The Bay Of Bengal Once Touched Sikkim

See this satellite imagery of the Himalaya.  The Indian State of Sikkim occupies the region just east of Darjeeling.

The Siwaliks (green arrows) appear as a forested linear band forming the southernmost hilly terrain of the Himalaya. The hills abut against broad alluvial plains. Rivers traversing the Himalaya carrying enormous sediment load encounter a gentler gradient upon exiting the hilly terrain. A loss of stream power results in sediment being dumped in the channel, so much so, that rivers get chocked on their own sediment. As a result, channels split and bifurcate forming a braided river system. These rivers  also suddenly change course, abandoning their channel and carving out new ones. Such course changes may occur during floods or by tilting of the land by structural movements.  Over time, the deposits of these ever changing rivers coalesce to form cone shape aprons of sediments known as alluvial fans. These rivers like the Kosi and the Tista, which flow transverse to the mountain range, meet an axial river like the Ganga and the Brahmaputra flowing parallel to the mountain front. The axial river flows into the Bay of Bengal.

The Siwalik hills were once these type of alluvial fans.  Just as today, during Miocene and Pliocene times, sediment was being deposited in front of the rising Himalayan mountains. Beginning about half a million years ago or so, these ancient alluvial fans were crumpled up and uplifted to form the Siwalik ranges. Active alluvial fan formation shifted southwards to its present locus. This process continues. In a few million years, the present day alluvial fans deposited by rivers like the Kosi and the Teesta will be deformed into a newer mountain range south of the Siwaliks. The Himalaya are growing southwards.

How do we know that the Siwaliks were once alluvial fans? Geologists rely on analogy, comparing the Siwalik sediments with what is accumulating in the present day alluvial fans. They find a striking similarity. Siwaliks are made up of alternations of coarse gravel layers and finer sand and silt layers with characteristic bed orientations and structures like cross beds and rippled sand. The gravel layers are inferred to be the river channel deposits while the finer sand and silt layers are the river bank, levee and floodplain deposits. An important finding made throughout the length of the Siwalik ranges has been the paleo-current directions preserved in the rocks.  Geologists have measured the orientation of bedding and ripple marks and found out that rivers were flowing south and south east i.e. perpendicular to the mountain chain. There is no evidence of an axial river like the Ganga in these Siwalik sediments. The thinking is that such an axial river must have flowed much to the south of the region of deposition of Siwalik sediments.

And what about evidence of a delta? Where did these Miocene and Pliocene rivers meet the sea? The logical geographic place to look for a coast would be towards the east. And in fact, that evidence has come from the Siwalik sediments of West Bengal and Sikkim. In a really interesting paper published recently in Current Science, Suchana Taral, Nandini Kar and Tapan Chakraborty describe sedimentary structures and marine trace fossils from Middle Siwalik sediments exposed along the Gish River and its tributaries in the Tista Valley. Siwalik rocks in the central and western part of the Himalaya show current structures that indicate south flowing rivers. In this easterly location however, the sediments show evidence of being deposited in a wave influenced environment. Sedimentary structures like wave ripple laminations and hummocky-swaley stratification indicate deposition in wave dominated marine bay.  Paleo-current indicators like ripple marks preserved on sandstone surfaces show a south as well as north directed current. This suggests an environment influenced by tides and north directed waves. Associated sediments show indicators of different delta environments like distributary channels, delta mouth bar and delta flood plain deposits.

Apart from current direction indicators, the sediments contain plant fossils indicative of mangrove vegetation and brackish water environments. They also contain trace fossils i.e. impressions and burrows made by creatures moving and disturbing the sediment surface. Cylindrichnus, Chondrites, Rosselia, Taenidium, Skolithos, Planolites are some of trace fossils reported in this study. The assemblage of trace fossils is similar to those reported from marine settings.

All this suggests that during the time of deposition of these Middle Siwalik sediments in Late Miocene-Pliocene times, about 5-10 million years ago, a branch of the Bay of Bengal had invaded as far north as present day Sikkim. Rivers carrying sediment from the Himalaya were debouching them in a delta and a shallow marine bay. The Sikkim Middle Siwalik strata are ancient deformed delta and marine deposits.  

A paleo-geographic reconstruction of this eastern part of these Siwalik depositional environments in shown below.

 Source: Suchana Taral, Nandini Kar and Tapan Chakraborty 2017

The  upper graphic shows the reconstructed delta and marine depositional environment. The lower graphic shows the regional paleo-geography. The pin shows the environmental location of the study area. The yellow rose diagram shows the paleocurrent directions measured in the Siwalik sediments.

Interestingly, some earlier work by geologists has shown that in Late Miocene times the Brahmaputra was flowing along a much more easterly route towards the Bay of Bengal. They used sand thickness and sand/shale ratios from wells drilled in the delta and found lobate sand bodies, which they inferred were brought in by a large river flowing from a ENE source. Their interpretation is shown in the graphic to the left (Uddin A. and Lundberg N. 1998). At the time the Shillong Plateau did not exist. The river flowed into the Bay of Bengal from the Upper Assam valley and through the Sylhet depression in to the Bengal Basin. The uplift of the Shillong Plateau in Pleistocene times forced the Brahmaputra to turn west and wrap itself around the newly emerging uplands.

Since Pliocene times, the tremendous amount of sediment being delivered by Himalayan rivers, coupled with Pleistocene sea level fall, has caused a retreat of this arm of the Bay of Bengal southwards.

In the satellite image below, based on the location of the Sikkim Siwalik deposits and other work on the Bengal Basin paleogeography, I have drawn in brown the coastline as it would have existed 5-10 million years ago. The ancient drainage systems are shown in blue. South directed arrows shows the extent of the growth of the Bengal/Bangladesh alluvial plains and delta and the retreat of the sea since then to its present location.

Pretty amazing finding.

Evo Devo Musical Video

This is cool!

How do we develop from one cell to a complex multicellular creature? Tim Blais who runs A Capella Science has a musical video out explaining the genetic basis for this wondrous transformation. This is a field of study known as evolutionary developmental biology.



Book recommendation: Sean B. Carroll's Endless Forms Most Beautiful: The New Science of Evo Devo is a good introduction.

Environment Links: River Issues In India

Sharing a few interesting and informative articles I came across in the past few weeks on rivers.

Endangered Himalayan Rivers: This one is from 2012. A large number of dams are planned on the Alaknanda and Bhagirathi rivers in the state of Uttarakhand.  Parineeta Dandekar writes about the weaknesses and bias in the Environment Impact Assessment process.

Rally For Rivers Plan. Will It Help?: The Rally For Rivers campaign by the Isha Foundation is calling on creating a 1 km wide tree plantation along the river banks. This, they claim, will help rejuvenate India's dying rivers. Veena Srinivasan, Sharad Lele, Jagdish Krishnaswamy and Priyanka Jamwal with the Ashoka Trust for Research in Ecology and the Environment, Bengaluru examine their claims in detail and find them wanting.

Caution Warranted For River Linking Project: The gargantuan river linking project envisages a series of dams and canal systems to transfer water from Himalayan rain and snow fed river basins to the drier Peninsular rivers in the south. Is it worth it?

Reuter's Erroneous Reporting On The Ken-Betwa River Linking Project: Two rivers in Madhya Pradesh and Uttar Pradesh are to be linked. SANDRP clarifies that the permissions process has yet to be completed. The two states don't even have a water sharing agreement! Reuter's screwed up.

Environment Ministry Panel Reject's Uttar Pradesh's Religious Smart City Plan: I'm including this to give an example of the utter indifference to ecology and environment shown by "planners and developers". The plan is for a smart city to be built inside the Hastinapur wildlife sanctuary, along the banks of the Ganga, which would have destroyed dolphin habitat and river ecology along a 7 km stretch. How does one even come up with such ideas? Fortunately, the usually pliant Environment Ministry has balked at approving this outrageous plan.

Mapping: In Praise Of The Triangle

Jerry Brotton in his book A History Of The World In 12 Maps writes about France's National Map Project. Begun around the 1670's upon the establishment of the Academie de Sciences and the Paris Observatory and headed by the astronomer Cassini I, it first attempted to create an accurate geodetic survey of France using the latest surveying instruments. At the heart of the survey was the calculation of distances and directions using the method of triangulation. Latitude was calculated using a quadrant that measured the altitude of celestial bodies. Then, using a measuring stick, a baseline of a known length was established. A third point on the landscape was sighted. The angles between the three control points were measured. Using trigonometric tables the lengths of the remaining two sides of the triangle could be calculated.

I liked this passage:

In 1744 the survey was finally completed. Its geometers had completed an extraordinary 800 principal triangles and nineteen base lines. Cassini III had always envisaged printing regional maps as they were produced, and by 1744 the map was published in eighteen sheets. Its new map of France, on an approximately small scale of 1: 1,800,000, shows the country represented as a network of triangles, with virtually no expression of the land's physical contours,and with large areas such as the Pyrenees, the Jura and the Alps left blank. It was a geometrical skeleton,a series of points,lines and triangles following coasts, valleys and plains in connecting key locations from which observations were carried out. Over it all lay the triangle, the new immutable symbol of rational, verifiable scientific method. On Cassini III's map the triangle almost takes on its own physical reality, a sign of the triumph of the immutable laws of geometry and mathematics over the vast, messy chaos of the terrestrial world. The Babylonians and the Greeks had revered the circle, the Chinese celebrated the square, the French now showed that it was the application of the triangle that would ultimately conquer the earth.

Cassini III was the grandson of Giovanni Domenico Cassini (Cassini I). The directorship of the Paris Observatory remained in the Cassini family over four generations.

This surveying method was quickly adopted and adapted by others. The Ordnance Survey began mapping the British Isles using this method in the late 1700's.  William Lambton took the Ordnance Survey's acquired expertise and began the Great Trigonometrical Survey of India in the year 1800, a feat that took nearly 50 years to complete. John Keay's book The Great Arc details that mammoth effort.

Field Photos: Glacial Deposits Of The Darma Valley, Kumaon Himalaya

During my recent trek to the Panchachuli Glacier in the Kumaon Himalaya, I obsessed about observing changes in metamorphic grade of the Greater Himalayan Sequence on the trek route and also about finding the South Tibetan Detachment fault system. I wrote about this in an earlier post.

But there were other interesting geological observations too. The Panchachuli Glacier has left a thick record of glacial deposits. The river Dhauliganga originates from this glacier. Along this river valley, glacial deposits can be observed to a distance of at least 5 kilometers downstream of the present location of the snout of the glacier, indicating that the glacier was much more extensive in the past. Tributary glaciers flowing out of the ranges east of the Dhauliganga have also left an extensive record in the form of thick fluvio-glacial deposits. These can be observed as far south as the village of Baaling.

We heard anecdotes in village Dugtu about how this glacier was much bigger in living memory and how it has been receding rapidly in the past few decades. On one level such stories are believable because studies of Himalayan glaciers have shown that many of them have been shrinking over the past few decades (ref). This is partly due to anthropogenic global warming, but glacial response to warming may be varied due to local variations in topography, precipitation and wind conditions. Some glaciers don't show retreat while some are actually seen to be expanding. Overall though, there a substantial ice loss observed across the Himalaya. Exactly how much of that is due to recent global warming and how much, as some scientists caution, due to natural factors is still being studied. Sustained warming though will cause these glacier to shrink further over the next century.

There is also a longer geological story of glacial advance and retreat written in these deposits.

I've embedded below an annotated interactive map of the glacial deposits of the Dhauliganga river valley in the Panchachuli Glacier area. This will enable readers to zoom in and recognize the various glacial landforms present in the valley. You can also access it via this Permanent Link.



The annotations depict:

a) The dark blue lines are the snout of the glacier.
b) The light blue lines are the recent terminal moriane fields.
c) The pink lines are older lateral moraines.
d) The yellow lines are outlines of older fluvio-glacial deposits
e) Numbers 1 -12 mark the locations of glacial deposits.

I have mapped only a few representative examples of each of the feature types. Readers can use these to explore similar features scattered throughout the valley. 

Location 1: This is the snout of the glacier. It is a mass of ice and frozen mud. The river Dhauliganga emerges out of an ice cave.

Location 2: Taken from near the snout of the glacier looking downstream. Ridges of the terminal moraine can be seen in the foreground. The arrows in the background outline a ridge of an older lateral moraine. Notice how the ridge decreases in elevation downstream suggesting that the terminus of this older glacial phase in somewhere nearby downstream.

Location 3: The older lateral moraine can be clearly seen as a sharp ridge line (arrow) separated from the valley wall by a depression. 

This moraine top is a few hundred meters above the valley floor implying that the glacier was thicker in the past. When was this lateral moraine deposited? It may be at least a few hundred years old. In the Garhwal Himalaya, similar older lateral moraines close to the glacier has been dated to be several hundred years old. They have been interpreted to be a result of glacial growth and deposition during the Little Ice Age, a period of earth cooling and climate instability that lasted from around the 1300's to the mid 1800's (for more on this climatic episode, I recommend Brian Fagan's book The Little Ice Age: How Climate Made History 1300-1850).

Location 4: A view of the glacier and an older lateral moraine (arrow) on the other side of the valley.

Location 5: Further downstream are thick glacial deposits. The river has incised or cut through these sediments. As a result the deposits form flattish plateaus or terraces that hug the mountain slopes. Village Dugtu, where we stayed, has been built on top of one such glacial terrace. The arrow in the top picture points to an exposure of these glacial deposits. A close up of this deposit is seen in the bottom picture. Notice the extremely ill sorted texture. Such ill sorted sediment deposited by glaciers is called Till. Large boulders are mixed in with  gravel, pebbles and much finer sized rock flour (the light to brown colored matrix).

Location 6: Another exposure of a glacial deposit near Dugtu. Again, notice the ill sorted deposit. However, at the top is a well sorted pebbly layer. This suggests deposition in more vigorous flowing water. Glacial retreat from time to time would have resulted in the establishment of a fluvial regime and deposition in these streams. These deposits may be a few hundred to several thousand years old.

Location 7: The glacial terrace on which village Dugtu is built is seen in the lower right corner. Farther away is village Philam built on the thick fluvio-glacial deposits of a tributary glacier originating in the range east of Dugtu. At village Dugtu, the east flowing river Dhauliganga makes a sharp southerly turn. The river has cut through these deposits and the slopes of the valley are thickly forested suggesting the great antiquity of these deposits.

Location 8: A nice view of glacial deposits south of village Baun along a smaller tributary of the  Dhauliganga. Notice the waterfall!

Location 9: A walk right through these thick fluvio-glacial deposits along a forested section of the valley slope. Again, notice the ill sorted nature of the deposits. Glacier are viscous and cannot sort sedimentary particles like water or air can. The result is a jumble of boulder, gravel and rock flour.

Location 10: Another cliff made up of fluvio-glacial deposits. I'm calling the deposits east of Dugtu as fluvio-glacial, since I observed intervals which show layering. This suggest deposition in water, either in streams or in melt water lakes and ponds that form in front of glaciers.

Location 11: A thick sequence of fluvio-glacial deposits along the Dhauliganga river. If you zoom and pan the satellite image you can recognize these terraces  southwards almost up to the village of Baaling.

I did not observe such deposits south of Baaling. However, there are smaller glaciers, such as the Naagling glacier, originating in the ranges on either side of the Dhauliganga. There would be smaller deposits scattered in these tributary valleys.

I have been vague about how old these deposits could be. If we assume that the Panchachuli glacier would have attained its maximum extent in the Pleistocene during the Last Glacial Maximum about 20,000 years ago, then the deposits furthest away from the present location of the glacier would be the oldest. As the glacier recedes one should find younger and younger deposits closer to the active glacier.

A study by Dirk Scherler and colleagues in the Garhwal Himalaya found such a pattern. They studied deposits of the prominent Jaundhar Glacier and the Bandarpunch Glacier in the Tons Valley. I've posted below a map showing the interpreted ages of deposition of glacial sediments.

 Source: Scherler et. al. 2010

Notice how the oldest deposits are further away from the present location of the glaciers (eastern most extremity of the map). These oldest deposits point to the maximum extent of the glacier that was reached in the Pleistocene during the Last Glacial Maximum.  However, the decreasing ages of the deposits upstream aren't the result of a uniform recession of the glacier. Instead, they point to several glacial episodes during which the glacier advanced, then receded, and then advanced again during the Holocene. Their data shows five such episodes of glacial growth dated to approximately 16 ka (ka = thousand years ago), 11-12 ka, 8-9 ka, 5 ka and less than 1 ka.


It turns out that the climate history of the Holocene is not one of uniform warming since the end of the last glacial period. The earth has gone through several minor cooling phases during the Holocene. The well known Younger Dryas Event around 12.9 -11.7 ka is one example.  Some studies suggest cooling episodes around 8.2 ka  and around 4.2 ka . And there is the Little Ice Age during the last millennium.

Another climate dynamic is fluctuating monsoon strength through the Holocene. The authors don't favor the explanation that these periods of glacial growth were triggered by global cooling events.  They argue that glacial growth corresponds to small phases of increased monsoon strength interrupting a longer trend of decreasing monsoon strength. More moisture means more snow and glacial growth. Since the long term trend in this part of the world is one of decreasing monsoon strength, every successive phase of glacial growth was smaller than the previous, resulting in younger and younger deposits upstream. The Little Ice Age deposits (which were likely driven by global cooling and not necessarily increased precipitation) mark the last major phase of glacial growth.

How are these deposits dated? Scherler and colleagues use a technique known as cosmogenic nuclide dating. This technique is one way to date the timing of surface exposure. Glaciers carry rock debris. These form a layer below the moving ice. When the glacier recedes the rock debris is deposited as a moraine or as an erratic boulder. It is exposed to the atmosphere and starts getting bombarded by cosmic rays. Energetic cosmic ray neutrons falling on atoms of minerals like quartz results in spallation reactions. This means that the collision of neutrons is energetic enough to fragment the nucleus. Oxygen bound up with silicon in the mineral quartz gets converted to an isotope of Beryllium (10Be). The amount of nuclides generated this way is proportional to the length of exposure. By measuring the amount of 10Be and comparing it with other isotopes, an 'exposure age' is estimated. This is essentially the age of glacial recession and the deposition of glacial sediment.

Samples have to been selected carefully for this method to give a true estimate of surface exposure and deposition. Care must be taken to avoid sampling rocks that have been repeatedly buried and exposed. Rocks which show signs of being subjected to prolonged glacial erosion are selected since  erosion will remove outer shells of material that may have accumulated nuclides during an earlier period of exposure.  Debris with a polished surface or with striations and grooves generally suggest subglacial transport and prolonged glacial erosion and are preferred samples.

 The figure below taken from the same study shows the reconstructed glacial extents using exposure dates of the moraine sequences in the upper Tons Valley.

Source: Scherler et. al. 2010

Such dating of glacial deposits at other locations in the Garhwal Himalaya (ref) tell a similar story of glacial growth and decay over the Holocene. And what about the Pleistocene? Is there evidence of older glacial cycles in the Himalaya? There are many studies that have identified glacial phases during the Pleistocene as well. For example, in northwest Garhwal, the Bhagirathi Glacial Stage has been dated to 63 ka (ref). And in the Ladakh Himalaya the oldest glacial stage has been dated to 430 ka (ref). Pleistocene ice ages have impacted glacial dynamics in the Himalaya too although more work needs to be done to understand the specific mechanisms of glaciation.

Location 12: Its back to the Dhauliganga valley floor. This moraine ridge (arrows) may be the remnant of an older terminal moraine. It is located about 2 kilometers downstream of the glacier.

The Panchachuli and other glaciers in the Kumaon region to the east of the Garhwal will also have their own history of past glory and recession. How much of the retreat of the Panchachuli and other Kumaon glaciers due to recent global warming?  And what is its fate? Hopefully, someone will study them with more precision in the future.