Why the ground shakes here
A mountain range still being squeezed
India pushes north
India still moves north into Asia by about 3 cm a year at Kyrgyzstan's longitude. Satellite positioning shows nearly two thirds of that, about 2 cm a year, going into squeezing the Tien Shan: its mountains are still being built.
Shallow faults under the ranges
Under Kyrgyzstan almost every measured quake lies in the upper 40 km of the crust, along faults at the feet of the ranges and beneath the basins: Chuy, Issyk-Kul, Fergana, Naryn, At-Bashy and the Alai.
A deep zone to the south
At the southern edge of the block, beneath the Pamir in Tajikistan and China, quakes reach depths of 100–250 km. Here a slab of Asia's lithosphere is sinking into the mantle. Only the northern edge of this zone lies within the block.
Rare large quakes
Most recorded quakes are too small to be felt. A few large ones were among the most destructive in the country's history, among them Kemin in 1911, Chatkal in 1946 and Suusamyr in 1992.
What moves beneath
How the ground breaks under Kyrgyzstan
Kyrgyzstan sits in the middle of a slow collision between India and the rest of Asia. Scientists measure it with satellite positioning, read it in the scars that faults leave on the land, and listen to it with seismometers. This is what their studies have found, in plain words.
Press play to zoom into one fault under the Tien Shan and watch a quake, from the first break to the shaking at the surface.Strain builds. The squeeze bends the rock, but friction keeps the fault locked, often for centuries.The break. The fault slips at the focus, or hypocentre, here about 15 km down, and the rupture runs along it. The epicentre is the point on the surface straight above.Body waves. P-waves (dotted) push and pull the rock at about 6 km a second and arrive first. S-waves (wavy) shake it side to side at about 3.5 km a second.Surface waves. At the ground the waves roll along it, slower still. In the soft gravel and clay of the basin the shaking grows; on hard rock it is weaker.
- India's push. India keeps moving north into Asia by about 3 cm a year at this longitude.
- The Pamir front. The edge of the Pamir rides over the Alai valley by 1–1.5 cm a year.
- Thrust faults. Ranges are pushed up and over the basins beside them, along faults at their feet.
- Brittle crust. The upper part of the crust is cold and snaps. Of the well-measured quakes in Kyrgyzstan, 96 % are less than 25 km down.
- Hot lower crust. Deeper down, the rock is hot and bends slowly instead of breaking. The crust here is 50–70 km thick.
- The sinking slab. Under the Pamir a slab of Asia's own outer shell is peeling away and sinking into the mantle. Quakes inside it reach about 250 km.
- Stable ground to the north. Between the Tarim basin and the northern plains, the Tien Shan is squeezed shorter by about 2 cm a year.
A push from far away
India ran into Asia tens of millions of years ago and is still pushing north: about 3 cm a year at Kyrgyzstan's longitude. Satellite positioning shows that nearly two thirds of that push, about 2 cm a year, is spent squeezing the Tien Shan. The ground across Kyrgyzstan gets shorter about as fast as a fingernail grows.
Ischuk et al. 2013 · Zubovich et al. 2010
Young mountains
At today's rate, all the squeezing that built the Tien Shan would have taken only about 10 million years: a late chapter of the collision. The mountains are still rising, and earthquakes are how they rise, one sudden step at a time.
Abdrakhmatov et al. 1996 · Zubovich et al. 2010
Stuck, then sudden
Faults do not slide smoothly. Friction keeps them locked while the squeeze bends the rock around them, storing strain for centuries. When the strain beats the friction, the fault slips in seconds: that is the earthquake. Then the slow loading starts again. In 1911 near Kemin the faults slipped 3–4 m on average, and up to 14 m, along 155–195 km.
Arrowsmith et al. 2017
Why most quakes are shallow
Only the upper crust is cold enough to snap. Lower down, the rock is hot and creeps, the way warm wax bends. That is why the points in the block crowd into a thin layer under the surface: half of the well-measured quakes in Kyrgyzstan begin less than 15 km down. Where the lower crust is cold and dry, at the edges of the range, a few quakes start deeper.
Sloan et al. 2011 · Zhang et al. 2023 · this atlas
Two ways to break
Most faults here are thrusts: one block of rock is pushed up and over another, so ranges climb over the basins beside them, as at Kemin in 1911 and Suusamyr in 1992. Others slide sideways. The best known, the Talas-Fergana fault, cuts diagonally across the west of the country, and trenches dug across it show large earthquakes in the past few thousand years.
Arrowsmith et al. 2017 · Rust et al. 2018 · Rizza et al. 2019
Alai: where the Pamir pushes hardest
The northern edge of the Pamir rides over the Alai valley by 1–1.5 cm a year, faster than across any other single fault zone measured in the Kyrgyz Tien Shan. That is why the Alai has the most frequent strong quakes in this atlas: about one of magnitude 5 or more a year within 60 km.
Zubovich et al. 2010 · Ischuk et al. 2013 · this atlas
The deep zone
Under the Pamir, India's north-western corner presses into Asia so hard that a slab of Asia's own lithosphere, the rigid outer shell of crust and upper mantle, peels away and sinks into the hot mantle below. Quakes inside the sinking slab reach about 250 km. The deepest well-measured one in this block is 160 km down, at its southern edge.
Kufner et al. 2016 · Sippl et al. 2013 · this atlas
When several faults break together
The largest earthquakes here do not stay on one fault. In 1911 the Kemin rupture ran across 13 fault patches in six sections. In 1889 the Chilik earthquake, north of Issyk-Kul, broke several faults over at least 175 km. In 1992 the Suusamyr earthquake struck a basin thought to be quiet, on a fault whose break barely showed at the surface. A quiet fault is not a dead one.
Arrowsmith et al. 2017 · Abdrakhmatov et al. 2016 · Mikolaichuk et al. 2020
Still debated
- How fast the Talas-Fergana fault moves: dated offsets in the landscape give 2–16 mm a year, satellite positioning less than 2 mm a year. The fault may move in bursts.
- Whether the deep zone under the Pamir and Hindu Kush is one twisted slab or two separate ones.
- How large the 1889 Chilik earthquake was (estimates run from magnitude 8.0 to 8.3) and which faults broke in it.
- Whether hot mantle rising under the western Tien Shan has thinned the range's deep roots. Seismic waves slow down there, which points to warm rock.
Learn more about earthquakes in Kyrgyzstan: read the literature.
Further reading
- Abdrakhmatov K.Ye. et al. (1996). Relatively recent construction of the Tien Shan inferred from GPS measurements of present-day crustal deformation rates. Nature 384, 450–453. doi:10.1038/384450a0
- Abdrakhmatov K.E. et al. (2016). Multisegment rupture in the 11 July 1889 Chilik earthquake (Mw 8.0–8.3), Kazakh Tien Shan. Journal of Geophysical Research: Solid Earth. repository
- Arrowsmith J.R. et al. (2017). Surface rupture of the 1911 Kebin (Chon-Kemin) earthquake, Northern Tien Shan, Kyrgyzstan. Geological Society Special Publication 432, 233–253. doi:10.1144/SP432.10
- Ischuk A. et al. (2013). Kinematics of the Pamir and Hindu Kush regions from GPS geodesy. Journal of Geophysical Research: Solid Earth. record
- Kufner S.-K. et al. (2016). Deep India meets deep Asia: lithospheric indentation, delamination and break-off under Pamir and Hindu Kush. Earth and Planetary Science Letters 435, 171–184. doi:10.1016/j.epsl.2015.11.046
- Mikolaichuk A.V. et al. (2020). Seismotectonics of the Inner Tien Shan: Suusamyr basin and adjacent areas. Geodynamics & Tectonophysics 11(1), 39–52. doi:10.5800/GT-2020-11-1-0461
- Rizza M. et al. (2019). Rate of slip from multiple Quaternary dating methods and paleoseismic investigations along the Talas-Fergana Fault. Tectonics. repository
- Rust D., Korjenkov A., Tibaldi A. (2018). Geologic slip-rate determinations on the Talas-Fergana fault: mismatch with geodetic slip rate. Geophysical Research Letters 45, 3880–3888. doi:10.1002/2017GL076990
- Sippl C. et al. (2013). Geometry of the Pamir-Hindu Kush intermediate-depth earthquake zone from local seismic data. Journal of Geophysical Research: Solid Earth 118, 1438–1457.
- Sloan R.A., Jackson J.A., McKenzie D., Priestley K. (2011). Earthquake depth distributions in central Asia, and their relations with lithosphere thickness, shortening and extension. Geophysical Journal International 185, 1–29. doi:10.1111/j.1365-246X.2010.04882.x
- Zhang S. et al. (2023). Seismic velocity structure of the crust and uppermost mantle beneath the Tien Shan and its adjacent areas. Frontiers in Earth Science 11, 1215103. doi:10.3389/feart.2023.1215103
- Zubovich A.V. et al. (2010). GPS velocity field for the Tien Shan and surrounding regions. Tectonics 29. doi:10.1029/2010TC002772
A fault under the capital
How the Issyk-Ata fault lifts a mountain front, one jolt at a time
The Kyrgyz Range rises straight out of the Chuy basin at the southern edge of Bishkek. The fault between them has been dug open in trenches by the Institute of Seismology. Press play to watch the cycle the trenches record: millennia of quiet loading, then a few seconds of slip.
- gravel, loam: last 2 million years
- red clay, sandstone: 25–2 million years
- old hard rock: granite, schist
- ≈ 20–25 km: below, rock is too warm to snap
- Tien Shan pushes north · Kazakh shield slides south, beneath
Loading: the squeeze bends the rock but the fault stays locked. On one segment of the Issyk-Ata fault this lasts 2,000–5,000 years.Slip: in seconds, the range side jumps up to 2 m up and over the basin. A quake of magnitude 6.8–7.4.Jolt after jolt, the range climbs. Trenches at its foot cut through steps in the ground up to 7–8 m high.
- Two segments. The fault runs 150–160 km along the foot of the Kyrgyz Range in two segments of 65–75 km. Each could produce a quake of magnitude 7.1–7.3.
- A thrust. In the trench walls the fault dips 18–42° to the south: the range is pushed north, up and over the basin.
- Jolts, not creep. Four surface-breaking quakes of magnitude 6.8–7.4 in about the last 5,000 years, each moving the ground up to 2 m.
- Slow on average. Averaged over 14,000–21,000 years, 0.6–0.9 mm a year; on one segment, 2,000–5,000 years pass between strong quakes.
- Building on it. The Institute has long warned against building in the fault zone; construction there continues.
Abdrakhmatov & Dzhumabaeva 2014 · Dzhumabaeva 2019 · Tilek kyzy 2023 (Vestnik of the Institute of Seismology) · K. Abdrakhmatov, interview to Kaktus.media, 15 September 2026
Felt or not felt
Magnitude is the quake. Points are what you feel.
The Institute of Seismology reports each quake felt in or near Kyrgyzstan with two kinds of number, and says whether it was felt in the country: «Ощущалось» or «Не ощущалось».
Two numbers
Magnitude, and the energy class K that Kyrgyz seismologists also use, measure the energy released at the source: one number per quake. Intensity, in points (баллы) of the 12-point MSK-64 scale, measures the shaking at a given place. It is strongest above the source (I₀) and weakens with distance, so one quake has many intensities.
What the points mean
1: not felt by people. 3: lamps sway. 4: felt indoors. 5: felt indoors and outdoors. 6: furniture shifts, dishes fall, windows crack, people run outside. 7: hard to stand; brick walls crack. 8: brick buildings collapse. 9–12: the ground cracks, landslides, the land surface itself changes.
Most quakes are not felt
The Institute's network records about 9,300–9,500 quakes a year. On its website it lists only those felt in or near the country: 704 from January 2024 to 27 October 2026, of which 399 were marked as felt in Kyrgyzstan.
Same quake, different shaking
The ground under a building changes how hard it shakes. In Bishkek, the same distant quake is felt more strongly near the centre and in the north of the city than in southern micro-districts such as the 12th, because the soils differ.
The map of the strongest shaking
The Institute's 2018 map shows the shaking to expect from the strongest likely quakes. By our tracing, 9 points covers about 55 % of the country and 8 points about 35 %. More than 9 points marks zones where a quake of magnitude 7.6 or more can start: the Talas-Fergana fault, the north shore of Issyk-Kul, the Kyrgyz Range front near Bishkek and the south-east of Issyk-Kul. Switch the block above to «Shaking map».
No date and hour
No one can say when and where the next strong quake will strike to the day. The Institute makes long-term forecasts of where strong quakes are more likely, from changes in seismicity, groundwater and the magnetic field, and passes them each year to the Ministry of Emergency Situations.
seismo.kg: felt-quake list, 12-point intensity scale, intensity map (2018), forecasting methodology · K. Abdrakhmatov, Kaktus.media, 15 September 2026
From Kyrgyz seismologists
Where the quakes come from: what the Institute's journal has found
Since 2013 the Institute of Seismology of the National Academy of Sciences has published its own journal, «Вестник Института сейсмологии НАН КР», 27 issues so far. We read the articles on the causes and sources of earthquakes. Their main findings, in plain words:
Two pieces of crust meet in the north
Where the Tien Shan meets the Kazakh shield, the shield's lithosphere slides south beneath the mountains while the Tien Shan pushes north over it. That is how the Kyrgyz, Trans-Ili and Kungey ranges rose, and the Chuy, Ili and Issyk-Kul basins sank between them.
Abdrakhmatov, Kalysova, Murataliyeva 2023
Blocks, and breaks at their edges
The crust of the northern Tien Shan is a mosaic of blocks: Kyrgyz, Issyk-Kul, Trans-Ili, Chu-Kendyktas, Charyn. The blocks store stress; the faults between them release it. Most of the region's earthquakes happen along the block edges.
Abdrakhmatov & Dzhanabilova 2016 · Abdrakhmatov et al. 2023
Old scars wake up
The strongest quakes of the north, Kemin 1911 and Chilik 1889, broke along the Kemin-Chilik zone: an ancient break in the crust that is moving again. Smaller background quakes cluster in knots around old granite bodies deep in the crust.
Abdrakhmatov et al. 2022 · Ashirov & Kalysova 2021
The direction of the squeeze
From how 1,790 quakes broke between 1991 and 2020 around the Chuy and Issyk-Kul basins, the crust is squeezed horizontally, roughly from north to south, and free to move up. So most quakes there push one block up over another.
Muraliev, Dzhenaliev, Maldybaeva 2021
Thousands of years between big breaks
Trenches show that a given fault breaks the surface every 2,000–5,000 years, sometimes 10,000. Each time the ground moves 2–4 m, in quakes of magnitude 6.9–7.4; averaged over time, faults move less than 1 mm a year.
Abdrakhmatov 2021 · Abdrakhmatov, Mukambaev, Ashirov 2019
Quiet millennia, busy centuries
Strong quakes in the northern Tien Shan come in clusters: thousands of quiet years, then several within a few centuries. The last cluster began in the late 1800s: 1885, 1887, 1889, 1911, 1938.
Abdrakhmatov et al. 2013
Alai: the Pamir front
Trenches across the Pamir front in the southern Alai valley found five surface-breaking quakes in about 7,000 years. Three were of magnitude 7.4 or more and probably broke 35 km of fault or more. The front moves 4.7 ± 1.7 mm a year.
Dzhumabaeva et al. 2022
Talas-Fergana: a silent segment
On the Kyldau segment of the Talas-Fergana fault, at least two destructive quakes struck in the last 3,800 years. The segment has not broken the surface since at least 420 BC, so it is a candidate for a quake of magnitude 7 or more.
Abdrakhmatov et al. 2021
Under Issyk-Kul
The floor of Issyk-Kul is still deforming, but slowly: no epicentres of even moderate quakes lie under the lake. The region's strong quakes come from faults in the ranges around it.
Abdrakhmatov 2017
Uqturpan, 2024
The magnitude 7 quake of 22 January 2024, just beyond the Kokshaal range, broke on a fault of the North Kokshaal zone. The rock moved up and sideways at once (a thrust with a strike-slip part), under horizontal squeeze. The Institute's solution agrees with international centres.
Abdyraeva, Maldybaeva, Sabirova 2024
Learn more about earthquakes in Kyrgyzstan: read the literature.
Articles in the Institute's journal (in Russian, with Kyrgyz and English abstracts)
- Абдрахматов К.Е., Стром А.Л., Дельво Д., Хавенит Х.В., Виттори Е. (2013). Временная кластеризация сильных землетрясений Северного Тянь-Шаня. Вестник ИС НАН КР 1(1). pdf
- Абдрахматов К.Е., Джумабаева А.Б. (2014). Сегментация Иссык-Атинского разлома (Северный Тянь-Шань). Вестник ИС НАН КР 1(3): 19–23. pdf
- Абдрахматов К.Е., Джанабилова С.О. (2016). Новейшая разломно-блоковая структура Северного Тянь-Шаня и сейсмичность. Вестник ИС НАН КР 1(7). pdf
- Абдрахматов К.Е. (2017). Активные деформации дна озера Иссык-Куль. Вестник ИС НАН КР 2(10). pdf
- Джумабаева А.Б. (2019). Хронология палеоземлетрясений в зоне Иссык-Атинского разлома (Чуйская впадина, северный Тянь-Шань). Вестник ИС НАН КР 1(13): 26–43. pdf
- Абдрахматов К.Е., Мукамбаев А.С., Аширов Б.М. (2019). Оценка сейсмической опасности Тянь-Шаня на основе данных об активных разломах. Вестник ИС НАН КР 2(14): 8–15. pdf
- Абдрахматов К.Е. (2021). Современные данные о повторяемости сильных землетрясений на территории Тянь-Шаня. Вестник ИС НАН КР 1(17). pdf
- Абдрахматов К.Е., Ризза М., Уолкер Р., Кэмпбел Г., Джексон Д., Маккензи Д. (2021). Повторяемость сильных землетрясений в зоне Таласо-Ферганского разлома в позднем плейстоцене-голоцене. Вестник ИС НАН КР 1(17): 8–12. pdf
- Аширов Б.М., Калысова Ж. (2021). Новейшие структуры Северного Тянь-Шаня и фоновая сейсмичность. Вестник ИС НАН КР 2(18): 11–22. pdf
- Муралиев А.М., Дженалиев А.А., Малдыбаева М.Б. (2021). Механизмы очагов землетрясений и современное поле напряжений Бишкекского прогностического полигона. Вестник ИС НАН КР 2(18): 74–83. pdf
- Абдрахматов К.Е., Аширов Б.М., Мураталиева Ж., Джумабаева А. (2022). Сейсмотектоническая позиция сильнейших землетрясений Северного Тянь-Шаня. Вестник ИС НАН КР 1(19). pdf
- Джумабаева А.Б., Байкулов С., Абдрахматов К.Е., Патыняк М., Ландграф А., Эроусмит Р., Стрекер М. (2022). Новые данные по палеосейсмологии на территории Алайской впадины (Юго-Западный Тянь-Шань). Вестник ИС НАН КР 1(19): 30–35. pdf
- Абдрахматов К.Е., Калысова Ж.К., Мураталиева Ж.З. (2023). Новейшая структура Северного Тянь-Шаня и землетрясения. Вестник ИС НАН КР 1(21): 9–15. pdf
- Тилек кызы Г. (2023). Зона Ысык-Атинского разлома — новые площади для застройки на территории города Бишкек. Вестник ИС НАН КР 2(22): 93–99. pdf
- Абдыраева Б.С., Малдыбаева М.Б., Сабирова Г.А. (2024). Механизм очага главного толчка землетрясения 22.01.2024 г. (Mpv = 6.9). Вестник ИС НАН КР 1(23). pdf
Strong earthquakes in the block
Named earthquakes on the map
| Earthquake | Date (UTC) | Magnitude | Depth |
|---|
Magnitudes are moment magnitudes from the harmonised Central Asian catalogue or the ISC-GEM catalogue. A dash means that the depth is not well measured, so the quake is drawn as a ring on the surface.
Method
How this was made
The quakes come from four catalogues. The Harmonized Earthquake Catalogue for Central Asia (World Bank, 2023) joins Soviet-era and national records up to 2020. The International Seismological Centre (ISC) Bulletin gathers reports from seismic networks around the world, including the Kyrgyz networks, and is reviewed up to September 2024. After that come EMSC, which includes reports from the Institute of Seismology of the National Academy of Sciences of Kyrgyzstan, and the US Geological Survey. A quake listed in two catalogues within 10 seconds and 20 km is counted once.
For the position and depth of each quake we use the best-checked solution available: the ISC-EHB relocations, then the ISC-GEM catalogue, then the ISC Bulletin. Mining blasts, explosions and other events that the ISC marks as not earthquakes are removed.
Magnitudes are measured on several scales. Where the harmonised catalogue or ISC-GEM gives a moment magnitude, we use it. Otherwise we use the best magnitude the networks reported, which can differ by a few tenths, so point size is a guide, not a measurement.
The quake points are a map of recorded earthquakes, not of every earthquake, and they do not forecast earthquakes. The shaking map is the Institute of Seismology's 2018 map, traced onto the block for viewing: its zone edges are accurate to about a kilometre and carry the original's 1:1,000,000 generalisation. For siting and design, use the Institute's original and the building code. Official seismic hazard assessment and warnings remain with the responsible state agencies.
Collaborate
Work with us on Kyrgyzstan's natural hazards
KG Labs Public Foundation builds national datasets, reproducible pipelines and tests of AI and Earth-observation models for climate and hazard research in Kyrgyzstan. We are looking for partners in monitoring, data sharing and research.