after the development of pollen analysis as a tool for reconstructing Holocene climate change. It was not until the mid-1970s that peat stratigraphy was revived in Scandinavia using new techniques for estimating peat humification, detecting changes in bog moisture using testate amoebae, and establishing detailed chronologies by radiocarbon dating(Aaby and Tauber 1975, Aaby 1976).

In 1916, Lennart von Post presented and demonstrated its potential as a technique for relative dating and for reconstructing past vegetation and past climate. He proposed the concept of regional parallelism, namely that the same overriding climate change will be reflected in different ways by different taxa in different geographic and climatic regions. Pollen stratigraphy soon became the only approach used to detect past climate changes and to identify the Blytt – Sernander periods. In the 1920s-1940s, many investigators made pollen-analytic studies in Europe, North America, South America, New Zealand, and China (Battarbee and Binney 2009). Continuous investigation on pollen diagrams from Europe, New Zealand, Tierra del Fuego, North America, Hawaii, and China, Lennart von Post proposed that there is a consistent three-fold division of Holocene pollen stratigraphies. Using an early form of smoothing and spectral decomposition, von Post identified two long-term trends and three cycles with period lengths of 1700, 800-900, and 200-400 years within the birch pollen curve. In 1946, He identified areas on Earth where pollen-analytic studies should be made to determine Holocene changes inthe Earth’s circulation systems. Major advances in pollen-analytic methodology and interpretation were made in the 1940s-1960s by Johs. Iversen, Knut Fægri, and others. From fossil pollen occurrences in Holocene sediments in southern Scandinavia, Iversen (1944) inferred that mid-Holocene summer temperaturesmay and winter temperaturesmay warmer than today. Iversen’s brilliant analysis were confirmed by Walther et al. (2005), who show shifts in its northern margin in the past 50 years in response to recent climate changes. Moreover, there was also evidence for the effects of prehistoric people on vegetation, and interest naturally centered on whether the extent of human activity and resulting land-use change were related to climate change.

The development of radiocarbon dating by Willard F. Libby in the early 1950s provided a means of deriving an absolute chronology for events in the Holocene. Harry Godwin (1966) was early applications in Holocene research, along with Eric Willis, Donald Walker, and others. Smith and Pilcher (1973) showed that major pollen-stratigraphic changes and pollen-zone boundaries were not synchronous, even within an area as small as the British Isles. A similar picture emerges at broader scales such as Europe or eastern North America. Szafer (1935) made a mapping pollen values at selected time intervals from as many localities as possible in Poland. Such maps have been compiled at a range of spatial scales, ranging from the European continental scale (Huntley and Birks 1983) to single countries. An alternative mapping approach, so-called iso-chrone maps, has also been developed (Moe 1970, Davis 1976, Birks 1989) to display the spatial patterns of the ages at which different pollen types have their first consistent occurrence or expansion, thereby illustrating the patterns of range expansion and contraction. An important development in Holocene climate research involved a combination of pollen analysis and radiocarbon dating to study pollen-stratigraphic changes in space and time. This development took advantage of natural climate gradients and major vegetational ecotones to study Holocene palynologic and hence vegetational changes in areas of potentially high sensitivity to climate change. By the early 1970s, Holocene researchers had accumulated large amounts of pollen-stratigraphic and other biologic data, often with radiocarbon dates providing an independent chronology. Climate reconstructions based on these data were primarily qualitative and based on indicator species or a comparative approach where modern and fossil assemblages were compared visually.

The international COHMAP project started in 1977 as climates of the Holocene. Paleoclimate modelers was built with pollen data and it quickly became global in its research area. Comparison between the results calculated which used to the Community Climate Model (CCM) for simulating past climates at 18 000, 15 000, 12 000, 9000, 6000, 3000, and 0 years ago and the available paleoclimate data, particularly terrestrial pollen, lake-levels, and pack-rat midden data and marine plankton data (Wright 1993). COHMAP was a major turning point in Holocene climate research. This model used what were then state-of-the-art climate models to simulate paleoclimate under specified boundary conditions. Moreover, it resulted in detailed compilations and syntheses of paleoclimate proxy data that were then used in data-model comparisons. In addition to it considered the global climate system as a whole and revealed the strong regional interconnections between different components of the Earth’s climate system. Further more it it revealed the remarkable spatial and temporal variation in circulation patterns and climate. Many of the researchers involved in COHMAP then attempted a unified global synthesis of paleovegetation data at the level of biomes for 6000 radiocarbon years ago (BIOME 6000 project) and designed to be comparable to other global syntheses of sea-surface temperatures, ice sheets (CLIMAP Project Members) and lake-level changes. Climate models and paleoclimate modeling have made enormous advances and their contribution to our understanding of Holocene climate history (Valdes 2014).

Although Holocene climate research have been made advances in reconstructing the temporal and spatial patterns of variation in Holocene climate over the past 200 years, it also has several problems which relates to dating, data interpretation, basic gaps in our knowledge, and data availability (Battarbee and Binney 2009).

Holocene climate variability between 2.800-2.000 years BP and 1.500 year BP were determined and demonstrated in some areas in the world as the North Atlantic. Holocene climate changes as inferred from ice and marine core data. Besides, base on marine core data also shows some lines of evidence about past climates in other areas as West Africa (Tudhope et al. 2001). Reconstruction of paleoclimatic changes from a detailed stable-isotope record for the ice core of the Greenland which suggested that contrasts markedly with amplitude of temperature variations and mean temperature over the late Pleistocene and Holocene (Dansgaard et al. 1993). However, mean temperature trend variation from low values with a great fluctuation range to high values with a small fluctuation range on on decadal to millennial time-scales. Thus, if using the result from the Greenland as a basic establish for climate change during the Holocene, consequently limitation of evidences support to Holocene climate variability. However, along with the development of science, evidences about climate change in the world and the knowledge about external influences on the climate system, are very complicated.

1.2. Climate forcing factors during the Holocene

Climate is defined as an area's long-term weather patterns. It include temperature, humidity, atmospheric pressure, atmospheric particle count and other meteorological variables in a given region over long periods of time. A region’s climate is determined by the average weather in the location over a long period of time. The relative importance of external forcing factors such as orbital forcing, solar variability, and volcanic activity, and the complex interactions between these factors and their impacts on the Earth’s climate system. Moreover, the role of human activities in determining vegetation cover and land-use and in influencing Holocene climate.

1.2.1. The change in the solar irradiance

During the Holocene, solar irradiance has been changed due to different latitudes. In the late Holocene (about 6000 cal. year BP), solar irradiance was more influence than present time. In the early Holocene, The gradually influence by external forcing drives the Earth's climate system and it caused the abrupt climate change and occurred similar in many parts of the world, however, these variations is not immediately apparent. In the Northern Hemisphere, for instance, the ice melts slowly and can not raise sea levels fast. Thus, sea level could middle Holocene sea level after some millennials. Similarly, species evolve too slowly to adapt to global warming during Holocene. The processes of migration, competition and evolution, are contributed to changing the surface of the Earth.

1.2.2. Orbital forcing

The main factors that affected climate change during the Holocene and related to three orbital parameters including eccentricity, obliquity and precession. These parameters have little impact on the total amount of solar radiation received by the Earth from the Sun, however, they are strongly influenced by solar radiation variations on different seasons and latitudes. In the early Holocene, precessional changes due to perihelion at the time of the northern hemisphere summer solstice (today it is closer to the winter solstice). Therefore, daily insolation values at various latitudes (northern hemisphere) in the summer are higher than others (ranging from ~40 °W/m2 higher than today at 60 °N to 25 °W/m2 higher at the Equator). Beside, July insolation (radiation at the top, or outside, the atmosphere) has slowly decreased over the last 12,000 years. Thermal anomalies were generally smaller during southern hemisphere summers and it centred at lower latitudes (Mackay et al. 2003).

1.2.3. Solar activity and solar forcing

Orbital forcing includes the redistribution of incoming solar energy, both attitudinally and seasonally. Consequently, there are differential effects on the climate system that can lead to circulation changes, and there could be different responses to the forcing in the northern and southern hemispheres. The variation in solar irradiance might be expected to affect all parts of the Earth equally. However, this is not so because the response to solar irradiance forcing is amplified regionally, cause by feedbacks and interactions within the atmosphere (Rind 2002).

1.2.4. Volcanic forcing

Volcanic eruptions are the main cause of strong short-term (annual) climate forcing, cause by the injection of large amounts of gases (SO2, CO2, H2O, N2) and dust into the atmosphere perturbs the radioactive balance via enhanced absorption and scattering of solar radiation leading to a warming in the upper atmosphere and a cooling in the lower atmosphere. The variation of atmospheric circulation made some temperature anomalies in somewhere which mean climate in somewhere is cool and others is warm. The climate over mainland is warm during winter time (in high attitude areas), cause mainly by volcano eruptions in the 20th century.

Figure 1.1. A record of volcanic sulphate concentrations in ppb derived from H2O4 measurements on the GISP2 ice core, Greenland (Zielinski et al. 1994)

We could be investigated almost temperature variations which causes by volcano eruptions after some years. Thus, volcano eruptions made climate change which occurs short-time during Holocene. However, volcano eruptions occur regularly in the past or they happen with a large scale at the same time, are affected by long time (decades). Follow volcano eruptions during Holocene, a large of heat release which led to a widespread of the decreasing of temperature. Thus, mountain glaciers advance which made intensified atmospheric circulation and generally lowered temperatures (Battarbee and Binney 2009). During the Holocene, volcano eruptions occur regularly in period between 9500 and 11,500 yr BP (Lam 2003). In the early Holocene, volcanic activity released 110 ppb sulphate to the stratosphere (in the central Geenland) which can be many time higher than the amount of sulphate was released from the Tambora eruption in 1815 (the largest known historic eruption in the past century).

Ice core provide a wealthy information on volcanic eruptions. According to Zielinski et al. (1994) shows a record of volcanic sulfate derived from the GISP2 ice core. Base on the results of Crowley (2000) about the volcanoes responsible for the emissions, He shows the corresponding forcing for the past 1000 years (Crowley 2000). The study of volcanic ash layers in Siple Dome (Antarctica), Bay et al. (2004) investigated more evidences support to a causal connection between volcanism and millennial climate change (Battarbee and Binney 2009). The high volcanic eruption during the main deglaciation of the early Holocene shows that ice-sheet unloading and/or sea-level rise was responsible for increased volcanism during this period. Volcanic eruption, solar variability, are imported role of some factors that influence on climate change during Holocene in the last millennium. For example, abrupt cooling events over Europe and northern Alaska during the Maunder Minimum, cause by volcanic eruption occur at the same time during this time (Nishimura and Nishino 2003). Moreover, global warming due to CO2 concentration increases, deforestation occurs in 1850s. Consequently, climate is quite cool in late 19th century.

1.3. Climate variability during Holocene in the world

According to the results study base on the GISP2 Greenland ice core, Denton and Kalén (1973) proposed that glacier fluctuation record can be identified at 9000-8000, 6000-5000, 4200-3800, 3500-2500, 1200-1000, and 600-150 cal year BP (Denton and Karlén 1973, Mayewski et al. 2004, Martin, Leorri and McLaughlin 2007).

1.3.1. The period between 9000-8000 cal. year BP

This period was characterized by severe climatic disruption and it is unique among the Holocene rapid climate change (RCC) intervals because it occurs at a time when large Northern Hemisphere ice sheets were still present. In the Northern Hemisphere, an abrupt climate event about 8200 years ago brought generally cold and dry conditions (the “8k” event). The temperature of the North Atlantic appears to have been generally cool over much of the Northern Hemisphere throughout this interval on the basis evidences as ice rafting, strengthened atmospheric circulation over the North Atlantic and Siberia (Mayewski et al. 1997), and increased frequency of outbreaks of cold air from the northeast over the Aegean Sea (Rohling et al. 2002). Glacier retreat occurs in the European Alps which suggested the influence of dry northerly winds in the Northern Hemisphere (Hormes, Müller and Schlüchter 2001). Additional support to this hypothesis comes from northwestern North America and Scandinavia where mountain glacier advances occur and treeline limit is lower in Sweden (Mayewski et al. 2004).

In the early Holocene, widespread aridity midway through humid period in low latitudes (deMenocal et al. 2000, Martin et al. 2007), Moreover, In this RCC interval, tropical Africa summer monsoons over the Arabian and the India subcontinent weakened (Maley 1982, Gasse 2000) and trade wind and rainfall fluctuated dramatically over the Caribbean with evidences as persistent drought occurs in Haiti, the Amazon basin, Pakistan, and Africa, the water level of Lake Titicaca fell sharply and increased precipitation in the Middle East (Mayewski et al. 2004).

In the Southern Hemisphere, this period was characterized by weakness of polar atmospheric circulation over East Antarctica, decreased rates of snow accumulation (Steig et al. 2000) and temperature trends change in different areas between East and West Antarctica (Ciais et al. 1994, Masson et al. 2000), evidence as grounded ice in the Ross Sea retreat (Conway et al. 1999), the warm sea surface temperatures (SSTs) in the eastern and western flanks of the southern Africa, precipitation increases in Chile, cause by the intensification of southern mid-latitude wesrelies (Mayewski et al. 2004).

1.3.2. Classic “cool poles, dry tropics” RRC

Following the period between 9000-8000 cal. year BP, RCC interval occurs with varied intensity and geographic extent, but generally, it characterized by high latitude cooling and low latitude aridity (Mayewski et al. 2004). The most extensive of these reorganizations occurred from 6000-5000 to 3500-2500 cal. year BP, with less widespread events occurring at 4200-3800 cal. year BP and 1200-1000 cal. year BP.

In the Northern Hemisphere, at least two of the re-organizations in the North Atlantic (6000-5000, 3500-2500 cal. year BP) bracket and it features by ice rafting events (Martin et al. 2007), alpine glacier advances (Denton and Karlén 1973) and strengthened westerlies over the North Atlantic and Siberia (Meeker and Mayewski 2002). In Scandinavia, the treeline limit is extended and mountain glaciers advance in the first interval (from 6000 to 5000 cal. year BP), but the situation opposite in the second interval (from 3500 to 2500 cal. year BP) (Martin et al. 2007). Similarly, strengthened westerlies over central North America in the first interval, but westerly winds are weak in second interval (Mayewski et al. 2004). In otherwise, cooling event over the northeast Mediterranean is related to increasing of winter time continental/polar air (Mayewski et al. 2004).

At lower latitudes, the fist RCC interval marks by humid period in tropical Africa that began from the end of early Holocene to mid Holocene. This period was characterized by a long term trend of increasing rainfall variability and acidification (Gasse 2000), although in some areas (Pakistan, Florida and the Caribbean, etc), the climate became wetter (Mayewski et al. 2004). While, the decrease of rainfall over northwest India (Enzel et al. 1999) and southern Tibet, and the water level of Lake Titicaca drop during this interval. Trade wind over the Cariaco Basin and rainfall in Ecuador are relatively stable during RCC interval between 6000 and 5000 cal. year BP, but highly erratic during from 3500 to 2500 cal. year BP. Although, the climate became drier in East Africa, the Amazon Basin, Ecuador and the Caribbean region in the second RCC interval, the climate still is wet due to the stepwise decrease in Asian monsoon intensity (Dykoski et al. 2005).

In the Southern Hemisphere, mountain glacier advance in New Zealand, polar ice core records show the intensification of the atmospheric circulation that are occurred on a long term trend of increasing summer insolation (Mayewski et al. 2004). The climate overs South Georgia Island, SSTs of southern Africa and eastern South Africa is cool. At the middle latitudes, the climate of Chile is drier during the first interval, but wetter during the second interval (van Geel et al. 2000). According to the discontinuous lake sediment records from Antarctica, Ingolfsson et al. (1998) suggested conditions warmer than present due to increased southern summer insolation (Ingólfsson et al. 1998).

Although evidence for the RCC interval during from 4200-3800 and 1200-1000 cal. year BP appear in fewer of the record, however they are characterized by the apparent synchrony and large spatial distributions. Those record also contains evidence which suggested global scale teleconnections as for the other intervals (Mayewski et al. 2004). In the Northern Hemisphere, the westerly winds over the North Atlantic and Siberia are weak during this RCC interval, temperature drop in western North America (Scuderi 1993). Although other climatic disruptions, but those records generally are synchronous, distributed in a wide range, signs and intensities (Mayewski et al. 2004). Some evidences as during the interval 4200-3800 cal. year BP, mountain glacier advance in North America, but retreat in Euro, while Scandinavian ice seems great unaffected. Moreover, North Atlantic Deep Water production is low from 4200-3800 cal. year BP, however, it increased during the 1200-1000 cal. year BP (Mayewski et al. 2004).

At the latitudes, the climate is characterized by dry conditions in much of tropical Africa (Gasse 2000)and monsoonal Pakistan during two RCC interval. While the water level of Lake Titicaca fall, however, climate of Haiti is generally intenwet. Trade wind strength over the Cariaco Basin are intensified (Haug et al. 2001). Otherwise, mountain glacier advance in Kenya (Karlén et al. 1999) and aridity in Ecuador during the RCC interval from 1200 to 1000 cal. year BP.

In the Southern Hemisphere, the wind strength is only change in polar and temperatures in Taylor Dome, Antarctica are fluctuated. At the middle latitudes, the climate of Chile is dry during both of these RCC interval. During the 4200-3800 cal. year BP, climate over South Georgia Island is warm. Similarly, the results from lake sediment records in the Antarctic Peninsula and Victoria Land (Ingólfsson et al. 1998, Hjort et al. 1998). Besides, mountain glaciers advance in New Zealand and climate over eastern South Africa is cool and dry during the interval from 1200-1000 cal. year BP.

1.3.3. “Cool poles, wet tropics” RCC starting at ~600 cal. year BP

During this RCC interval, the climate in both regions are cold and windy, however, the low latitude aridity that prevailed during earlier intervals does not generally characterize the tropics during this most recent interval. Moreover, several records are missing recent sections, cause by artifacts of sampling and anthropogenic influences (Mayewski et al. 2004). Thus, this event in this study only from 600 to 150 cal. year BP.

In the Northern Hemisphere, mountain glacier advance and strengthened westerlies wind over the North Atlantic ad Siberia which showed the climate changes in this interval have the fastest and strongest during Holocene (O'brien et al. 1995), with exception of the short-lived about 8200 years ago (the “8k” event). At the low latitudes, the climate over the Cariaco Basin, Haiti and Florida became arid (Haug et al. 2001). Inversely, the climate in equatorial East Africa is humid conditions. Moreover, river discharge in Pakistan and Ecuador increased which indicated that both Indian monsoon and El Nino-southern Oscillation (ENSO) systems are affected (Mayewski et al. 2004).

In the Southern Hemisphere, the climate over a large proportion of the Antarctica Peninsula became warmer, however, East Antartica is cold (Morgan et al. 1997). Trade wind over East Antarctica and Amundsen Sea strengthen (Kreutz et al. 1997). The climate over South Georgia is cool, mountain glacier advance in New Zealand and rainfall in Chile increased during this interval. While, the SSTs of Benguela are cool, climate over southern Africa marks with climatic cooling and drying (Mayewski et al. 2004) .

1.4. Climate variability during Holocene in the Northern Vietnam

Study of the characteristic paleoclimate in Vietnam was accomplished by analyses pollen, diatoms, radiometric and conventional 14C ages (Tanabe et al. 2003, Hori et al. 2004, Funabiki et al. 2007). Analysing radiometric 14C and characteristics of Quaternary sedimentary facies in sediment core of the Red River Delta (RRD) are invested the transgression during the 9000-6000 cal. year BP which made the sea level was about 2-3 m high the present level and sea level was stable during the 6000-4000 cal. year BP, then sea level fell, at fist rapidly and later gradually, reaching the present level at 1000 cal. year BP (Tanabe et al. 2003, Funabiki et al. 2007).

Analysing characters in sediment Holocene of Red river mouth was assessed changing processes characterizing paleoenvironment based on comparing to climate directive in areas surrounding (China, Cambodia, East Sea) (Tanabe et al. 2006). Analysing pollen and radiometric 14C in sediment core showed characterizing climate in RRD during the last 5000 years, it included three cycles of cooling and warming: a cool and wet climate during 4530-3340 cal. year BP, 2100-1540 cal. year BP, and 620-130 cal. year BP; a warm and dry climate during 3340-2100 cal. year BP, 1540-620 cal. year BP và the present warm climate (Li et al. 2006b). The result also showed to affecting by human activities in RRD by beginning appearance pollen with origin wet rice (at 3.340 cal. year BP). Studying characteristics of pollen and spore in sediment core of the Hanoi area for reconstruction of paleoclimate changes during 10,500-5800 cal. year BP. In this period, the weather was characterized by warm and humid conditions, however, there is a short time of cold and dry climate during 8200-8400 cal. year BP.

Stable isotope ratio δ18O và δ 13C of the primary carbonate components (formed by the chemical deposition processes) carbonate biochemical (formed from the growth of crustaceans and mollusks) which was stored stable, long term and shows a strong connection with variable cycles condition temperature, humidity, rainfall and characterized by climate in the last time. In Vietnam, some results of research on restoration of climate condition in Vietnam have been made based on the stable isotope analysis method. The result of δ18O value in tree ring Tree ring samples were collected from old growth F.hodginsii (Po Mu) at the Mu Cang Chai area, Yen Bai province, Vietnam show δ18O significantly correlated with temperature, precipitation. Based on that, it is possible to calculate the PDSI drought index and the operation of ENSO in the last 300 years (Sano, Xu and Nakatsuka 2012). The result analysis of δ 18O value in cave stalactites in Ninh Binh province (Vietnam) also shows characterized by climate during the last 5500 year (Lin et al. 2006). Therefore, using stable isotopes method for reconstruction condition paleoclimate is a modern and reliable method higher than others used traditional directives (diatoms, pollen, facies…) (Berger et al. 2012, Stocker et al. 2014).

In Besides using stable isotopes method, determination of grain particles and content of clay minerals in sediment cores also illustrating the depositional environment and reconstructing the paleoenvironment and paleoclimate . (Sionneau et al. 2008, Li et al. 2017). Clay minerals are a dominant component of most marine sediments, and are mainly land-derived. Their geographic distributions and sources have been extensively investigated since the 1960s (Biscaye 1965). The content of clay minerals provided some information about climate condition in the last time (rainfall, runoff,…) in Northern Gulf of Mexico, bay of Bengal,…(Sionneau et al. 2008, Li et al. 2017). Semi-quantitative estimation of clay mineral abundances is based on peak areas, weighted by empirically estimated factors (Biscaye 1965, Biscaye et al. 1997). Since in most continental environments, smectite and kaolinite are generally formed by chemical weathering, while illite and chlorite are mostly inherited from ancient rocks modified by physical or moderate chemical weathering, the clay mineral ratio (Smectite + Kaolinite)/(Illite + Chlorite) indicates the type of weathering (chemical or physical) that affected the erodible sediments. According to the distribution of this (S + K)/(I + C) ratio, the North American continent east of the Rockies can be divided in different dominant weathering-regime areas (Sionneau et al. 2008).

1.4.1. Environment variability in the Northern Vietnam

Red River Delta is a vast triangular region located around the Red River basin area in the Northern Vietnam. The region consists of 11 cities and provinces, namely, Ninh Binh, Bac Ninh, Ha Nam, Ha Noi, Hai Duong, Hai Phong, Hung Yen, Thai Binh, Nam Dinh, and Vinh Phuc. Red River Delta located in the western coast of the Guft of Bac Bo (Tonkin), East Sea, Vietnam.

a. The Quaternary stratigraphy

The Quaternary sediment – the top deposition of Cainozoic, which unconformly overlies the Neogene deposits, are composed mainly of sands and gravels with lenses of silt and clay. The sediment thicken seaward with a maximum thickness about 200 m under the coastal area of the delta (Mathers and Zalasiewicz 1999). The Song Hong valley formed during the LGM and located southwest of the present Song Hong channel (Tanabe et al. 2006). The narrow, elongate valleys is oriented NW-SE, with approximately 20 km wide and more than 80 m deep. Those valleys is filled by gravel, sand, and clay which made during lowstand, transgressive, and high stand deposits, respectively. While, a large proportion of the surrounding sediments are massive clay intercalated with peaty organic layers which formed in highstand deposits (Lam 2003).

The Quaternary Period, comprising the Holocene and Pleistocene Epochs. The Pleistocene is divided into three period, consisting of Lechi (Q11 lc), Hanoi (Q12-3.1 hn) and Vinhphuc (Q13.2 vp) formations. The Holocene is divided into two period: Haihung (Q21-2 hh) and Thaibinh (Q23 tb) formations (Tran and Nguyen 1991).

The first sequence that formed in early Pleistocene and corresponded with Le Chi formation. This formation is located west and northwest of the RRD and consisted mainly of coarse sediment deposited pebble (quartz, silica, marble), grave, sand, silt, brown-gray clay… Distribution depth of this formation from 45-50 m to 65-70 m in the edges and the thickness varies from 5-10 m to 20-25 m and their origin were alluvial, proluvial. The sediment divides this formation into three layers: (1) the bottom layer includes cobble, pebble, gravel, coarse sand; (2) the middle layer includes predominantly middle-fine sand and silty sand; (3) the upper layer with mainly of fine sediment includes clay, silty clay, mixed with a little fine-grained sand that is light-gray, grayish-yellow (Lam 2003).

The Ha Noi formation aged in middle-late Pleistocene and formed from the fluvial and diluvial sediment. This formation distributed in edges of mounds, hills and plains of Ba Vi, Soc Son (Hanoi). The sediment divides this formation into two layers: (1) the under layer with mainly of coarse sediment deposited, their component from cobble, pebble, gravel, coarse sand in edges to coarse-middle sand in center part. The sediment is characterized by poor sorting and low roundness; (2) the upper layer with mainly of fine sediment includes silty sand and sandy silt (Lam 2003). The Ha Noi formation lies unconformable upon the Lechi formation.

The Vinh Phuc formation was formed in late Pleistocene and the sedimentary origin were fluvial, fluviolacustrine, fluviomarine. This formation with mainly of sand mixed with a little pebble, gravel in the bottom, silt sand, silty clay. The laterized sedimentary surface were mottled yellow-gray and brown-red. This formation has paleontological complex and pollen which characteristic of brackish, brackish-water environments (coastal estuaries areas). The sedimentary fluvio-marine origin of the Vinh Phuc formation only found them in drill hole at the coastal areas of Ninh Binh, Nam Dinh, Thai Binh provinces with mainly of silty clay that is blue-gray and green-gray (Lam 2003).

The boundary between the last two geological epochs, the Pleistocene and the Holocene, is placed at the date 10,000 BP. A large portion of the laterized sedimentary surface of Vinh Phuc formation was yellow-gray and brown-red with some iron coatings cause by weathered. This formation is covered by sedimentary of Hai Hung formation. The widespread of this surface has remarked and accepted as the Pleistocene-Holocene boundary. After the last glacial maximum and lateglacial sediments, Holocene sediments consist of two formations named Hai Hung and Thai Binh (Lieu 2006).

The origin of Hai Hung formation was fluvial, fluviolacustrine, coastal swamp, and marine facies types. According to the result of which suggested Hai Hung formation was formed in early-middle Holocene. The thickness of this formation varied from 2-5 m in edges to 15-20 m in centre of plain and reached to 30-35 m in coastal (Lam 2003).

Thai Binh formation was formed after the marine regression period (3000 years BP) and lies upon Hai Hung formation. The thickness of this formation varied from 1-2 m in the edges to 15-20 m in coastal (Lam 2003). Thai Binh formation is subdivided into two parts. Their composition was mainly of fine-grained sand and silty clay and the origin was bog lake, fluviomarine and marine sediment which was formed in late Holocene.

b. Natural features in Northern Vietnam

Northern Vietnam is located in a transition zone between tropical and subtropical ecosystems. The study area is a complex geological environment with the contact granite and limestone, high mountains and deltas, rugged terrain peaks and low plains, tropical and subtropical species. Because this area is located in interaction zone between tropical and subtropical with biological influences of three biogeographic regions (Indochina, South China and East Sea) (Sterling, Hurley and Le 2008). The north of Vietnam has a vast coastal zone running along the northwest side of the Gulf Tonkin (East Sea). The area study is characterized by lower topography in the northwest-southeast direction which could have a significant effect on the flow of the biggest rivers. The northern mountainous area consists of Dong Trieu, Bac Son, Ngan Son, Gam River and they are arc-shape mountain ranges which are gradually low toward the sea. Below these mountain ranges are the vast Red River Delta. Transition region from Northeast Mountain to Red River Delta, from Vinh Phu to Quang Ninh is hills with rounded peaks and slopes (altitude 200 – 300 m above sea level). They are the midland area of Northern Vietnam.

Northern Vietnam is outstanding characterized by limestone terrain, consisting of field on limestone (naturally eroded limestone), Karst valley, cave and underground rivers and streams.

c. Sea level evolution during Holocene

The previously published data showed clearly the sea-level curve for the Song Hong delta region over the past 20.000 kyr (Fig) (Geyh, Streif and Kudrass 1979, Hanebuth, Stattegger and Grootes 2000, Tran and Ngo 2000, Doan and Boyd 2001). During the LGM, the sea level was about 120 m below the present level. It reached approximately 50, 30, 15, and 5 m below the present level at about 11, 10, 9, and 8 cal. kyr BP, respectively. The Holocene sea-level rise began to decelerate between 10 and 9 cal. kyr BP. The sea reached its present level between 8 and 7 cal. kyr BP. After attaining a high of 2-3 m above the present level at 6-4 cal. kyr BP, sea level fell, at first rapidly and later gradually, reaching the present level from 4 to 0 cal. kyr BP. Tanabe el al (2003) divided the Holocene sea-level into three phases. In phase I (9-6 cal. kyr BP), sea level rose from 15 m below the present level to 3 m above the 15 present level at a rate of 6 mm/ year. In phase II (6-4 cal. kyr BP), sea level was stable. During Phase III (4-0 cal. kyr BP), sea level dropped from 3 m above the present level to the present level at an average rate of 0.67-0.1 mm/year (Tanabe et al. 2003).

Figure 1.2. Sea-level curve for the western margin of the East Sea during the past 20 kyr (Tanabe et al. 2003).

1.4.2. Climate variability in the Northern Vietnam

Vietnam has a tropical monsoon climate, with humidity averaging 84 percent throughout the year. The Northern Vietnam is situated in tropical climate, but is influenced by northeast monsoon and differentiation in climate. Thus, this climatic region is characterized by instability in seasons and temperature. The Northern Vietnam has four distinct seasons of spring, summer, autumn, and winter. The average rainfall is ranging from 1,127 mm (Nam Dinh) to 4,802 mm (Bac Quang, Ha Giang province). The weather in summer (from May to October) is hot, humid, and rainy until the presence of monsoons. The average of temperature varied from 27-29 °C and the highest is from 31-33 °C. In the winter, the temperature falls, especially in December and January. The Vietnam weather in the northeast is colder than other. The average of temperature varied from 16.3 °C to 20.9 °C and the lowest is from 14.4 °C to 19 °C.

Table 1.1. Climate stations in the North of Vietnam ((Sterling et al. 2008))

According to the results of palynological research on two cores (VN and GA cores) from the Song Hong (Red River) delta in the sub-tropical zone of Asia. Li et al. (2006) investigated vegetation and climate changes during the last 10,000 year, particularly centennial- to millennial-scale climate changes after 5340 cal. year BP, which are correlative with oceanic changes recorded in the East China and South China seas.

1.5. Objective of this study

Reconstructing and characterizing the environmental changes in Holocene in the North of Vietnam, supply base science for assessment and prediction climate change and environment, to enhance the effectiveness of programs to respond and adapt to climate change and extreme weather phenomena are increasingly complex movements. In order to reach those goals the following tasks were established:

– Taking cores covering the entire Holocene and are located in Northern Vietnam with two case study in Red River Delta and Ao Tien Lake.

– Calculating of textural parameters (Md, So, Sk, KG), describing sediment grain size distributions which used to reconstruct the depositional environment of sediments. Correlation between size parameters and transport processes/depositional mechanisms of sediments.

– Clay minerals association (during X-ray diffraction analyses) in the sediment core were investigated and applied to connect with the weather process in the past during Holocene monsoon climate in the RRD, Vietnam.

– Reconstructions of Paleoenvironmental and Paleoclimate change by using stable isotopes analysis.

– The overall goal of the investigation is to understand processes controlling sedimentation and identificate of possible climatic changes in the RRD and Ao Tien Lake which reflected by the lithology and mineralogy during Holocene.

Results are presented in two part. The first, dealing with the lithology, geochemical, and clay mineral content of Holocene sediments in two case study in Northern Vietnam (Red River Delta and Ao Tien Lake), is a comprehensive description of the cores from the study area. The second part, reconstruction of Paleoenvironment and Paleocliamte in the Northern of Vietnam as inferred from sedimentalogical and geochemical data during Honocene.

. STUDY AREAS LOCATION AND STATE OF THE ART

2.1. Red River Delta (Vietnam)

Spanning some 150 km in width, the Red River Delta (RRD) is located in the western coastal zone of the Gulf of Tonkin. The RRD with about 120 km long and 140 km wide is the fourth-largest delta in Southeast Asia, after the Mekong, Irrawaddy, and Chao Phraya deltas (Chan et al. 2012). Its catchment covers parts of China and Vietnam and its water and sediment discharge greatly influence the hydrology in the Gulf of Tonkin (Fig).

Figure 2.1. Location of Red river delta, Vietnam (modified from Tanabe et al., (2006))

The RRD was formed in Red river fault systems have been minor since the late Miocene (Lee and Lawver 1994, Tanabe et al. 2003) situated in a Neogene NW-SE-trending sedimentary basin (Song Hong Basin) (Nielsen et al. 1999) filled with Neogene and Quaternary sediments to a thickness of more than 3 km (Mathers and Zalasiewicz 1999). The Quaternary sediments, which unconformable overlie the Neogene sediments, are composed mainly of sands and gravels with subordinate lenses of silt and clay, and those sediments have thicken seaward to a maximum thickness of 200 m beneath the coastal area of the delta (Lam 2003, Tanabe et al. 2006).

From this information, this study was collected drilling cores taken from the RRD with geographical location at latitude 20°25'39.86" N, longitude 106°24'7.46" E, altitude +0.5 m (core VL) (Fig ).

2.1.1. Geographical setting

The RRD plain can be divided into three subsystems based upon surface topography and hydraulic processes, including of fluvial-dominated, tide-dominated, and wave-dominated (Mathers and Zalasiewicz 1999, Tanabe et al. 2006). The fluvial-dominated subsystem includes meandering rivers, meandering levee belts, flood plain, and fluvial terraces. It is located in the western part of the delta, where the fluvial flux is relatively stronger than others. The wave-dominated system spreads in the southwestern section of the delta, where wave energy is high due to strong summer monsoon. The system is characterized by alternating beach ridges and muddy tidal lagoon deposit. Besides, tide-dominated subsystem reaches into the northeastern part of the delta, where Hainan Island shelters the coast from strong waves. The system consisting of tidal flats, marshes, and tidal creeks/channels.

The regional climate is characterized by a tropical monsoon climate with four seasons (spring, summer, autumn, winter) and humidity averaging from 84-100 % throughout the year (Lieu 2006). While in the summer monsoon from May to October with heavy rainfall, hot and humid weather. In the dry season lasts from November to April of the following year, the climate is dry and cold. The average temperature is 29 – 30 °C the highest is 42 °C in summer (but it lasts a few days per month): May, June, July and the lowest is 9 °C (but it lasts a few days per season) between January and February. The climate in the investigation area is therefore by cool, dry winters (December until March) and warm, wet summers (May until October) characterizes (Van Maren, Hoekstra and Hoitink 2004, Pruszak et al. 2005).

The behavior of wave nearshore part is varied from direction west-southwest (from June to September) to west, northwest (from December to March of the following year). The mean tidal in this area ranges 1.9-2.6 m with the maximum height of the wave is 2-3 m (in the winter) and 5-6 m (in the summer). The tidal is characterized by there is one time of high level and one time of low level. The range of the tide at the coast is about 4 m. In the summer monsoon season, tidal influences within the delta are restricted because of the overwhelming effect of the high freshwater discharge, but in the dry season, tidal effects are evident in all of the major distributaries almost as far inland as Hanoi (Mathers et al. 1996).

According to the data given by the General Department of Meteorology and Hydrology (Vietnam), from 1884 to 1989, there were 1,993 storms and tropical depressions influenced on Vietnam territory (about 5 storms/tropical depressions per year) and 148 of which (30 %) came to the RRD (Lieu 2006). The statistics of data showed an increase in the frequency and duration of storms and typhoons during the second half of the 20th century. The increased storms and tropical depressions lead to the raising of the annual average wave height. It results in changing the geomorphology and sedimentology (erosion, accretion, shoreline, distribution of sediment) of this area (Lieu 2006).

2.1.2. Geological setting

The RRD is surrounded by mountainous areas formed of Precambrian crystalline rocks and Palaeozoic and Mesozoic sedimentary rocks and the structure is dominated by NW-SE aligned faulting trending sedimentary basin approximately 500 km long and 50-60 km wide. RRD developed overlying one trough valley which was formed by faults. The NW-SE aligned Red River fault system regulates the distribution of the mountainous areas, the drainage area, and the straight course of the Song Hong. However, fault movements have been considerably minor since the late Miocene (Pruszak et al. 2005). The trough valley was developed from early Cainozoic and filled with Neocene and Quaternary sediments with a thickness of more than 3 km and the subsidence rate of the basin is 0.04-0.12 mm/year (Mathers and Zalasiewicz 1999).

Figure 2.2. Simplified onshore Quaternary stratigraphic column of the RRD Basin
(Tran and Nguyen 1991)

The Quaternary sediment, which unconformably overlies the Neogene deposits and consists mainly of sands and gravels with lenses of silt and clay. In the RRD, the sediment is thick approximate 100 m beneath Hanoi and thickens eastwards to attain 200 m beneath parts of the coastal area (Mathers and Zalasiewicz 1999, Pruszak et al. 2005). In the coastal area of delta, the shallow water depths in the Gulf of Tonkin (< 50 m) suggest that much of the sequence is preserved in the floor (Mathers and Zalasiewicz 1999). The Quaternary depression in the RRD was mainly filled by continental deposits in five geological cycles as follows: early Pleistocene (Lechi formation), middle to late Pleistocene (Hanoi formation), late Pleistocene (Vinhphuc formation), early to middle Holocene (Haihung formation) and late Holocene (Thaibinh formation) (Nghi et al. 1991) (Fig).

The simplified onshore and nearshore Quaternary stratigraphy developed by Vietnamese workers comprises two main series: sea-level lowstand sediments (Pleistocene) and sea-level highstand sediments, the latter building the modern delta (Holocene) (Tran and Nguyen 1991).

2.1.3. Hydrology

The catchment area of the Red River (RR) is about 169.000 km2 with the annual discharge of the RR is about 137 × 109 m3 of water and 116 × 106 tons of suspended sediment, ranking among the 15 largest sedimentary discharges in the world (Pruszak et al. 2005, Milliman and Meade 1983). The Red River distributes its flow through five branches with 25 % of the flow of the Red River discharging into the sea via the Ba Lat mouth (Song Hong mouth). The water discharge in RRD region varies seasonally because most of the drainage area is under a subtropical monsoon climate regime. The averages of year precipitation in the summer is about 1,600 mm (occupy 85-95 % of the total yearly rainfall occurs). Approximately 90 % of the annual sediment discharge occurs during the summer monsoon season, when the sediment concentration may reach 12 kg/m3 (Mathers and Zalasiewicz 1999).

Table 2.1. Features of the Red River (Milliman 1995)

2.2. Ao Tien Lake, Ba Be National Park (Vietnam)

Figure 2.3. Location of Ao Tien Lake, Ba Be National Park, Vietnam

Ao Tien located within Ba Be National Park (Bac Kan province) in northeastern Vietnam (Fig. 1). Ao Tien Lake is the natural limestone mountain, formed by the collapsed of a limestone cave. The Ao Tien Lake has a surface area of 1.5 ha, with the average depth from 10-11 m, the maximum of depth is 16 m (Field work was conducted on May 2017).

In this study, sediment core was collected by a simple device which followed design by Davis and Steinman (Davis and Steinman 1998), Somrisi et al. (Somsiri et al. 2006). This device included of drill bit PVC at length of 2 m and 9 cm in diameter. The tip of drill bit was bevelled edge and other tip connected with cross hand which included of four pipe which the same at length. The PVC pipes were connected by adapter. Holes (1 cm in diameter) were drilled in a whorled pattern along the length of the 2 pipes over the drill bit with the distance between two close holes was 30 cm. These holes allowed water to drain out of the corer and water pressure reducing on sediment core in drilled bit.

A continuous sediment core (AT-01) was recovered by a rotary drill (9 cm in diameter). The core sample located at latitude 22°26'51.35" N, longitude 105°37'2.64" E, in 14 m deep water. The total core length was 110 cm and sediment recovery approached 100 %. Immediately following collection, the core sample was placed in PVC tubes and transported to the laboratory in cool condition.

In the laboratory, based on grain size, color, sediment structure and fossils, the sediment core was split and examined. The outer layer (1 cm in thickness) was removed which minimized contamination, then sliced into 54 samples at 2 cm intervals. The sediment samples were packed in labelled polyethylene bags for further analysis.

2.2.1. Geographical setting

The surrounding of Ao Tien is limestone rock system with tropical forest primeval. The regional climate is characterized by a tropical monsoon climate with two seasons. While in the summer monsoon from May to October with heavy rainfall, hot and humid weather. In the dry season last from November to April of the following year, the climate is dry and cold (Weide 2012, Ha et al. 2017). The temperature of this area varied in across a wide range between two seasons. The highest temperature in January and the lowest temperature in July. In the summer season, sediment from surrounding tended to increase because the increased rainfall and temperature changes lead to the raising of the deposition input at Ao Tien Lake. The source of water in Ao Tien Lake from water meteoric and groundwater exchanged with Ba Be Lake.

Figure 2.4. Overview of Ao Tien Lake, Ba Be National Park, Bac Kan Province, Vietnam

2.2.2. Geological setting

The surrounding Ba Be Lake is mainly karst terrain with interspersed shale and igneous rocks. Ba Be Lake has lithology diversity (Limestone dominates the regional lithology) and tectonic unique which is resulted by long, complex geological conditions. Ba Be Lake was formed from the destruction of the South-East Asia continental mass at the end of Cambria era (about 200 Ma) (Ha et al. 2017), while the surrounding the ancient limestone mountains date back more than 450 million years.

In this area has four fault systems: NW-SE, NE-SW, sub-longitudinal, sub-latitudinal. Ba Be Lake is very different in comparison with other Karst lakes in the world with a special geological structure. It is a tectonic basin with the bed sediment formed by deposition which consisted mainly of clay. Therefore, the lake will not drain as the bottom and the water kept in long term storage.

2.2.3. Hydrology

Table 2.2. Hydrological features in the Ba Be National Park

Ba Be Lake is the largest natural freshwater lake in Vietnam with four rives and streams which connected to the lakes. Cho Leng River, Ta Han and Bo Lu streams, all in the south and southwest which pour water in the lake with a total catchment area of 420 km2. Lake outflow water in the Nang River through several subterranean caves and waterfalls, then continues to flow in the Gam River.

Figure 2.6. Ecological landscape Ba Be National Park, Vietnam

In the rainy season, the water level up to 2-3 m in comparison with the previous state was noticed, cause by Gam River falls water into Ba Be Lake in this time. When the water level of Nang River decrease, water in Ba Be Lake continuous flows Nang River.

. METHODOLOGY

Numerous methods have been used to characterize the material. An overview of samples taken and methods employed is illustrated in a flow chart in Fig 3.1.

Figure 3.1. Illustrating flow chart sampling and methodological procedures in RRD

3.1. Sediment core sample

3.2. Grain size analysis

Variation of the grain size can be an important facies indicator, sedimentary development and environmental change (Folk and Ward 1957, Friedman 1979, Bui, Mazzullo and Wilding 1989, Blott and Pye 2001).

In the laboratory, all samples were first dried in an electric oven at 60 0C for 48h. The dried sample put in a mortar, and slightly crushed by a rubber pestle. Organic matter, carbonate and silica were removed using 30% H2O2, 20% Ac and 2 mol/l sodium carbonate, respectively. Sediment grain size was used laser beam diffraction, using a Particular LA-950 (Horiba) instrument at the GEO – CRE (Key Laboratory of Geoenvironment and Climate Change Response), Vietnam National University (VNU). This device can measure suspension samples liquid in the grain size range 0.01 – 3000 μm. Each sediment a sample was analysed in triplicate to yield the percentages of the related fraction of a sample with a relative error of less than 1 %. Mean, mode, sorting, skewness and other statistics were calculated by a grain size distribution and statistics program (GRADISTAT program) (Blott and Pye 2001).

Figure 3.3. Preparation sample for grain size analysis

Table 3.1. Size scale adopted in the GRADISTAT program, a modified Udden (1914), Wentworth (1922) and Friedman and Sanders (1978) (Blott and Pye 2001)

The parameters used to describe a grain size distribution include:

Mean grain size (Md) designates grain size. It lies in cumulative curves with the y-coordinate 50% (Q50).

Sorting coefficient (So) measures a total width of the particle size distribution, which is expressed most completely by the standard deviation.

Skewness (Sk) shows the degree to which a cumulative curve approaches symmetry

Kurtosis (KG) provides the degree of concentration of the grains relatives with the average.

Figure 3.4. Preparation and measurement of the grain size analyses were completed at the Key Lab (Vietnam National University – VNU)

3.2. X – ray diffraction analyses (XRD)

The sample of Holocene age in RRD were investigated for their clay mineral distribution by X-ray diffraction method. Semi-quantitative determination by peak areas with expediently estimated calculations (Biscaye 1965, Biscaye et al. 1997).

X-ray diffraction analyses were performed at the University of Bucharest, Romania. For the determination of clay minerals by X-ray diffraction was required complex sample preparation to increase the intensity of diffraction reflexes. Wet sediment sample (25g) were split and disaggregated with distilled water because dry grinding can cause changes of the phase, and in extreme cause, can lead to stress defects in the structure of crystals which will show XDR line diffracted or even structural collapses (the production of X-ray amorphous material) (Moore and Reynolds Jr 1997).

All sample were treated to remove organic matters, carbonate and silica with 10% H2O2, 20% H2COO3 and 2 mol/l sodium carbonate (Kunze and Dixon 1986, Rabenhorst, Wilding and West 1984). At the moment the effervescence ceased, the excess acid was removed by repeated washings with distilled water and centrifugations. Particles grains were separated by settling according to Stoker’s law and concentrated by centrifugations.

After treatment with sodium polyphosphate the samples were centrifuged several time for fraction separation < 2µ. The separation of dispersed suspension in 2 rotating test tubes with 5 ml. The first tube were saturated by MgCl2 and second tube were saturated by KCl and each tube was transferred to two slides by wet smearing. The smear sample were carried out also on oriented specimens including air – dried; ethylene – glycolated; heated at 330 0C and 550 0C (Jackson 2005).

The analysis was performed semi – quantitatively using a PANalytical diffractometer in the following conditions (Fig 5.2):

– Oriented sample was saturated by MgCl2 and air-dried at 20 0C;

– Oriented sample was saturated by MgCl2 and solvations with ethylene glycol and glycerol, heated to 80 0C for 8 -10 hours (vapour method);

– Oriented sample was saturated by KCl and air-dried at 20 0C;

– Oriented sample was saturated by KCl and heated to 330 ° C for 1 hour;

– Oriented sample was saturated by KCl and heated to 550 ° C for 1 hour.

The heated, solvated, saturated and air-dried sample were kept in a desiccator until they were analysed.

X-rays are electromagnetic radiation similar to light. The wavelength of X-ray and the structural spacings of crystals both have dimensions about angstroms (1Å = 10-8 cm). They are produced when electrically charged particles of sufficient energy are decelerated (Poppe et al. 2001). This observation is an example of X-ray wave interference (Roentgen diffraction), commonly known as X-ray diffraction (XRD) (Moore and Reynolds Jr 1997, Poppe et al. 2001).

If an incident X-ray beam encounters a crystal lattice, general scattering occurs. Although most scattering interferes with itself and is eliminated (destructive interference) diffraction occurs when scattering in a certain direction is in phase with scattered rays from other atomic planes. Under this condition the reflections combine to form new enhanced wave fronts that mutually reinforce each other (constructive interference). The relation by which diffraction occurs is known as the Bragg law or equation. Because each crystalline material has a characteristic atomic structure, it will diffract X-rays in a unique characteristic pattern.

Figure 3.6. Sample processing flow chart to prepare clay films and analysis clay minerals
(Moore and Reynolds Jr 1997)

This Bragg’s law is expressed in a mathematical form a equation:

nλ = 2d sinθ

where:

d – lattice interplanar spacing of the crystal;

θ – X-ray incidence angle (Bragg angle);

λ – wavelength of the characteristic X-ray;

n – range of diffraction.

According to the model of Bragg the layers of the clay minerals can be regarded as network levels. The distances between the layers are the network level distances. Continuously varying λ or θ over a range of values are done by rotating the crystal or using a powder or polycrystalline specimen give information of crystals.

Oriented specimen (grain size < 2 μm) preparation should be used for the investigation of the clay minerals. The analyses were run on a PANalytical X’Pert at the Mineralogy Laboratory. The department has a PANalytical X’Pert Pro diffractometer equipped with a Co X-ray tube and the very fast X’Celerator detector. The unit has a 15 position sample changer and a set of low background sample holders for samples as small as 1 milligram. Diffraction data is acquired by exposing powder samples to Cu-Kα X-ray radiation, which has a characteristic wavelength (λ) of 1.5418 Å. X-rays were generated from a Cu anode supplied with 40 kV and a current of 40 mA. The data was recorded with a scan speed of 1.0o min-1 and a step size of 0.01o in the 2θ range of 4-40 degree.

Mineral identification

Diffractograms were visually interpreted with the help of a computerized search via PANalytical X’Pert HighScore Plus, v3.0 program. The peak area values of clay minierals used to calculate percentages of major clay minerals which is based upon the fact that clays minerals have many crystal lattices, hence they reflect X-ray beams in different direction and. If a mineral has high amount in the smear sample, after analysis, its peak will appear with high intensity and high value of peak area. According to the semi-quantitative method, clay minerals in RRD sediment are identified and interpreted (Biscaye 1965, Starkey, Blackmon and Hauff 1984). Remember that diffraction peaks should appear for all the other d (00l) reflections within the two theta interval scanned if there is no mixed layer.

a. Illite

Illite is one of mica clay mineral series. It is structurally similar to muscovite with characterized by intense 10Å (00l) and 3.3 Å (003) peaks that remain unaltered by glycerol or ethylene glycol solvation, K-saturated, and heating to 550 0C (Fanning, Keramidas and El-Desoky 1989, Poppe et al. 2001). The value of area of peak 10 Å from air-dried sample was chosen to calculate. The degree of crystallization of illite was evaluated by illite well-ordered (10.0 Å) and illite poor ordered (10.2 Å) (Meunier 2005) (Table 3.2).

Table 3.2. The identification of clay minerals base on peak of X-ray diffraction

b. Kaolinite and chlorite

Kaolinite and chlorite were divided in to the relative intensities of their 3.58Å (002) and 3.54Å (004) peak, respectively (Biscaye 1965). Heating alone will not distinguish the dioctahedral kaolinite group minerals from Fe-rich chlorite because the 002, 003, and 004 chlorite peaks are also weakened by this heat treatment (Moore and Reynolds Jr 1997). The area of peaks in air-dried smear samples represent behaviour of kaolinite, chlorite, respectively. Kaolinite and chlorite can be differentiated by heating because Chlorite (about 14 Å) is a typical heat stability. When the sample being heated at 550 0C to destroy the kaolinite and may weaken or destroy the chlorite (7Å), the chlorite structure is left largely intact with only partial dehydration of the octahedral layer (Brindley and Gillery 1956) (Table 3.2).

c. Smectite

Smectite is characterized by the 00l reflection will swell to about 17Å when Mg-saturated with ethylene glycol (about 17.8Å with glycerol); when heated to 550 0C, the 00l reflection will collapse to about 10Å (the exchange cations) (Moore and Reynolds Jr 1997). In this study, d-value in 17Å used to identify smectite (Table 3. 2).

3.3. Stable isotopes analysis

Figure 3.7. Preparation sample for analysis

Figure 3.8. Vibration mixer with 12 tubes in the laboratory

For analysis of OM, stable carbon isotopes (δ13C), and C/N ratio, about ten grams of fresh sediment were put into a beaker and completely dried at 60 0C for 48h, and then ground to a fine powder by using an agate mortar and pestle. The visible organic matter particle (roots and small plant matter) and shell fragments were removed by a stainless steel forceps. The measurement followed the methods outlined by Tue et al. (2014). The organic matter (OM) content was obtained via loss on ignition measurement. About two grams of the pulverized sediments was first dried at 100°C in a drying oven for 2 h and then heated at 550°C in a temperature monitored muffle furnace for 5 h. OM content is calculated as the difference between the weight from before and after combustion at 550°C divided by the initial sample weight times 100%. For analysis δ13C, about 0.2 g of pulverized sediment was placed in an Eppendorf tube and treated with HCl 1N for 24h to remove carbonate at room temperature. Then, the samples were rinsed by Milli-Q filtered with distilled water and re-dried at 60 0C in an electric oven for 48h. After dried, about 10 g samples were weighed and wrapped in tin capsules. The δ13C, C/N ratio were simultaneously analyzed using an elemental analyser (EuroVector EA3000) connected to an isotope ratio mass spectrometer (Nu-Perspective Instrument) at the GEO – CRE (Key Laboratory of Geo-environment and Climate Change Response), Vietnam National University (VNU). During analysis, a certified reference material (L-alanine) was used to quantify the analytical results. The δ13C was expressed in ‰ (per mil) and following equation:

Where, δX is δ13C ro δ14N; R = 13C/12C, Rsample is the isotope ratio of the sample and Rstandard is the isotope ratio of a standard referenced to Pee Dee Belemnite (PDB) limestone carbonate. Analytical errors were 0.05% for δ13C.

Figure 3.9. Stable Isotope Mass Spectrometry (ISMS) – Nu Instruments systems
(Vietnam National University)

Figure 3.10. Gas chromatography system
(Vietnam National University)

Figure 3.11. EuroVector EA3000 system
(Vietnam National University)

. THE LITHOLOGY, GEOCHEMICAL PROXIES, CLAY MINERAL CONTENT OF SEDIMENT

4.1. Red River Delta (Vietnam)

4.1.1. Lithological characteristics

The parameters shows sediment grained size distribution, which used to reconstruct the sediment environments (McLaren and Bowles 1985). They are controlled by the direct of transport and the sedimentary processes. In this study, Md values showed the sediments were classified from the fine silt to very fine sand, with a dominance of very coarse silt and coarse silt followed geometric Folk and Ward graphic measures (Folk and Ward 1957).

Figure 4.1. Photographs of typical sediment facies of the VL core in the RRD, Vietnam

Figure 4.2. Tri-plot for textural analysis of all sediments in core sample RRD

Sediments of core samples mainly consist of sand, silt and clay, which ranged from 3.3-73.4, 22.9-87.2, and 1.5-25.5 %, respectively. Sediment facies of core sample can be classified as sandy silt, silty sand and sand (Fig. 4). Base on lithology, color and grain-size parameters, the sediment can be divided into six sections and two second sedimentary units with the following depth range: unit 1- estuarine sediments (from 36 m to 30.1 m); unit 2 – deltaic sediment (from 30.1 m to 0 m) (Fig. 6). The depositional environment during the Holocene of the RRD was resulted of the interaction between sea level rise and fluvial inputs.

Figure 4.3. Grain-size distribution of different kinds of sediment: (A), (B), (C) and (D) sediments in the core VL, RRD

According to the character (lithology) of sediment core VL, it showed a correlation with the core VN from the last study in RRD by Tanabe et al (2006). Beside, the location of core VL was also nearly with the core VN. Consequently, the geochronology of the sediment core VL was calculated based on the sediment rates which was calculated due to fifteen accelerator mass spectrometry (AMS) 14C for this area. The result showed that the last of core length correspond to dates ranging 11,260 cal. years BP {Tran, 1991 #471}. With this time, the sediment core VL was covered during the Holocene.

Figure 4.4. Sector plot showing the bivariate relationship between (a) the grain-size (µm) and sorting, (b) skewness and sorting, (c) skewness and kurtorsis

Grain-size analyses for the sediment in core VL revealed fourth types of grain-size distribution. Type A and C, which account for 31 % and 19 % of the total samples, respectively show a trimodal grain-size distribution. Type B, a large portion of the samples (46 %), has a bimodal grain-distribution with a modal size of 10-40 µm and a small amount of fine material. Type D shows bimodal grain-size distributions, with a similar modal size of the fine compound and a different modal size of the coarse compound.

According to the assumption of these statistical parameters reliability reflect differences between different deposition settings. Figure 5a shows the relationship between mean grain-size and sorting. For third deposition environments (sub- and intertidal flat, tidal flat, floodplain), there is a clustering in fine sized and very poorly sorted. Both mean size and sorting are hydraulically controlled (Rajganapathi et al. 2013), so that in all sedimentary environment the poorly sorted sediments have mean size in the medium silt fraction. Figure 5b shows the relationship between skewness and sorting. For the sub-to intertidal flat and delta front flatform, sediment are very poorly sorted and symmetrical to very fine skewed. By contrast, poorly sorted to very poorly sorted are mainly clustered around symmetrical to fine skewed and have negative skewness values. For the delta front slope and shelf prodelta sediment, poorly sorted to very poorly sorted are mainly clustered around very coarse to symmetrical skewed and have positive skewness value. In the figure 5c shows the relationship between skewness and kurtosis. For the delta front slope and shelf prodelta sediment, the positive skewness/very platykurtic to very leptokurtic range. In others deposition environments, the negative skewness/ very platykurtic to very leptokurtic range. This suggested that the dominance of silt grain-size population and the subordinate of very fine sand grain-size which gives a small proportion negative skewness.

Figure 4.5. Grain-size distribution of the core VL in RRD, Vietnam (Tue et al., 2019)

a. Core section 1: at the depth from 36 to 30.1 m

This section is characterized by reddish gray and consisted of very fine sand and clay lamination (as faint lenses 3-8 mm thick) and a large portion of this facies is bioturbated. Mud content of the sediment slightly decreases from the core bottom to the smallest value at the depth of 32.3 m (Fig. 6). The mean grain-size (Md) tended to gradually increase upward with average value was 18.39 µm. The average value of sorting (So) was 4.45±0.46 and a large portion of value is classified as very poorly sorted. The mean of skewness (Sk) ranged -0.29±0.1, it divided into very fine skewness and fine skewness sediment. Kurtosis (KG) values tended to increase upward with two high value were 1.65, 1.28 at the depth 32.3 m, 31.3 m, respectively. The KG value can be divided three, but a large portion value indicated platykurtic sediment (Fig. 5).

In the early part of the Holocene, the sea level in Vietnam was about 31 m below prevent and increased at a relatively constant rate of about 9 mm per year (Tjallingii et al. 2014). The reddish-gray color of the sediment showed erosion processes during low stands. In addition, below 30.1 m in depth, sediments consist of the reddish-gray color, which corresponds to erosion processes during low-stands. The sedimentary parameters illustrated a large portion of coarse silt and fine sand sediment which were mainly formed in high energy environments (McLaren and Bowles 1985, Rajganapathi et al. 2013).

Plant and shell fragments are scattered in core sediment and a large portion of shell fragments is concentrated from 31.6 to 32.5 m. Therefore, this section was interpreted as a sub-tidal and inter-tidal flat environment with a high frequency of tidal flooding (Tanabe et al. 2006). Sedimentation rates in this section interpreted as 0.42 cm/year (Li et al. 2006a).

Figure 4.6. Sedimentary parameters in grain-size distribution from core sample showing variations with depth in sediment core (m)

b. Core section 2: at the depth from 30.1 to 18.9 m

In the section, surface erosion formed during a transgression and separated by others section (depth core sample about 30 m) (Tanabe et al. 2006). Sediment was characterized by blue-gray bioturbated clay and consisting of laminated fine sand, silt with reddish-gray in color. Besides, sediment was bioturbated by gray clay and rich in shell fragments. Mud values in this section ranged from 69.16 to 95.48 %. Md values tended to gradually increase upward from the erosion surface to 27.5 m and decreased to the depth 18.1 m. The value of sorting (So) were divided into two types with a high portion of values were very poor sorted sediments (30-21.9 m of the depth sediment core) and other was poorly sorted sediments. Based on Sk values, sediments can be divided into 4 types, consisting of very fine skewness, fine skewness, symmetrical and coarse skewed sediments. KG values showed the stability of values which indicated for platykurtic sediments (from 30-26.5 m in the depth) and the upper layer fluctuated in a high range which indicated form very platykurtic to leptokurtic.

In the middle part of the Holocene, the sediments can be divided into two part. At the lowest part, sediment facies suggested as a ravinement surface during the transition (Tanabe et al. 2003, Tanabe et al. 2006). Between 30.1 and 27.5 m in depth sediment core showed the transgressive sand sheet was overlain the erosion surface, characteristics of Sk corresponds to lower energy environment and abundant sources of sediment inputs. Between 26.7 and 18.7 m in depth of core sediment, the Md and Sk values varied over a small range, which corresponds to low hydrodynamic energy condition. A large portion of mesokurtic and leptokurtic sediments suggested continuous addition of coarse grain-size after the winnowing and retention action of tidal currents.

Table 4.1. Sediment parameters in grain size distribution

The environment of this section was interpreted as shelf to pro-delta sediments because shell fragments in this section were decreased than lower layers and the increased mud content (Tanabe et al. 2006). The sedimentation rate in this section ranged from 0.06 to 1.14 cm/year (Li et al. 2006b).

c. Core section 3: at the depth from 18.9 to 11.7 m

This section was characterized by reddish gray or black and laminated silt clay that was rich in shell fragment and plant matter. Mud content tended to slightly decrease with ranged from 70.77 to 96.70 %. Md values between from 6.40 to 20.49 µm and tended to gradually increase upward. Sk values were gradually decreased upward and classifying it as symmetrical and coarse skewed sediment. The values of kurtosis showed relative low variation from 0.69 – 1.29, classified as platykurtic, mesokurtic and leptokurtic sediment. Between 18.9 and 11.7 m in depth sediment core, values of Md and sand content slightly increased upward which indicated higher hydrodynamic energy.

The sediment with abundant shell and plant fragments is typical of delta, therefore the environment of this section changed to delta front slope sediment (Tanabe et al. 2006). The sedimentation rate in this section ranged from 0.26 to 2.13 cm/year (Li et al. 2006b).

d. Core section 4: at the depth from 11.7 to 4.1 m

The section was characterized by gray, fine sand and clay sediment with rich of plant matter. Mud values in sediment tended to decrease and the minimum value at a depth of 8.7 m (32.4 %). Md values tended to increase and ranged between 11.06 and 56.55 µm. So, values was a small range from 2.99 to 4.97 with mainly categorized very sorted sediment. Sk value fluctuated and indicated as very fine skewed, fine skewed and symmetrical sediment. KG values fluctuated and predominantly indicating platykurtic and mesokurtic sediment.

Figure 4.7. Sedimentary columns of the core sample in RRD. (A) Lithological characteristic of sediment core; (B) Mud content in sediment core; (C) Sedimentation rates were calculated by linear interpolation between 14C ages (Tanabe et al. 2006, Li et al. 2006b)

Between 11.7 and 4.1 m in depth sediment core, the sedimentation rate markedly increased to the high value of 1.94 cm/year within the study area (Li et al. 2006b). The sediment volume supplied by the Red River during the last 2000 cal. year BP was much higher than others, correspond to an increase in the progradation of the delta system (Hori et al. 2004). The coarsening succession upward which suggested that the energy of the environment increased.

The environment of this section was interpreted as delta front platform sediment (Tanabe et al. 2006). The sedimentation rate in this section ranged from 0.82 to 1.94 cm/year (Li et al. 2006b).

e. Core section 5: from 4.1 to 0.5 m

This section consisted of gray fine sand, silt and peat lenses 1-5 cm thick. Mud content tends to decrease. Inversely, sand content tended to slightly increase upward with a maximum value at the depth of 0.5 m (36.5 %). Md values varied in across a wide range, continuously decreasing from underlying facies to the depth of 1.7 m upward. The So values tended to slightly increase from underlying facies and reached a peak at a depth of 1.7 m. In this layer, most of the sediment was categorized as very poorly sorted. Sk values continuously fluctuated widely from very fine to fine-skewed sediments. KG values varied marginally with the exception at depth 0.9 m in which there was an indication of platykurtic sediments. The increasing of clay content and a decrease in Md values suggested decreasing tidal and wave energy.

Between 4.1 and 0.5 m in depth sediment core suggested decreasing tidal and wave energy. In this section, abundant plant fragment and root and laminated clay indicated it influenced by tidal environmental (Tanabe et al. 2006). The sedimentation rate in this section interpreted 0.82 cm/year (Li et al. 2006b).

f. Core section 6: at the depth from 0.5 to the core surface

Sediments in this section consisted of reddish brown clay silt, fine sand and abundance of fine roots because it is corresponds to a lateritic weathering profile develop in the floodplain (Tanabe et al. 2006). The mean of grain-size (Md) tended to increase upward. Base on the changes in sediment parameters, sediment in this section can be classified as very poorly sorted, very fine skewed and platykurtic sediment. Sand content was form 36.5 to 64.9 % and mud content markedly dropped to the lowest levels of 35.08 % which suggested the sediments were mainly formed in high energy environments. The Md values in the surface were high because it was influenced by channel-levee sediment at the land surface of the core site.

4.1.2. Geochemical proxies (LOI, TOC, δ13C)

a. Loss-on ignition (LOI)

In the depth sediment core from 36 to 30.1 m, LOI values fluctuated with an average of 4.4±2.2 %. LOI tended to gradually decrease at the depth of from 30.1 to 26.7 m, then increased to a maximum at 21.3 m. LOI values continuously decreased in the depth sediment core between 21.3 and 4.1 m, then slightly increased again from 4.1 to surface sediment core.

b. C/N ratios

C/N ratios fluctuated slightly between 3.53 and 20.08 with a mean of 11.09 ± 3.42 (Fig. 4C). Most notably, it suddenly dropped to a value of 6.7 at the erosion surface (at the depth of 30.3 m). In the depth from 36 to 30.1, C/N rations fluctuated slightly, then it tended to increase slightly upward, with an average of 10.0 ± 2.3 in the depth between 30.1 and 17.3 m. At the depth from 17.3 to 7.7 m, C/N ratios tended to decrease slightly and reached a peak of 3.5 at 7.7 m in depth. C/N ratios fluctuated widely around a mean of 10 at the depth from 7.7 m to the core surface.

c. Total organic content (TOC)

Below the depth of 30.1 m, TOC content gradually fluctuated with an average of 1.1±0.4 % and notice a suddenly decreased at the depth of 30.3 m. In the depth sediment core from 30.3 to 26.5 m, TOC content varied over a small range, then it tended to decrease at the depth between 17.3 and 4.1 m with the exception of three points at 14.7, 7.5 and 4.9 m in depth. From 4.1 to surface sediment core, TOC values tended to increase with contained a large proportion of rootlets.

Figure 4.8. The variation of LOI (%), C/N ratios, TOC (%) and δ13C (‰) in core VL, Red River Delta with depth (m)

TOC content was positively correlated with TN content and LOI content. The regression line of TOC and TN showed very close which suggested that the inorganic nitrogen content was an insignificant contribution to the TN pool and it could be disregarded (Andrews, Greenaway and Dennis 1998, Hedges et al. 1986). Thus, the TN can be used instead of organic nitrogen for calculating C/N ratios to examine sedimentary sources and to reconstruct the paleoenvironment (Müller and Mathesius 1999).

d. Stable isotope carbon

δ13C values ranged from -29.09 to -21.95 ‰ with an average of −26.79 ± 1.44 ‰. Most notably, it suddenly increases to a value of -23.84 ‰ at the erosion surface (at the depth of 30.3 m). Below the depth of 30.5 m, δ13C values had small fluctuations from 27.02 to 29.09 ‰, then gradually increased and formed a peak of −21.9‰ at 28.9 m in depth. From this depth, δ13C values tended to decrease to the depth of 20.3 m. At the depth between 20.3 and 8.7 m, δ13C value was relatively invariant, then displayed small variation around the mean of −27.1‰ to the core surface sediment. Generally, δ13C values displayed an opposite trend with C/N ratios.

4.1.3. Clay mineral content

a. Illite

Illite contents ranged from 13.29 to 78 % with an average of 60.46±17.45 %. Below the depth of 30.1 m, illite contents fluctuated around the mean 49.78±15.90 %. Illite contents varied over a large range from 28.1 to 24.3 m, then slightly fluctuated between 28.1 and 2.9 m with a mean of 59.37±12 %. From 3 m to the core surface Illite contens tended to increased.

b. Kaolinite

Kaolinite contents ranged from 5 to 25 % with an average of 10.75±3.95 %. Below the depth of 30.1 m, kaolinite contents fluctuated around the mean 12.19±5.05 % and reach a peak to a maximum at 33.9 m. In the section between 30.1 and 21.8, kaolinite contents tended to increased and then suddenly increased at the depth of 21.6 m. Kaolinite contents continuous increased at the depth between 21.6 and 11.3 m, then slightly decreased at the depth of 2.9 m. From 2.9 to 0.9 m, kaolinite tended to increase and then drop at the depth from 0.9 to to the core surface.

c. Smectite

Figure 4.9. Variation in clay mineral (smectite, kaolinite and illite) proportion (%) in the core sediment of RRD, Vietnam

Smectite contents ranged from 1 to 49.33 % with an average of 14.24±13.34 %. Below the depth of 28.1 m, smectite contents fluctuated widely from 4 to 49.3 %, then suddenly decreased at the depth of 27.9 m., smectite contents fluctuated again at the depth from 27.9 to 14.2 m, then increase upward at the depth of 2.9 m. From 2.9 m to the core surface, smectite tended to fluctuated. In comparison, variations in the two most abundant ones (smectite and illite) have opposite trends.

4.2. Babe Lake

4.2.1. Lithological characteristics

The parameters shows sediment grained size distribution, which used to reconstruct the sediment environments (McLaren and Bowles 1985). They are controlled by the direct of transport and the sedimentary processes. In this study, Md values showed the sediments were classified from the sandy silt to silt, with a dominance of coarse silt followed geometric Folk and Ward graphic measures (Folk and Ward 1957).

Sediment of core sample were mainly composed of sand and silt, which ranged from 6.04-27.67 % and 72.33-93.96 %, respectively (Fig. 2). The clay fraction content displayed a very small (Fig. 3). The color of sediment is characterized by dark gray bands, it suggested that sediments could be formed in aerobic environment because depositional environment occurred in the lake bottom.

Figure 4.10. Triangle graph of grain-size composition shows that the sediment of AT core was mainly composed of sandy silt and silt

a. Core section 1: from 110 to 90 cm

This section was characterized by sandy silt and composed of the fine sand and coarse silt. Mud content varied in cross a wide range from 76.01 to 83.58 %. The mean sediment grain size (Md) fluctuated slightly between 23.73 and 33.63 µm with a mean of 29.12±3.21 µm. In this section, the mean sorting (So) value ranged from 2.96 to 3.42, classifying it as being very poorly sorted sediment (Fig. 3). The skewness (Sk) values were ranged between 0.3 and 0.36 with a dominance of very coarse skewness sediment.

b. Core section from 90 to 23 cm

This section was composed fine sandy, medium silt and coarse silt. Mud content slightly increased upward and reached a maximum at the depth of 51 cm (93.96 %). Md values fluctuated widely and reached a maximum at the depth of 81 cm (28.97 µm). So values were closely correlated with the Md values. Base on the variation of So values, sediment can be divided into two types, consisting of poorly sorted and very poorly sorted. Sk values fluctuated widely and were classified into coarse skewness and very coarse skewness (Fig. 3).

Figure 4.11. Sedimentary parameter in grain size distribution from sediment core showing with depth (cm)

c. Core section from 23 to 0 cm

This section was characterized by fine sandy and medium silt. Mud content markedly decreased to the depth of 13 cm (72.33 %) upward, before continuously increasing to the sediment core surface. Md values tended to increase to the depth of 13 cm (30.75), before decreasing upward. So values were also closely correlated with the Md values. In this layer most of the sediment was categorized as very poorly sorted. Sk values can be divided into two part. In the first part (from 23 to 9 cm), the Sk tended to increase with the maximum value was 0.48. While, the Sk values from 9 cm to surface sediment core tended to decreasing (Fig. 3).

6.2.2. Geochemical proxies (OM, C/N ratio, δ13C)

a. Organic matter (OM)

Organic matter is estimated from performing the loss on ignition (LOI). LOI values varied in across a wide range from 5.19-20.69 %, with an average of 12.69±4.88 %. Below the depth of 91 cm, LOI values tended to slightly increase with a maximum at the depth of 95 cm (20.69 %), then decreased at the depth of 53 cm. LOI values content fluctuated with a small range at the depth from 53 to 27 cm, although LOI values displayed a two peak at depth of 41cm, 31 cm with ranged 10 %, 10.45 %, respectively. At the depth between 27 and 15 cm, LOI values tended to increase, then continuously decreased upward (Fig. 4).

b. C/N ratios

Organic carbon content ranged from 16.67-36.55 % with an average of 26.2±6.44%. At the depth below 57 cm, OC content varied over a small range and tended to slightly increase. From 57 to 27 cm, OC content displayed higher variation, and then decreased to the surface sediment (Fig. 4).

c. Stable isotopes carbon (δ13C and δ15N)

δ13C values ranged from -28.57 to -36.8 ‰ with an average of -32.23±2.67 ‰. The δ13C values showed an inverse trend with C/N. Below the depth of 69 cm, δ13C values had small fluctuations , then suddenly dropped to a value to the depth of 53 cm. From 53 to 23 cm, δ13C values displayed small variation, before increasing to the depth of 11 cm and then decreased to the surface sediment (Fig. 4).

δ15N values slightly fluctuated from 5.33 to 7.67 ‰, with an average of 6.2±0.6 ‰ and can be divided into two part. Below the depth of 47 cm, δ15N values tended to gradually decrease, then displayed small variation around the mean of 5.86 ‰ upward (Fig. 4).

Figure 4.12. The variation of LOI (%), OC (%), C/N ratios, δ15N and δ13C in the sediment core with depth (cm)

d. Correlation between LOI and other sedimentary parameters

LOI values is positively correlated with density of sediment, C/N ratio and δ13C values follow by equations: Density (g/m3) = 68.34LOI – 37.3 (R2=0.66), C/N ratios = 0.23LOI + 10.25 (R2 = 0.69), δ13C = 0.45LOI – 37.87 (R2=0.66), respectively. Inversely, LOI values is negative correlated with OC content, the linear regression line has an equation of the form TOC = -0.84LOI + 36.83 (R2=0.4) (Fig. 5), which means the sample with high of LOI values, as well as OC content was low.

Figure 4.13. Relationship between LOI and density of sediment (a), LOI and OC content (b), LOI and C/N ratios (c), LOI and δ13C values

e. Correlation between OC content and other sedimentary parameters

OC content is positively correlated with organic nitrogen (Norg), the linear regression line has an equation of the form Norg = 0.1OC-0.5 (R2=0.94). The relationship between OC content and Norg suggested that organic carbon and total nitrogen in sediments mainly originated from organic matter. In contrast, OC content is negative correlated with C/N ratios and δ13C, follow by equations: C/N = -0.13OC+ 16.7 (R2=0.69), δ13C=-0.36OC-22.82 (R2=0.75), respectively (Fig. 6).

Figure 4.14. Relationship between OC content and Norg (a), OC content and C/N ratios (b), OC content and δ13C values

. RECONSTRUCTION OF PALEOENVIRONMENT AND PALEOCLIMATE IN THE NORTHERN OF VIETNAM AS INFERRED FROM SEDIMENTALOGICAL AND GEOCHEMICAL DATA

5.1. Recontruction of paleoenvironment and paleoclimate in the Red River Delta

5.1. Source of the organic matter

Investigating source of sediment organic matter provide a wealth of information on the sedimentation processes (Lorente et al. 2014, Leng and Marshall 2004), the paleoenvironment, palaeoclimate (Tue et al. 2011, Reotita et al. 2014, Dykoski et al. 2005), marine-terrestrial transport (Liu et al. 2016) and fluctuation of sea level (Wilson et al. 2005).

Stable isotopes (e.g., carbon stable isotope δ13C) have been largely used to investigate the provenance of sediment organic matter from the marine environment, estuary, lagoon, inter-tidal zone and delta plain .

5.2. Provenance of clay minerals in the RRD during the Holocene

Clayey sediment, weathering provenance, sediment transport and depositional processes, are all factors affected to the homogenous association of clay minerals in coastal and marine sediment (Chamley, 1989, Pandarinath, 2009, Wang & Yang, 2013). Estuarine and delta areas of large rives receive huge detrital loads from their vast drainage basic. The features of this locations are determined by a region's geology and influenced by physical, chemical, and climatic conditions. Thus, the location of the receiving basin represent an average composition of clay minerals sourced from the whole catchment (Wang & Yang, 2013). The multi sourced clayey sediments derived from the Song Hong catchment might have been well mixed in the marginal marine environments whereas the clays in the floodplain and river channel better reflect their source characteristics due to rapid transport and deposition. The presence of gibbsite-kaolinite assemblage shows high chemical weathering degree of supply source during last glacial maximum and fluvial activity of regression stage in the warm and wet periods (Li, Saito, Matsumoto, Wang, Tanabe, et al., 2006). The concentration of smectite assemblage in marine environments is controlled not only by provenance but also by particle size, and neoformation. Due to the fact that smectite remains in suspension for long time it is transported further offshore the sediment could be inherited from delta plain soil and transferred to shallow marine. The enrichment of Si in environment will slowly lead to the substitution of Al by Si and form illite-smectite mixed layed and smectite. Consequently, by transportation and segregation processes it concentrated in marine environment (Chamley, 1989). However, the replace of K+ by Na+ in structure in sea water leached illite transformed to illite-smectite mixed layed and smectite was other explaination for the neoformation of smectite (Meunier, 2005).

5.2. Reconstruction of Paleoenvironmental and Paleoclimate in Ao Tien Lake

The variation in textural parameters (Md, So, Sk) describing sediment grain-size distribution have long been used to reconstruct the depositional environment of sediments (Angusamy and Rajamanickam 2006, McLaren and Bowles 1985, Visher and Hughes 1969). Those studies suggested that the Md, So, Sk of sediment grain-size are controlled by the direction of transport and depositional processes. In this paper, the Md values showed that the sediments were categorized from fine sand to medium silt with a dominance of coarse silt. This pattern suggested the prevalence of comparatively low energy condition. The Md values reflected lake water level with time (Xiao et al. 2009) and corresponded to stronger warmer/wetter climate in surrounding lake (Yanhong et al. 2006).

Figure 7. Bi-plot of δ13C and C/N ratios for sediment core in Ao Tien Lake. According to Lamb et al. (2006), C/N ratios of organic matter originated from fresh water algea, marine phytoplankton and microalgae generally <10, while originating from terrestrial plants, fresh water plants and detrital organic matter are often >10 and the δ13C value of C3 plant vary between -33 and -23 ‰

According to Weide (2012), sedimentation rates in Ba Be lake was calculated by 14C and 137Cs dating in sediment core with the depth of 212 cm. This result shows the sedimentation rate in the bottom of the core (150-165 cm) is estimated at 0.1 cm/year, from 130.5 to 139.5 cm, the sedimentation rates increases to 0.2 cm/year and slightly increasing to 0.23 cm/year from 61-26 cm (Weide 2012). Moreover, the source of water in Ao Tien Lake exchanged with Ba Be Lake by karst system. In core AT, the Md values, sand contents, silt content had small fluctuations which suggested the source of sediment input from surrounding areas must be maintained in a stable condition/environment. In the present study, the average of sedimentation rates in Ba Be lake below the depth of 130 cm could represent for the Ao Tien lake’s sedimentation rate, which means that the core length in Ao Tien correspond to the last 700 years (from 1300 AD to present).

In the unit 1 (from 108 to 90 cm), the Md values and sand contents tended to slight decrease and inversely, mud contents tended to gradually increase. This pattern indicated that hydrological regimes are a favourable environment for fine-grain particles (Liu et al. 2008). LOI values, C/N ratios and δ13C had small fluctuations and they are the highest than other sections, indicating high lake water level and warmer climate (Yanhong et al. 2006). LOI values and C/N ratios reached a maximum at the top sediment as 20.69 %, 17.01, respectively, suggesting a high amount of organic matter input at Ao Tien Lake. In addition, δ13C values in this section was quite stable with an average of -28.9 ‰, indicating a large proportion of C3 plants in organic matter (Lamb et al. 2007). The core length in unit 1 correspond to a period from 1300 to 1424 years. The record δ18O measurements of stalagmite calcite in the southern China (Dongge Cave) showed this period of high Asian monsoon intensity with a high precipitation, a decrease of δ18O values (Dykoski et al. 2005).

In the unit 2 (from 90 to 23 cm), the sediment core can be divided into three distinguish part. In the lowest part (from 90 to 57 cm), all LOI values, C/N ratios and δ13C, δ15N tended to decrease with an average of 13.97±2.97 %, 13.92±0.85, -30.20 ±1.04 ‰ and 6.23±0.39 ‰, respectively. In addition, the parameters as Md, So also tended to decrease but it is not clearl. Sand contents suddenly decreased with an average of 15.74±2.96 %. Md values tended to slight decrease. So values were classified as poorly sorted sediment. In addition, the organic matter tended to change from high C3 plants to lake microalgae (Talbot and Lærdal 2000). C/N ratios and δ13C values tended to decrease which suggested that organic matter with a dominance of microalgae (Lamb et al. 2007). δ15N values tended to decrease suggested that a large proportion of microalgae/cyanobacteria and they can concentrated nitrogen from atmosphere for growing up, thus reducing δ15N values in the sediment. The record δ18O measurements of stalagmite calcite in the southern China (Dongge Cave) and tree-ring base hydroclimate (reconstruction the Palmer Drought Severity Index – PDSI) showed this period of weak Asian monsoon intensity with a decrease of precipitation and variation in PDSI remained low (Buckley et al. 2010, Dykoski et al. 2005). In this period, the weakening of monsoonal activity in the Northern Vietnam was supported by the percentage decrease tropical rain forest and lake water level. In the middle part (from 57 to 45 cm), LOI values, C/N ratio, δ13C values and δ15N varied over a small range and had a decreasing trend. Variation of sand content and Md values remained low. It is suggested that the enhanced microalgae biomass with a low level of water in Ao Tien Lake. The decreasing δ13C and δ15N values showed the dominance of chrysophyceae and cyanobacteria in lake microalgae. The results of, Dykoski et al.(2005), Buckley et al., (2010) also showed the decreasing precipitation in the first the part, then continuously increased upward (Buckley et al. 2010). Thus, the δ18O values of stalagmite calcite in the southern China (Dongge Cave) tended to decrease (Dykoski et al. 2005). In the last part (from 45 to 23 cm), the parameters of sand contents, Md values and So values tended to slightly increase. LOI values, δ13C values, δ15N values and C/N in this part tended to increase with an average of 8.23±2.58, -34.57±0.95 ‰, 5.97±0.13 ‰ and 12.18±0.56. This pattern suggested a high amount of sediment and organic matter into the lake from surrounding area and the increasing of organic matter derived from microalgae and C3 plants which grown up in the area surrounding. Thus, the lake water level was increased due to a high amount of water meteoric input at Ao Tien Lake. Otherwise, the enhanced Asian monsoon regime. These results were similar to those of the record δ18O measurements of stalagmite calcite in the southern China (Dongge Cave) (Dykoski et al. 2005). At the end of this part, while the LOI values increased, C/N ratio and δ13C decrease, variation of δ15N changed with time. The pattern indicated microalgae biomass developed with high lake water level. At the some point, δ13C values increased, it can occurred when 13C content of inorganic carbon dissolved in water (Leng et al. 2006). The core length in unit 2 correspond to a period from 1424 to 1864 AD.

In the unit 3 (from 23 to the core surface), the sediment core can be divided into two distinguish part. In the lower part (from 23 to 9 cm), sediment strongly increased of sand contents. Thus, LOI values tended to slightly increase than the underlying part with an average of 15.31±2.73 %, C/N ratios and δ13C also tended to gradually decrease with an average of 13.49±0.41 and -33.06±0.7 ‰. This results indicated the enhanced organic matter which derived from C3 plants grown up in the area surrounding due to higher lake water level. The comparative with the record δ18O measurements of stalagmite calcite in the southern China (Dongge Cave) and tree-ring base hydroclimate (reconstruction the PDSI) showed an increase of precipitation at the first of the part and slightly decrease at the end of this part (Buckley et al. 2010, Dykoski et al. 2005). In the upper part (from 9 to surface sediment core), sand contents, Md values and So values markedly dropped with an average of Md as 21.52±3.93 µm. Similarly, LOI values, C/N ratio, and δ15N values tended to decrease upward. δ13C values suddenly decreased and reached a minimum at 3 m in depth. This pattern indicated the decreasing an amount of water input at Ao Tien Lake. Variation of δ13C and C/N ratios remained low, suggesting the organic matter contents had derived from microalgae with a dominance of chrysophyceae (Vuorio, Meili and Sarvala 2006, Lamb et al. 2007). The core length in unit 3 correspond to a period from 1864 AD to present.

CHAPTER 6. CONCLUSIONS

.

References

Aaby, B. (1976) Cyclic climatic variations in climate over the past 5,500 yr reflected in raised bogs. Nature, 263, 281.

Aaby, B. & H. Tauber (1975) Rates of peat formation in relation to degree of humification and local environment, as shown by studies of a raised bog in Deninark. Boreas, 4, 1-17.

Andrews, J., A. Greenaway & P. Dennis (1998) Combined carbon isotope and C/N ratios as indicators of source and fate of organic matter in a poorly flushed, tropical estuary: Hunts Bay, Kingston Harbour, Jamaica. Estuarine, Coastal and Shelf Science, 46, 743-756.

Angusamy, N. & G. V. Rajamanickam (2006) Depositional environment of sediments along the southern coast of Tamil Nadu, India. Oceanologia, 48.

Battarbee, R. W. & H. A. Binney. 2009. Natural climate variability and global warming: a Holocene perspective. John Wiley & Sons.

Berger, A., F. Mesinger & D. Sijacki. 2012. Climate change: inferences from paleoclimate and regional aspects. Springer Science & Business Media.

Birks, H. J. (1989) Holocene isochrone maps and patterns of tree-spreading in the British Isles. Journal of Biogeography, 503-540.

Biscaye, P., F. Grousset, M. Revel, S. Van der Gaast, G. Zielinski, A. Vaars & G. Kukla (1997) Asian provenance of glacial dust (stage 2) in the Greenland Ice Sheet Project 2 ice core, Summit, Greenland. Journal of Geophysical Research: Oceans, 102, 26765-26781.

Biscaye, P. E. (1965) Mineralogy and sedimentation of recent deep-sea clay in the Atlantic Ocean and adjacent seas and oceans. Geological Society of America Bulletin, 76, 803-832.

Blott, S. J. & K. Pye (2001) GRADISTAT: a grain size distribution and statistics package for the analysis of unconsolidated sediments. Earth surface processes and Landforms, 26, 1237-1248.

Brindley, G. & F. Gillery (1956) X-ray identification of chlorite species. American Mineralogist: Journal of Earth and Planetary Materials, 41, 169-186.

Brøgger, W. C. 1901. Om de senglaciale og postglaciale nivåforandringer i Kristianiafeltet (Molluskfaunan). I kommission hos H. Aschehoug & Company.

Buckley, B. M., K. J. Anchukaitis, D. Penny, R. Fletcher, E. R. Cook, M. Sano, A. Wichienkeeo, T. T. Minh & T. M. Hong (2010) Climate as a contributing factor in the demise of Angkor, Cambodia. Proceedings of the National Academy of Sciences, 107, 6748-6752.

Bui, E., J. Mazzullo & L. Wilding (1989) Using quartz grain size and shape analysis to distinguish between aeolian and fluvial deposits in the Dallol Bosso of Niger (West Africa). Earth Surface Processes and Landforms, 14, 157-166.

Chan, F. K. S., G. Mitchell, O. Adekola & A. McDonald (2012) Flood risk in Asia’s urban mega-deltas: drivers, impacts and response. Environment and Urbanization ASIA, 3, 41-61.

Ciais, P., J. Jouzel, J. Petit, V. Lipenkov & J. White (1994) Holocene temperature variations inferred from Antarctic ice cores. Annals of Glaciology, 20, 427-436.

Conway, H., B. Hall, G. Denton, A. Gades & E. Waddington (1999) Past and future grounding-line retreat of the West Antarctic Ice Sheet. Science, 286, 280-283.

Crowley, T. J. (2000) Causes of climate change over the past 1000 years. Science, 289, 270-277.

Dansgaard, W., S. Johnsen, H. Clausen, D. Dahl-Jensen, N. Gundestrup, C. Hammer, C. Hvidberg, J. Steffensen, A. Sveinbjörnsdottir & J. Jouzel (1993) Evidence for general instability of past climate from a 250-kyr ice-core record. Nature, 364, 218-220.

Davis, M. 1976. Pleistocene biogeog raphy of temperate deciduous forests. Geosci. Man.

Davis, W. P. & A. D. Steinman (1998) A lightweight, inexpensive benthic core sampler for use in shallow water.

deMenocal, P. B., J. Ortiz, T. Guilderson, J. Adkins, M. Sarnthein, L. Baker & M. Yarusinsky (2000) Abrupt onset and termination of the African Humid Period:: rapid climate responses to gradual insolation forcing. Quaternary science reviews, 19, 347-361.

Denton, G. H. & W. Karlén (1973) Holocene climatic variations—their pattern and possible cause. Quaternary Research, 3, 155-205.

Doan, D. L. & W. E. Boyd (2001) Some facts of sea-level fluctuation during the Late Pleistocene-Holocene in HaLoneg Bay and Ninh Binh area. Journal of Sciences of the Earth, 23, 86-91.

Dykoski, C. A., R. L. Edwards, H. Cheng, D. Yuan, Y. Cai, M. Zhang, Y. Lin, J. Qing, Z. An & J. Revenaugh (2005) A high-resolution, absolute-dated Holocene and deglacial Asian monsoon record from Dongge Cave, China. Earth and Planetary Science Letters, 233, 71-86.

Enzel, Y., L. Ely, S. Mishra, R. Ramesh, R. Amit, B. Lazar, S. Rajaguru, V. Baker & A. Sandler (1999) High-resolution Holocene environmental changes in the Thar Desert, northwestern India. Science, 284, 125-128.

Fanning, D. S., V. Z. Keramidas & M. A. El-Desoky (1989) Micas. Minerals in soil environments, 551-634.

Folk, R. L. & W. C. Ward (1957) Brazos River bar [Texas]; a study in the significance of grain size parameters. Journal of Sedimentary Research, 27, 3-26.

Friedman, G. M. (1979) Differences in size distributions of populations of particles among sands of various origins: addendum to IAS Presidential Address. Sedimentology, 26, 859-862.

Funabiki, A., S. Haruyama, N. V. Quy, P. V. Hai & D. H. Thai (2007) Holocene delta plain development in the Song Hong (Red River) delta, Vietnam. Journal of Asian Earth Sciences, 30, 518-529.

Gasse, F. (2000) Hydrological changes in the African tropics since the Last Glacial Maximum. Quaternary Science Reviews, 19, 189-211.

Geyh, M., H. Streif & H.-R. Kudrass (1979) Sea-level changes during the late Pleistocene and Holocene in the Strait of Malacca. Nature, 278, 441.

Ha, N. T. T., N. T. P. Thao, K. Koike & M. T. Nhuan (2017) Selecting the Best Band Ratio to Estimate Chlorophyll-a Concentration in a Tropical Freshwater Lake Using Sentinel 2A Images from a Case Study of Lake Ba Be (Northern Vietnam). ISPRS International Journal of Geo-Information, 6, 290.

Hanebuth, T., K. Stattegger & P. M. Grootes (2000) Rapid flooding of the Sunda Shelf: a late-glacial sea-level record. Science, 288, 1033-1035.

Haug, G. H., K. A. Hughen, D. M. Sigman, L. C. Peterson & U. Röhl (2001) Southward migration of the intertropical convergence zone through the Holocene. Science, 293, 1304-1308.

Hedges, J. I., W. A. Clark, P. D. Quay, J. E. Richey, A. H. Devol & M. Santos (1986) Compositions and fluxes of particulate organic material in the Amazon River1. Limnology and Oceanography, 31, 717-738.

Hjort, C., S. Björck, Ó. Ingólfsson & P. Möller (1998) Holocene deglaciation and climate history of the northern Antarctic Peninsula region: a discussion of correlations between the Southern and Northern Hemispheres. Annals of Glaciology, 27, 110-112.

Hori, K., S. Tanabe, Y. Saito, S. Haruyama, V. Nguyen & A. Kitamura (2004) Delta initiation and Holocene sea-level change: example from the Song Hong (Red River) delta, Vietnam. Sedimentary Geology, 164, 237-249.

Hormes, A., B. U. Müller & C. Schlüchter (2001) The Alps with little ice: evidence for eight Holocene phases of reduced glacier extent in the Central Swiss Alps. The Holocene, 11, 255-265.

Ingólfsson, Ó., C. Hjort, P. A. Berkman, S. Björck, E. Colhoun, I. D. Goodwin, B. Hall, K. Hirakawa, M. Melles & P. Möller (1998) Antarctic glacial history since the Last Glacial Maximum: an overview of the record on land. Antarctic science, 10, 326-344.

Jackson, M. L. 2005. Soil chemical analysis: advanced course. UW-Madison Libraries Parallel Press.

Karlén, W., J. Fastook, K. Holmgren, M. Malmstrfm, J. Matthews, E. Odada, J. Risberg, G. Rosqvist, P. Sandgren & A. Shemesh (1999) Holocene glacier fluctuations on Mount Kenya, East Africa, between 6000 cal. years BP and the present. Ambio, 28, 409-418.

Kreutz, K. J., P. A. Mayewski, L. D. Meeker, M. S. Twickler, S. I. Whitlow & I. I. Pittalwala (1997) Bipolar changes in atmospheric circulation during the Little Ice Age. Science, 277, 1294-1296.

Kunze, G. & J. B. Dixon (1986) Pretreatment for mineralogical analysis. Methods of Soil Analysis: Part 1—Physical and Mineralogical Methods, 91-100.

Lam, D. D. (2003) The evolution of Holocene sediments in Red River Delta. PhD thesis, Vietnam National University.

Lamb, A. L., C. H. Vane, G. P. Wilson, J. G. Rees & V. L. Moss-Hayes (2007) Assessing δ13C and C/N ratios from organic material in archived cores as Holocene sea level and palaeoenvironmental indicators in the Humber Estuary, UK. Marine Geology, 244, 109-128.

Lee, T.-Y. & L. A. Lawver (1994) Cenozoic plate reconstruction of the South China Sea region. Tectonophysics, 235, 149-180.

Leng, M. J., A. L. Lamb, T. H. Heaton, J. D. Marshall, B. B. Wolfe, M. D. Jones, J. A. Holmes & C. Arrowsmith. 2006. Isotopes in lake sediments. Springer.

Leng, M. J. & J. D. Marshall (2004) Palaeoclimate interpretation of stable isotope data from lake sediment archives. Quaternary Science Reviews, 23, 811-831.

Li, J., S. Liu, X. Shi, X. Feng, X. Fang, P. Cao, X. Sun, Y. Wenxing, S. Khokiattiwong & N. Kornkanitnan (2017) Distributions of clay minerals in surface sediments of the middle Bay of Bengal: Source and transport pattern. Continental Shelf Research, 145, 59-67.

Li, Z., Y. Saito, E. Matsumoto, Y. Wang, S. Haruyama & K. Hori (2006a) Palynological record of climate change during the last deglaciation from the Song Hong (Red River) delta, Vietnam. Palaeogeography, Palaeoclimatology, Palaeoecology, 235, 406-430.

Li, Z., Y. Saito, E. Matsumoto, Y. Wang, S. Tanabe & Q. Lan Vu (2006b) Climate change and human impact on the Song Hong (Red River) Delta, Vietnam, during the Holocene. Quaternary International, 144, 4-28.

Lieu, N. T. H. (2006) Holocene evolution of the central Red River Delta, Northern Vietnam lithological and mineralogial investigations. PhD thesis, Greifswald University, Germany.

Lin, Y., Y. Chen, H. Chiang, C. Shen, D. Lam & N. Dung. 2006. Asian monsoon in the past 5000 years inferred from high amplitude variation of geochemical records in speleothems from Ninh Binh, Vietnam. In AGU Fall Meeting Abstracts, 08.

Liu, X., U. Herzschuh, J. Shen, Q. Jiang & X. Xiao (2008) Holocene environmental and climatic changes inferred from Wulungu Lake in northern Xinjiang, China. Quaternary Research, 70, 412-425.

Liu, X., R. Rendle-Bühring, I. Meyer & R. Henrich (2016) Holocene shelf sedimentation patterns off equatorial East Africa constrained by climatic and sea-level changes. Sedimentary Geology, 331, 1-11.

Lorente, F. L., L. C. R. Pessenda, F. Oboh-Ikuenobe, A. A. Buso Jr, M. C. L. Cohen, K. E. B. Meyer, P. C. F. Giannini, P. E. de Oliveira, D. de Fátima Rossetti & M. A. Borotti Filho (2014) Palynofacies and stable C and N isotopes of Holocene sediments from Lake Macuco (Linhares, Espírito Santo, southeastern Brazil): depositional settings and palaeoenvironmental evolution. Palaeogeography, palaeoclimatology, palaeoecology, 415, 69-82.

Mackay, A., R. Battarbee, J. Birks & F. Oldfield. 2003. Global change in the Holocene. Arnold, Hodder Headline Group.

Maley, J. (1982) Dust, clouds, rain types, and climatic variations in tropical North Africa. Quaternary Research, 18, 1-16.

Martin, R. E., E. Leorri & P. P. McLaughlin (2007) Holocene sea level and climate change in the Black Sea: multiple marine incursions related to freshwater discharge events. Quaternary International, 167, 61-72.

Masson, V., F. Vimeux, J. Jouzel, V. Morgan, M. Delmotte, P. Ciais, C. Hammer, S. Johnsen, V. Y. Lipenkov & E. Mosley-Thompson (2000) Holocene climate variability in Antarctica based on 11 ice-core isotopic records. Quaternary Research, 54, 348-358.

Mathers, S., J. Davies, A. McDonald, J. Zalasiewicz & S. Marsh (1996) The Red River delta of Vietnam. British Geological Survey Technical Report WC/96/02.

Mathers, S. & J. Zalasiewicz (1999) Holocene sedimentary architecture of the Red River delta, Vietnam. Journal of Coastal Research, 314-325.

Mayewski, P. A., L. D. Meeker, M. S. Twickler, S. Whitlow, Q. Yang, W. B. Lyons & M. Prentice (1997) Major features and forcing of high‐latitude northern hemisphere atmospheric circulation using a 110,000‐year‐long glaciochemical series. Journal of Geophysical Research: Oceans, 102, 26345-26366.

Mayewski, P. A., E. E. Rohling, J. Curt Stager, W. Karlén, K. A. Maasch, L. David Meeker, E. A. Meyerson, F. Gasse, S. van Kreveld, K. Holmgren, J. Lee-Thorp, G. Rosqvist, F. Rack, M. Staubwasser, R. R. Schneider & E. J. Steig (2004) Holocene climate variability. Quaternary Research, 62, 243-255.

McLaren, P. & D. Bowles (1985) The effects of sediment transport on grain-size distributions. Journal of Sedimentary Research, 55, 457-470.

Meeker, L. D. & P. A. Mayewski (2002) A 1400-year high-resolution record of atmospheric circulation over the North Atlantic and Asia. The Holocene, 12, 257-266.

Meunier, A. 2005. Clays. Springer Science & Business Media.

Milliman, J. D. (1995) River discharge to the sea: a global river index (GLORI). LOICZ reports and studies, 2.

Milliman, J. D. & R. H. Meade (1983) World-wide delivery of river sediment to the oceans. The Journal of Geology, 91, 1-21.

Moe, D. (1970) The post-glacial immigration of Picea abies into Fennoscandia. Botaniska Notiser, 123, 61-6.

Moore, D. & R. Reynolds Jr. 1997. X-ray Diffraction and the Identification and Analysis of Clay Minerals. Oxford University Press, Oxford, England.

Morgan, V., C. Wookey, J. Li, T. Van Ommen, W. Skinner & Μ. Fitzpatrick (1997) Site information and initial results from deep ice drilling on Law Dome, Antarctica. Journal of Glaciology, 43, 3-10.

Müller, A. & U. Mathesius (1999) The palaeoenvironments of coastal lagoons in the southern Baltic Sea, I. The application of sedimentary Corg/N ratios as source indicators of organic matter. Palaeogeography, Palaeoclimatology, Palaeoecology, 145, 1-16.

Nghi, T., N. Q. Toan, N. D. Minh & N. Van Vuong (1991) Quaternary sedimentation of the principal deltas of Vietnam. Journal of Southeast Asian Earth Sciences, 6, 103-110.

Nielsen, L., A. Mathiesen, T. Bidstrup, O. Vejbæk, P. Dien & P. Tiem (1999) Modelling of hydrocarbon generation in the Cenozoic Song Hong Basin, Vietnam: a highly prospective basin. Journal of Asian Earth Sciences, 17, 269-294.

Nishimura, M. & N. Nishino (2003) Data collection of the archeological sites in the Red River Plain. Report on Grant-in-Aid from the Scientific Research Project of the Ministry of Education, Culture. Sports, Science and Technology of Japan (in Japanese with English abstract).

O'brien, S., P. A. Mayewski, L. D. Meeker, D. A. Meese, M. S. Twickler & S. Whitlow (1995) Complexity of Holocene climate as reconstructed from a Greenland ice core. Science, 270, 1962-1964.

Poppe, L., V. Paskevich, J. Hathaway & D. Blackwood (2001) A laboratory manual for X-ray powder diffraction. US Geological Survey Open-File Report, 1, 1-88.

Pruszak, Z., P. Van Ninh, M. Szmytkiewicz, N. M. Hung & R. Ostrowski (2005) Hydrology and morphology of two river mouth regions (temperate Vistula Delta and subtropical Red River Delta). Oceanologia, 47.

Rabenhorst, M., L. Wilding & L. West (1984) Identification of Pedogenic Carbonates using Stable Carbon Isotope and Microfabric Analyses 1. Soil Science Society of America Journal, 48, 125-132.

Rajganapathi, V., N. Jitheshkumar, M. Sundararajan, K. Bhat & S. Velusamy (2013) Grain size analysis and characterization of sedimentary environment along Thiruchendur coast, Tamilnadu, India. Arabian Journal of Geosciences, 6, 4717-4728.

Reotita, J., F. Siringan, J. Zhang & R. Azanza (2014) Paleoenvironment changes in Juag Lagoon, Philippines based on sedimentology, bulk geochemistry and stable isotopes and their implication to nutrification. Quaternary International, 333, 110-121.

Rind, D. (2002) The Sun's role in climate variations. Science, 296, 673-677.

Rohling, E., P. Mayewski, R. Abu-Zied, J. Casford & A. Hayes (2002) Holocene atmosphere-ocean interactions: records from Greenland and the Aegean Sea. Climate Dynamics, 18, 587-593.

Sano, M., C. Xu & T. Nakatsuka (2012) A 300‐year Vietnam hydroclimate and ENSO variability record reconstructed from tree ringδ18O. Journal of Geophysical Research: Atmospheres (1984–2012), 117.

Scuderi, L. A. (1993) A 2000-year tree ring record of annual temperatures in the Sierra Nevada mountains. Science, 259, 1433-1436.

Sionneau, T., V. Bout-Roumazeilles, P. Biscaye, B. Van Vliet-Lanoe & A. Bory (2008) Clay mineral distributions in and around the Mississippi River watershed and Northern Gulf of Mexico: sources and transport patterns. Quaternary Science Reviews, 27, 1740-1751.

Somsiri, T., D. T. H. Oanh, S. Chinabut, N. T. Phuong, M. Shariff, F. M. Yusoff, K. Bartie, M. Giacomini, M. Robba & S. Bertone (2006) A simple device for sampling pond sediment. Aquaculture, 258, 650-654.

Starkey, H. C., P. D. Blackmon & P. L. Hauff (1984) The routine mineralogical analysis of clay-bearing samples. Geological Survey bulletin (USA). no. 1563.

Steig, E. J., D. L. Morse, E. D. Waddington, M. Stuiver, P. M. Grootes, P. A. Mayewski, M. S. Twickler & S. I. Whitlow (2000) Wisconsinan and Holocene climate history from an ice core at Taylor Dome, western Ross Embayment, Antarctica. Geografiska Annaler: Series A, Physical Geography, 82, 213-235.

Sterling, E. J., M. M. Hurley & M. D. Le. 2008. Vietnam: a natural history. Yale University Press.

Stocker, T., D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex & P. M. Midgley. 2014. Climate change 2013: The physical science basis. Cambridge University Press Cambridge, UK, and New York.

Talbot, M. R. & T. Lærdal (2000) The Late Pleistocene-Holocene palaeolimnology of Lake Victoria, East Africa, based upon elemental and isotopic analyses of sedimentary organic matter. Journal of Paleolimnology, 23, 141-164.

Tanabe, S., K. Hori, Y. Saito, S. Haruyama, V. P. Vu & A. Kitamura (2003) Song Hong (Red River) delta evolution related to millennium-scale Holocene sea-level changes. Quaternary Science Reviews, 22, 2345-2361.

Tanabe, S., Y. Saito, Q. Lan Vu, T. J. J. Hanebuth, Q. Lan Ngo & A. Kitamura (2006) Holocene evolution of the Song Hong (Red River) delta system, northern Vietnam. Sedimentary Geology, 187, 29-61.

Tjallingii, R., K. Stattegger, P. Stocchi, Y. Saito & A. Wetzel (2014) Rapid flooding of the southern Vietnam shelf during the early to mid‐Holocene. Journal of Quaternary Science, 29, 581-588.

Tran, M. & V. D. Nguyen (1991) Groundwater resources in Hanoi area. Geological survey of Vietnam.

Tran, N. & Q. T. Ngo (2000) Development history of deposits in the Quaternary of Vietnam. In: Nguyen, T.V. (Ed.), The Weathering Crust and Quaternary Sediments in Vietnam. Department of Geology and Minerals of Vietnam, Hanoi, 177-192.

Tudhope, A. W., C. P. Chilcott, M. T. McCulloch, E. R. Cook, J. Chappell, R. M. Ellam, D. W. Lea, J. M. Lough & G. B. Shimmield (2001) Variability in the El Niño-Southern Oscillation through a glacial-interglacial cycle. Science, 291, 1511-1517.

Tue, N. T., H. Hamaoka, A. Sogabe, T. D. Quy, M. T. Nhuan & K. Omori (2011) The application of δ13C and C/N ratios as indicators of organic carbon sources and paleoenvironmental change of the mangrove ecosystem from Ba Lat Estuary, Red River, Vietnam. Environmental Earth Sciences, 64, 1475-1486.

Valdes, P. J. 2014. An introduction to climate modelling of the Holocene. In Global Change in the Holocene, 36-51. Routledge.

van Geel, B., C. J. Heusser, H. Renssen & C. J. Schuurmans (2000) Climatic change in Chile at around 2700 BP and global evidence for solar forcing: a hypothesis. The Holocene, 10, 659-664.

Van Maren, D., P. Hoekstra & A. Hoitink (2004) Tidal flow asymmetry in the diurnal regime: bed-load transport and morphologic changes around the Red River Delta. Ocean Dynamics, 54, 424-434.

Visher, F. N. & G. H. Hughes (1969) The difference between rainfall and potential evaporation in Florida.

Vuorio, K., M. Meili & J. Sarvala (2006) Taxon‐specific variation in the stable isotopic signatures (δ13C and δ15N) of lake phytoplankton. Freshwater biology, 51, 807-822.

Weide, D. M. (2012) Freshwater diatoms as a proxy for Late Holocene monsoon intensity in Lac Ba Be in the Karst Region of Northern Viet Nam. California State University

Wilson, G. P., A. L. Lamb, M. J. Leng, S. Gonzalez & D. Huddart (2005) Variability of organic δ13C and C/N in the Mersey Estuary, UK and its implications for sea-level reconstruction studies. Estuarine, Coastal and Shelf Science, 64, 685-698.

Wright, H. E. 1993. Global climates since the last glacial maximum. U of Minnesota Press.

Xiao, J., Z. Chang, B. Si, X. Qin, S. Itoh & Z. Lomtatidze (2009) Partitioning of the grain-size components of Dali Lake core sediments: evidence for lake-level changes during the Holocene. Journal of Paleolimnology, 42, 249-260.

Yanhong, W., A. Lücke, J. Zhangdong, W. Sumin, G. H. Schleser, R. W. Battarbee & X. Weilan (2006) Holocene climate development on the central Tibetan Plateau: a sedimentary record from Cuoe Lake. Palaeogeography, Palaeoclimatology, Palaeoecology, 234, 328-340.

Zielinski, G. A., P. A. Mayewski, L. D. Meeker, S. Whitlow, M. S. Twickler, M. Morrison, D. A. Meese, A. J. Gow & R. B. Alley (1994) Record of volcanism since 7000 BC from the GISP2 Greenland ice core and implications for the volcano-climate system. Science, 264, 948-952.