Geological process

Himalayan fault lines

Print edition : May 29, 2015

The destroyed village of Barpak, at the epicentre of the quake, seen from a helicopter during a relief and rescue mission in Nepal, on May 3. Photo: DANIEL BEREHULAK/NYT

Villagers sort through rubble in search of their belongings in Barpak on May 6. Photo: DANIEL BEREHULAK/NYT

Figure 1: Colour-coded shakemap depicting the severity of shaking. The table shows the Mercalli intensity categories along with the corresponding qualitative strength of shaking, peak-ground acceleration as percentage of g (the acceleration due to gravity) and the peak velocity of ground motion in cm/s from the observed impact on structures above ground (Credit: USGS).

Figure 2: The northward drift of India from 71 million years ago to present time. Note the simultaneous counter-clockwise rotation of India. Collision of the Indian continent with Eurasia occurred at about 55 million years ago (Credit: Wikipedia sourced from

Figure 3 (left): Epicentres of greater than M4.0 earthquakes within the India-Eurasia collision zone from 1990 to date. The inset yellow border defines the boundary of Nepal (Credit: Incorporated Research Institutions for Seismology (IRIS) at 6 (right): The aftershock distribution outlines the rupture zone of the main shock. The rupture during the main shock initiated beneath the epicentre and propagated towards southeast (Credit: IRIS

Table 1: The estimated population affected by the different levels of shaking intensity on ground (Credit: USGS).

Figure 4: Picture showing the main tectonic features of the Himalayan geology (Credit: Geological Society of America Special Papers, 2007, 419, pages 135-151).

Figure 5: Dip-slip faults occur in inclined fracture interface between two rock blocks. It is called ‘thrust fault’ when the interface inclination angle is gentle and the block above the fault plane overrides the block below the fault plane. When it is steep (above 45°) it is called ‘reverse fault’.

Figure 7: Map showing the April 25 Nepal earthquake among the large and great earthquakes along the Himalayan arc. The epicentre of the present quake is located in the central seismic gap closer to the 1934 Nepal-Bihar (M8.1) earthquake (Credit: Rajendran, C.P. et al., 2015, Journal of Geophysical Research, DOI: 10.1002/2014JB011015).

Graph 1: Plot of number of aftershocks with magnitude 4.0 or more till May 1. Most of the aftershocks were in the magnitude range 4-5 and also the number dropped significantly after the first 36 hours (Credit: Nepal Seismological Centre,

The geology of the Himalayas, determined largely by the ongoing collision of the Indian tectonic plate with the Eurasian plate, makes the region particularly vulnerable to earthquakes. There is a great need for better disaster preparedness as seismologists, through analysis of available data, expect a ‘great’ quake to hit the region in the near future.

It was waiting to happen for the past several centuries though geologists and seismologists were not in a position to predict when the “great Himalayan earthquake” would actually come. What hit the landlocked Himalayan country of Nepal at high noon on April 25 was not quite the “great” one, but it was certainly a grim foreboding of what could happen in this central region of the 2,400-km-long Himalayas in the next 50-100 years. Nepal occupies nearly a third of that 2,400-km stretch in the middle.

According to the United States Geological Survey (USGS), the (moment) magnitude of the earthquake was 7.8, which is two decimal points less than what makes a “great” earthquake and is categorised only as “major”. The India Meteorological Department (IMD) evaluated its magnitude to be 7.9. The USGS, too, first placed it at 7.9 but later corrected it to 7.8. (The moment magnitude scale, denoted as M, is the currently used measure of the energy released in an earthquake and differs slightly from the older Richter magnitude but, like the latter, it is also a logarithmic scale and a unit difference in magnitude is equivalent to a factor of 32 in the energy released.) The amount of energy released in an M7.8 earthquake is about 45 Peta (1015) Joules, which is equivalent to a nuclear explosion yield of about 7-8 megatons of TNT. It must, however, be remembered that this energy is released underground and much of it is dissipated by absorption in the crust of the earth itself and only a fraction of it impacts the surface.

Though the tremor does not qualify to be classified as a “great” earthquake, its impact was devastating nevertheless; it brought with it widespread destruction and death within a radius of about 200 km around Kathmandu, the capital city, flattening villages and bringing down buildings, houses and other masonry structures, historical as well as modern-urban. It was strong enough to be felt across most parts of northern India, Bangladesh, Tibet and Pakistan. It also triggered an avalanche on Mt. Everest, killing at least 19 people at the base camp of the mountain, making it the worst tragedy related to mountain expeditions in history. It triggered another huge avalanche in Langtang valley, where 250 people were reported to be missing.

The main shock struck at 11.41 hours IST (11.56 hours Nepal ST and 06.11 hours Coordinated Universal Time UTC, which is roughly the same as the GMT). The epicentre for the tremor was placed at Barpak village in Gorkha district (for which reason the USGS has now named the 2015 Nepal earthquake as the Gorkha earthquake). It is located at about 77 km northwest of Kathmandu, with latitude-longitude of 28.147° N, 84.708° W and, according to the USGS, the depth of the earthquake origin was 15 km. Other cities close to the epicentre include Lamjung (34 km E-SE), Bharatpur (58 km N-NE) and Pokhara (73 km E).

A depth of 15 km implies a shallow earthquake, which means the shaking of the earth above would be much more, causing far more severe damage to life and property than an earthquake originating deeper inside the earth’s crust. The severity of shaking on the earth’s surface is measured on the Modified Mercalli Intensity (MMI) scale, based on observed destruction to life and property on the earth’s surface and, on a scale of I-XII, given the enormity of destruction, the Nepal Earthquake was placed at level IX (category “violent”). Nearly 5.3 million people experienced severe ground-shaking during this earthquake (Figure 1 and Table 1).

Much of the damage was concentrated in and around Kathmandu. The manner in which the city is built makes it particularly vulnerable to earthquakes. It is located in a broad valley surrounded by the Himalayas. Historically, this valley was actually the site of a lake within which river delta and lake sediment had accumulated to a thickness of about 100 metres. So, basically, the city is built on a basin over the underlying strata of soft lake sediment.

Both the shallowness of the earthquake, the nature of energy release, and the fact that Kathmandu is built on basin-formed lake sediment with structures not conforming to earthquake-resistant building codes were the chief reasons for the extensive destruction seen. According to the information on the earthquake provided by the Incorporated Research Institutions for Seismology (IRIS), a consortium of universities engaged in seismological research, sedimentary basins have a large effect on the ground motion above them. While seismic waves travel at high velocity through the rigid crystalline rock of the crust below, they slow down greatly when they enter a basin. This increases the amplitude of the seismic waves as they travel through the basin. In addition, the sharp contrast in the densities of the soft basin rocks with the surrounding material causes seismic waves to reflect and trap energy within the basin for a longer time, which extends the duration of shaking.

As of May 1, the death toll is estimated to be around 7,500, and the injured would be much higher. Nepalese Prime Minister Sushil Koirala may not be off the mark when he said that the number could reach 10,000. The rural death toll was relatively low because the quake struck during midday when most villagers were outdoors working. The economic losses are estimated to be around a third of Nepal’s GDP, which is around $20 billion. Rebuilding the economy, it is estimated, could consume at least a fourth of Nepal’s GDP and would need sustained international support over the next few years.

Colliding plates

Himalayan quakes are the result of the geological process of plate tectonics because of which the earth’s upper solid part (comprising the crust and the upper mantle that make up the lithosphere) is made up of a number of relatively rigid plates (massive slabs of rock of continental size), much like a cracked egg-shell, that are constantly moving relative to one another, sometimes colliding with or sliding past one other. Slips on faults that define the plate boundaries commonly result in earthquakes.

About 50 million years ago, the Indian plate slammed into the Eurasian plate and began diving under it (subduction) at a great velocity of about 40-50 millimetre/year in the north-northeast direction. This process is still going on and the collision zone actually wraps around the northwest projection of the Indian subcontinent in the Hindu Khush region of Tadjikistan and Afghanistan and then extends to the southeast through Nepal and Bhutan (Figure 2). The compressive stress created by this northward motion is being absorbed at the inter-plate boundary and during this process, while the Indian plate is getting squeezed and becoming smaller, the Eurasian plate is getting lifted up and a fraction (15-20 mm/yr) of that relative motion of 40-50 mm/yr is what is driving the formation of the Himalayan mountain range across the 2,400-km-long frontal arc of the plate south of Tibet in the east-west direction. In the Nepal section of this arc, this northward motion is essentially perpendicular to the Himalayas.

This accumulation of tectonic compressive stress across the inter-plate boundary gives way when it exceeds a certain limit, resulting in an earthquake, usually along existing fault planes. Much like a spring, the rocky plates of the crust cannot absorb any more of the stress by their elastic deformation (strain) and the rock blocks slip past against one another, resulting in the rupture of a part of the fault plane and the release of that stored-up energy in the form of earthquake. Because of this continuing build-up of stress, the Himalayan region is a highly earthquake-prone region which has seen many large earthquakes and consequent heavy destruction (Figure 3 shows the epicentres of greater than M4.0 earthquakes within the India-Eurasia collision zone from 1990 to date).

Four earthquakes of M6.0 or more have occurred within a 250-km radius of the April 25 epicentre during the past century. One, a M6.9 event in August 1988, 240 km to the southeast, resulted in about 1,500 deaths. The largest was the “great” Nepal-Bihar earthquake of 1934, which had a magnitude of 8.1 and was located close to the 1988 event, close to the India-Nepal border on the Nepal side, which severely damaged regions of Nepal and Bihar and killed more than 10,700 people, with over 7,000 in Bihar alone. The April 25 earthquake is the most powerful one to strike after the 1934 temblor.

There have been other major pre-20th century earthquakes in Nepal—in 1408, 1505, 1681, 1803, 1810, 1833 and 1866. But none of these seems to have been a great earthquake (a >M8.0 event). The last great earthquake in this central region of the Himalayas was perhaps the one in 1255, according to Roger Bilham of the University of Colorado. Given this huge 750-year time gap, geologists have been expecting a great earthquake to occur in this region.

Continuing aftershocks

A string of aftershocks of M4.0 and above followed the main shock, and are continuing to occur even now. Within 26 minutes of the main shock, an M5.5 aftershock was recorded by the IMD. This was soon (just 7.5 minutes later) followed by an M6.6 aftershock, which would itself be categorised as a “strong” earthquake. In fact, within 24 hours of the main tremor, there were as many as 40 aftershocks of magnitude 4.0 or larger and the 41st one was another “strong” M6.7 aftershock (measured as M6.9 by the IMD). The epicentre of the M6.6 aftershock was around the same region as the main shock, at a distance of about 70 km northwest of Kathmandu and that of the next day’s M6.7 aftershock was actually 67 km east of Kathmandu. These two strong tremors of magnitude greater than 6.0 were felt even in India and Bangladesh. These big aftershocks compounded the already widespread and severe damage inflicted by the main shock and greatly hampered the recovery efforts following the main shock. Structures already weakened by the main shock also ended up being flattened. They also probably were the cause of more avalanches that were reported near Mt. Everest.

As of May 7, there have been as many as 147 aftershocks of magnitude 4.0 or more, but after the strong one of April 26 (M6.7), the magnitude of nearly all of them were in the 4-5 range and the frequency also dropped significantly with time (Graph 1).

Aftershocks, which occur close to the main shock, are normal after a large earthquake and are expected to continue, though they will occur less often with time. In general, the larger the main shock, the larger the area over which aftershocks occur, but the exact pattern varies from quake to quake and is not predictable. However, based on past events and the aftershocks of those, forecast of aftershock sequences can be made using modelling techniques.

The USGS, based on aftershock modelling using data of past Nepal events, said on May 1 that there was a 16 per cent probability of a ≥M6.0 aftershock and an 83 per cent probability of a ≥M5.0 aftershock occurring before May end. The modelling also shows that there is a finite probability of aftershocks persisting in small numbers even up to one year. However, the USGS added that the probability of aftershocks was lower on May 1 than they had forecast before on April 27 as the number of aftershocks decreased faster than expected.

The geology of the Himalayas on its southern side is characterised by three major tectonic units: The Main Central Thrust (MCT), the Main Boundary Thrust (MBT) and the Himalayan Frontal Thrust (HFT) or the Main Frontal Thrust (MFT). These structures essentially separate the different rocky formations that characterise the different stages of the Himalayan outcrop—the higher Himalayas, the lesser Himalayas and the sub-Himalayan Shivalik range. The highest and the oldest of these is the MCT, which is a north-dipping fault and marks the tectonic contact between the higher and the lesser Himalayas. The lesser and the sub-Himalayas are separated by the MBT, and the HFT or the MFT constitutes the southern-most and the youngest thrust.

All these three faults conjoin along the basal detachment plane—called decollément in geophysical language—called the Main Himalayan Thrust (MHT). The MHT defines the thrust interface between the subducting Indian and the overriding Eurasian plates where there is a plane of detachment from the Indian lithosphere. On the far Tibetan side of the Himalayas, there is another fault system called the South Tibetan Detachment (STD). All these four thrusts form the base of the Tethyan Sedimentary Series of the Himalayas (Figure 4). In the current era, both the MBT and the HFT faults are considered more active, nucleating regions of large earthquakes compared to the MCT. It is believed that the great plate-boundary earthquakes, such as the 1934 Nepal-Bihar earthquake, had their origins in the MCT. In fact, such great earthquakes are believed to involve the tectonics of the entire fault system, resulting in the rupturing of the very thrust front itself.

According to the USGS, the April 25 earthquake was the result of “thrust faulting” on or near the MFT/MHT. (A “thrust fault” (Figure 5) occurs in the inclined fracture interface between two rock blocks with an angle of 45° or less when the block above the fault plane overrides the block below the fault plane.) Earthquakes are usually shown as points in maps but large earthquakes of the Nepal kind are actually the result of slipping over a large fault area.

According to the USGS, events of the size of the April 25 earthquake are typically about 100 km x 50 km in size and the rupture area of this M7.8 event has been determined to be about 120 km (eastwards) x 80 km (southwards), which was initiated at the hypocentre (the point beneath the epicentre) and directed south-eastwards towards Kathmandu. The aftershock distribution (Figure 6) gives an indication of the rupture zone of the main shock. Significantly, the maximum amount of slip in this thrust fault event of about 3 metres seems to have occurred in the rupture area just about 20 km north of Kathmandu. As a result of this large crustal displacement close to the city, substantial energy release was directed towards the capital city. The 1934 Nepal-Bihar and the 1833 events too had ruptured a similar sized area of the fault plane. Susan Hough, a seismologist with the USGS, has tweeted that the rupture area of the present quake seems to overlap the segment that caused the 1934 Nepal-Bihar quake.

“This slip released the equivalent of only about a century of built-up strain,” says C.P. Rajendran, an Earth Sciences expert at the Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Bangalore. According to studies by Rajendran and his colleagues, the last great earthquake in the central Himalayan region occurred sometime between the 1259 and 1433. Whether, as contended by Roger Bilham, the 1255 event too corresponds to a great earthquake, which would mean large events in quick succession, remains to be settled still. Prior to that, there is evidence to suggest the occurrence of a great earthquake between 1050 and 1250. In any case, it is true that there has been an absence of a great earthquake in the last 700 years in the central Himalayas. According to Rajendran, this translates, based on GPS measurement-based studies, the accumulated strain over this period is about 12-14 metres. So about 10 m of strain remains to be released, which makes the occurrence of large earthquakes in this region, which is home to about 10 million people, imminent.

A great quake overdue?

A great earthquake of magnitude 8 or more would actually rupture the MFT itself. But there is a prominent stretch of about 700 km in the central Himalayas between the epicentres of the 1905 Kangra earthquake (M8.0) and the 1934 Nepal-Bihar earthquake (M8.1) where the MFT has shown no sign of any rupturing for several centuries, which is referred to as “Central seismic gap”. (A seismic gap refers to a region along an active fault where stress is accumulating but no large earthquakes have occurred recently. Seismic gaps are often flanked by areas that have experienced large earthquakes in the recent past and these regions are high-risk areas for large earthquakes in the future.) The Himalayan seismic gap theory holds that this prolonged quiescence in this 700 km central seismic gap implies that a great earthquake (>M8.0) is overdue. The present Nepal quake was located right in this central Himalayan seismic gap (Figure 7).

However, in a January 2015 paper published in the journal Lithosphere, Rajendran and his co-workers, which include Australian scientists, have pointed out that despite this recognised seismic risk, the geometry of faults in this region remains poorly characterised. Their analysis of the landscape and erosion patterns in the western segment of this gap suggests the emergence of a detachment fault beneath Uttarakhand that “provides a sufficiently large and coherent fault segment capable of hosting a great earthquake”. While this could imply that the probability of the “great Himalayan earthquake” occurring within the next 50-100 years or so remains fairly high, it need not necessarily be so, points out Rajendran. “A probabilistic estimate needs to be done using all the available information on the past earthquake through statistical methods; for example, a 30 per cent chance of this occurring in the next 10 years etc.,” he said. “However,” he added, “our investigations have shown that there are several seismogenic structures in the Himalayas. The strain need not be released through only one mode like great earthquakes which would break the frontal thrust. This earthquake indicates the importance of little known structures in the Himalayas, which also probably rules out any regularity (in time) of earthquake occurrence.”

That is, according to him, the release of pent up energy in the central region, notwithstanding the huge accumulated strain, need not be necessarily through a great earthquake. Many quakes similar to the Nepal quake, and perhaps even those of lesser magnitude and events of M6-M7, could result in the release of the accumulated strain. “But there will be great earthquakes in the region also because it had happened in the past. The past is the key to the present,” he said.

Need for greater preparedness

All these seismological insights into the quake-prone Nepal region in the Himalayas point to the need for far greater disaster preparedness not only in Nepal but also in the Himalayan States of India. “Earthquakes don’t kill people; buildings do,” is the belief among seismologists. Though the exercise of improving building codes in vulnerable districts in Nepal, Kathmandu in particular given the increasing urbanisation, was taken up by the Nepal government in all seriousness, the earthquake struck before even a significant beginning could be made in the huge task.

Brian Tucker, founder and president of GeoHazards International, a non-profit organisation devoted to reducing casualties from natural disasters, was quoted in The Washington Post soon after the April 25 earthquake saying, “It seems that the rural-to-urban migration of people has resulted in really rapid construction of housing which, as far as I can see from my visits, has been unregulated and is just very, very vulnerable.”

Tucker recalled his conversation in the late 1990s with a Nepalese Minister who, referring to the 1934 Nepal-Bihar earthquake, apparently told him: “We don’t have to worry about earthquakes anymore, because we already had an earthquake.” “I took him to the window and said to him, ‘As long as you see those Himalaya mountains there, you will know that you will continue to have earthquakes’,” Tucker told The Washington Post. The April 2015 quake is a grim reminder to not allow such complacency to set in. This holds equally true for vulnerable regions of Uttarakhand and other Himalayan regions of India.

This article is closed for comments.
Please Email the Editor