Somewhere beneath the streets of Kathmandu Valley, a hidden groundwater system is depleting completely unnoticed. The only visible warning sign? The ground has quietly begun to sink, collapsing by up to 21 centimeters a year in some places (Rawal & Wang, 2025).
Beneath Kathmandu’s urban surface lies a complex layered aquifer system composed of gravel, sand, and compressible clay with water filling every space in between. This boundary line marking the top of the saturated zone is known as the water table. The internal water pressure isn’t just statically present but holds the soil grains apart, similarly like water inside a soaked sponge keeping it from collapsing under its own weight.
However, when groundwater extraction supersedes the natural recharge rate, the water table drops drastically because natural surface recharge and deep percolation take anywhere from decades to millennia causing severe drops in the water table exposing empty spaces within clay and silt layers which collapse under the weight of the earth above it. This severe, permanent subsurface compaction across the entire valley manifests as land subsidence.
This isn’t a dramatic metaphor but a result of rapid urbanization in Kathmandu Valley that strains the aquifer system and causes significant land subsidence.
Satellite Proof
Fig: a) Sedimentation Thickness Map (Huang et al., 2024), b) Vertical deformation velocity map of Kathmandu Valley (2017–2024)
In recent years, remote sensing has enabled precise measurements of these surface movements. As shown in Figure 1a, central and western parts of the valley floor (shaded in red and orange) contain soft, compressible sediment layers stretching over 50-500m deep. These clay and silt deposits hold the highest volume of water and are the most vulnerable to the irreversible compaction from draining the aquifer system. A study published in 2025 analyzed 7 years of satellite radar data using InSAR (Interferometric Synthetic Aperture Radar), showing that between 2017 and 2024, parts of the northwest, south, and northeast valley floor of Kathmandu dropped at rates up to 21 centimeters per year indicated in Figure 1b. Central hotspot locations experienced cumulative drops exceeding 1.5 meters in less than a decade (Rawal & Wang, 2025).
This silent hazard causes slow, distributed settling which cracks foundations, subtly tilts the multi story buildings, and puts continuous pressure on water and drainage networks over years.
What is pulling out the water?
The main reason for this deficit is the gap between public supply and urban demand. The valley’s Kathmandu Upatyaka Khanepani Limited (KUKL), a central public utility, delivers an average of 92 million liters of water against an estimated demand of 470 million litres (S. Shrestha et al., 2023). To bridge this gap, residents, hotels, hospitals, industries, and tanker businesses turn towards the ground beneath them. Shallow and deep aquifers supply about 50 to 70% of all water consumed in the valley, especially during the dry season (G. Shrestha et al., 2023). These uncoordinated, unmonitored extraction networks are driven by private drilling and deep industrial boreholes severely depleting and straining the groundwater reservoir.
Extracting water is the first half of the equation, the other half is the simultaneous loss of natural recharge of monsoon rain. When hydrologists measured actual water infiltration rates across different land surfaces using double ring infiltrometers, the results varied widely from near sealed 0.01 cm/hour to permeable 37.2 cm/hour, averaging around 7.3 cm/hour across the base (G. Shrestha et al., 2023). In all, it shows the change of land cover as the second half of the equation, i.e. open soil absorbs rainfall and recharges deep layers whereas the paved, built surfaces acts like a sealant causing surface runoffs and forcing rainwater to drain.
Between 1990 and 2020, Kathmandu’s built-up area expanded by a staggering 368% (Rawal & Wang, 2025). Crucially, much of this concrete development spread directly over the valley's primary recharge belts specifically the northern foothills (Lapsephedi, Sundarijal, and Budhanilkantha) and southern pockets (Godavari and Nallu), where porous sediments naturally channel surface water into deep aquifers (Dahal et al., 2019; Lamichhane & Shakya, 2019). Hydrogeological flow models project that combining this rising extraction with the ongoing loss of recharge zones could force the water table down an additional 5.25 meters in the dry season and 7.39 meters in the wet season by the year 2050 (Lamichhane & Shakya, 2020).
Why does the system still respond?
The satellite data offers an encouraging answer. When the Melamchi Water Supply Project began piping water into the valley, the utility’s public groundwater pumping dropped from 58 million liters per day in 2021 to 26 million liters per day in 2024. Satellite radar captured the physical response immediately: in neighbourhoods where extraction dropped, the rate of subsidence slowed down (Rawal & Wang, 2025). While damage in long-stressed areas cannot be undone, newer development in eastern Lalitpur and Bhaktapur can still be saved if intervention happens now.
Fig : a) Manga Hiti (nepaltravellor.com), b) Pimbahal Pokhari (Prashant Shrestha)
Long before remote sensing, an ancient ground water monitoring system existed in Kathmandu valley through a network of stone water spouts known as hiti and neighbourhood ponds. A valley-wide survey cataloged 573 traditional stone spouts. Today, only 224 remain functional; 94 have completely disappeared — buried under roads or dried out permanently (Tripathi et al., 2019). The active spouts still deliver 2.4 million liters of free water daily, providing a vital lifeline for low-income households (Shrestha et al., 2022). Monthly monitoring of 16 stone spouts alongside 49 nearby wells confirms that both sources track the exact same water table (Smartphones for Water Nepal, 2022).
When deep wells over-extract, neighboring ancient hiti run dry. When communities step in, however, these systems recover. In Patan, a historic spout complex contaminated by a local factory in 2000 was saved through public advocacy. The factory was relocated, the stone filtration channels were restored, and the feeding pond (Pim Bahal Pokhari) was cleaned. Today, it supplies hundreds of local households once again.
What Restoring the Balance Looks Like
Reversing aquifer compaction requires pairing structural municipal policy with local infrastructure:
Rooftop Rainwater Harvesting: Kathmandu receives 1,300–1,600 mm of annual rainfall. A standard 100 m² rooftop can collect over 160,000 liters of water per year (SmartPaani, n.d.). Directing this clean water into recharge wells instead of storm drains recharges the soil directly beneath neighborhoods.
Protecting Key Recharge Belts: Cities must establish strict no-build zoning laws over the high-infiltration zones identified in Budhanilkantha, Sundarijal, and Godavari, treating these natural recharge areas as critical public infrastructure.
Restoring Urban Ponds (Pokhari): Community projects like the Gongabu-Ranibari recharge facility and Swayambhu pond restoration show that urban retention basins effectively absorb storm runoff and feed shallow aquifers.
The Broader View
Kathmandu is not an isolated case. Across the Hindu Kush Himalaya, rapidly growing urban centers like Pokhara, Itahari, and Dhangadhi are following the same playbook: paving over permeable soils before mapping their subsurface hydrology.
Kathmandu simply happens to be where satellite radar, citizen science, and hydrogeological modeling have intersected to make the process visible. The ground under the valley floor is sinking slowly, but the data proves that when extraction eases and water is allowed back in, the ground responds. The tools to fix it exist, they just need to be deployed at scale before temporary compression becomes permanent loss.
References
Dahal, A., Khanal, R., & Mishra, B. K. (2019). Identification of critical location for enhancing groundwater recharge in Kathmandu Valley, Nepal. Groundwater for Sustainable Development, 9, 100253.
Lamichhane, S., & Shakya, N. M. (2019). Alteration of groundwater recharge areas due to land use/cover change in Kathmandu Valley, Nepal. Journal of Hydrology: Regional Studies, 26, 100635.
Lamichhane, S., & Shakya, N. M. (2020). Shallow aquifer groundwater dynamics due to land use/cover change in highly urbanized basin: The case of Kathmandu Valley. Journal of Hydrology: Regional Studies, 30, 100707.
Rawal, S., & Wang, G. (2025). Long-term subsidence assessment by LiCSBAS and emerging hot spot analysis in Kathmandu Valley. Land, 14(4), 700.
Shrestha, A., Shah, D. N., Bajracharya, R. M., & Shrestha, S. (2022). Traditional stone water spouts status and its practical significance in urbanizing Kathmandu Valley, Nepal – a review. Environmental Challenges, 8, 100573.
Shrestha, G., Shakya, B. M., Shrestha, M. B., & Khadka, U. R. (2023). Water infiltration rate in the Kathmandu Valley of Nepal amidst present urbanization and land-use change. H2Open Journal, 6(1), 1–14.
Shrestha, S., Bista, S., Byanjankar, N., Shrestha, S., Joshi, D. R., & Prasai Joshi, T. (2023). Groundwater quality evaluation for drinking purpose using water quality index in Kathmandu Valley, Nepal. Water Science, 37(1), 239–250.
Shrestha, P. (2009, November 1). Pimbahal in HDR. Wikipedia Commons. https://commons.wikimedia.org/wiki/File:Pimbahal_in_HDR_%284084741920%29.jpg
Smartphones for Water Nepal. (2022). Traditional hiti system of the Kathmandu Valley. SmartPaani. (n.d.). Rainwater harvesting.
Huang, J., Sinclair, H., Pokhrel, P., & Watson, C. S. (2024). Rapid subsidence in the Kathmandu Valley recorded using Sentinel-1 InSAR. International Journal of Remote Sensing, 45(1), 1–20. https://doi.org/10.1080/01431161.2023.2283902
Nepal Traveller. (2022, April 8). The historical hitis of Kathmandu Valley. https://nepaltraveller.com/sidetrack/the-historical-hitis-of-kathmandu-valley
Tripathi, M., Hughey, K. F. D., & Rennie, H. G. (2019). The state of traditional stone spouts in relation to their use and management in Kathmandu Valley, Nepal. Conservation and Management of Archaeological Sites, 20(5–6).
