Why do landslides sometimes happen after the rain has stopped?
― What held the slope up was the "pushing match" between soil grains
The morning after days of heavy rain, once the sky brightens, a slope gives way. This has been reported again and again. Water doesn't just add weight to soil. It loosens the force with which soil grains press against each other, weakening the friction that was holding the slope in place. And water takes time to soak down deep. So the most dangerous moment isn't necessarily while it's raining.
Scoop up dry sand at a sandbox and let it trickle down, and it forms a neat cone-shaped pile. It never gets steeper than a certain angle — beyond that, it simply flows back down on its own.
Now gently pour water over that pile. For a while it keeps its shape, but once the water has soaked all the way through, the pile suddenly slumps and spreads out toward its base.
Sometimes it collapses even after you've stopped pouring. That "slightly delayed collapse" is what today's story is about.
There are only two reasons a slope collapses
Soil is made of fine grains pressing against one another. When the gaps fill with water, the water pushes back from inside, weakening how hard the grains press together. When that pushing weakens, so does the friction that resists sliding.
Water that soaks into the ground can take many hours to work its way down to the deeper parts of a slope. That's why friction is weakest not at the peak of the rain, but afterward.
A slope normally stands because the "force trying to slide it down" and the "friction holding it in place" are balanced. Water slightly increases the former and sharply reduces the latter. Let's look at each in turn.
What holds a slope up is the pushing match between grains
Why doesn't a pile of dry sand ever get steeper than a certain angle? Because the grains touch and catch against each other, rough and interlocking. How strong that catch is depends on how hard the grains press against one another. The harder they're pressed together, the harder it is for them to slide sideways.
Now bring in water. When the gaps in the soil fill up with water, that water is also under pressure, and it pushes the grains apart from within. The grains come closer to "floating," if only slightly. When the force between them weakens, so does their grip. The soil's weight barely changes, yet the force holding the slope up alone drains away.
It takes time for water to arrive
To reach the state shown on the right of Figure 1, water has to work its way down to the deep parts of the slope. But water inside soil doesn't move like a river current — it seeps slowly through fine gaps.
Because of this, it's thought that it can take anywhere from several hours to about half a day between the surface getting wet and a layer several meters deep becoming fully saturated. Even after the rain stops and the sky brightens, water underground may still be working its way downward. The moment when friction is weakest arrives later than the peak of the rain. That's why "it's stopped, so it's safe" doesn't hold true.
On top of that, mountain slopes don't all collect rainwater evenly. Valley-shaped spots gather water from a wide surrounding area. Even with the same amount of rain, such spots can become saturated first.
Tree roots are said to bind soil together like a net, making shallow collapses less likely. But roots typically reach only about 1 to 2 meters below the surface. For a "deep collapse," where the slide surface lies much deeper than that, roots offer almost no help. A forest on the slope doesn't mean it's safe.
Japan's Meteorological Agency uses a figure called the "soil water index," which estimates how much of the fallen rain has accumulated underground. It's designed to capture not just how hard it's raining, but how much has built up. When this index crosses a threshold, the prefecture and the Meteorological Agency jointly issue a landslide warning.
So what should you actually do?
- Even after the rain stops, stay away from cliffs and steep slopesUnderground, water may still be working its way down. It takes time before it's actually safe.
- If you notice odd sounds, smells, or changes in water, tell someone and evacuate right awayA rumbling from the mountain, pebbles pattering down, spring water turning cloudy — these are considered warning signs.
- When you flee, move sideways, away from the slope, not straight downhillMoving sideways, away from the slope, is thought to get you to safety faster than heading straight down the incline.
When a landslide warning or evacuation advisory has been issued, treat the danger as ongoing even if the rain has eased. Move to higher ground in a sturdy building, away from cliffs and valleys. If going outside feels riskier, move to a room on the side of the house farthest from any slope. If someone is caught in a slide or injured, call 119 and leave the rescue to firefighters and police. Trying to dig someone out yourself risks getting caught in a secondary collapse. Follow your municipality's instructions and official weather information when deciding whether to evacuate.
Summary
A slope stands because of the friction created by soil grains pressing against each other. Water that soaks in loosens that pushing match and weakens the friction. And because it takes time for water to reach deep down, the weakest moment can arrive after the rain has already stopped. How bright the sky looks and what's happening underground are two different things.
What decides whether a slope collapses isn't the rain that's falling.
It's how much water has already built up underground.
A river suddenly rising after the rain has stopped works on the same "delayed effect" principle. For more, see "Why does a river suddenly rise even when the sky is clear?" The way water in soil gaps can strip the ground of its support is also covered in "Why does liquefaction turn solid ground into something like water?" How tree roots carry water is explained in "How can trees draw water up to 100 meters high?" And how rainwater carves valleys over long stretches of time, creating waterfalls and cliffs in the path of someone lost in the mountains, is covered in "Why shouldn't you follow a stream downhill if you get lost in the mountains?"
- Build a small pile of dry sand on a tray. Note the steepest angle it holds without your hand supporting it.
- Using a spray bottle, wet just the top of the pile a little at a time. You'll notice that right after the surface gets wet, the shape actually holds even better.
- Keep adding water and wait until it reaches the base of the pile. Some time after you stop adding water, the base may suddenly slump and spread out.
Do not try this on a real outdoor slope. This is only a small-scale model — an actual slope behaves quite differently depending on how its soil layers are stacked and where groundwater flows.
Want to know more? ― Terms, formulas, and how this connects to textbooksWe label which level each part belongs to, from middle-school science to university-level courses
- MSCovered in middle-school science
- HSCovered in high-school "Physics Basics"
- HS+High-school advanced content, or textbook sidebar material
- Univ.Not covered in high school — university-level specialist content (geotechnical engineering, hydrology)
- ResearchNot yet settled even at university level — something researchers are actively investigating
MSTerms: this phenomenon has names
- Slip surface: the boundary along which the soil on a slope begins to slide. This is the yellow dashed line in Figure 1.
- Pore water pressure: the pressure of water filling the gaps between soil grains. It weakens the pushing match between grains from the inside.
- Effective stress: the force with which grains are actually pressing against each other. It equals the total force acting on the soil minus the pore water pressure.
- Soil water index: a figure estimating how much fallen rain has accumulated underground. Used to decide when to issue landslide warnings.
MSHSCheck it with a formula: how much does friction drop when water gets in?
Consider a soil mass weighing 1000 newtons sitting on a 30-degree slope. All we need is resolving a force into components along and perpendicular to the slope, plus the basic idea of friction.
| In symbols | F = W × sinθ R = μ × ( W × cosθ − U ) Collapse condition: R < F |
| In words | Sliding force = weight × the along-slope component. Holding friction = a slipperiness coefficient × (the force pressing onto the slope − the push-back force from water in the gaps) |
| Where this comes from | "Resolution of forces," which splits gravity into a component along the slope and one perpendicular to it, and the "law of friction," which says friction is proportional to the pressing force. Subtracting the water's push-back force is the geotechnical-engineering idea of "effective stress" |
What the symbols mean: W is the weight of the soil mass, θ is the slope angle, μ is the slipperiness coefficient, U is the push-back force from water in the gaps, F is the sliding force, and R is the holding friction (forces in newtons).
| Weight of the soil mass | 1000 newtons (roughly 100 kilograms' worth) |
| Slope angle | 30 degrees |
| Along-slope component (at 30°) | a value of 0.50 |
| Perpendicular-to-slope component (at 30°) | a value of 0.87 |
| Coefficient of soil slipperiness | 0.7 (a rough value for dry soil) |
| Push-back force from water in the gaps | 400 newtons (assuming rain has soaked in) |
| Symbol | Meaning and unit |
| Weight | Size of the gravitational force acting on the soil mass [newtons] |
| Component | Multiplier used to split the weight into slope-relative directions [no unit] |
| Coefficient | Multiplier for how much friction results from the pressing force [no unit] |
| Push-back force | Size of the force with which water in the gaps pushes the grains apart [newtons] |
| Sliding force | 1000 × 0.50 = 500 |
| Force pressing onto the slope (dry soil) | 1000 × 0.87 = 870 |
| Holding friction (dry soil) | 870 × 0.7 = 609 |
| Pressing force (after water gets in) | 870 − 400 = 470 |
| Holding friction (after water gets in) | 470 × 0.7 = 329 |
| Ratio of holding force to sliding force (dry soil) | 609 ÷ 500 ≒ 1.22 |
| Ratio of holding force to sliding force (after water gets in) | 329 ÷ 500 ≒ 0.66 |
While dry, friction of 609 outweighs the sliding force of 500, and the slope stays put. Once water gets in, friction drops to 329, below 500. The instant the balance breaks, the slope starts to move — even though the weight itself never changed from 1000. It's not that the soil gets heavier and collapses. It's that the water strips away the force holding it up. That's the crux of this phenomenon.
HSHS+Why does it kick in "late"?
HSHow fast water moves through soil depends on how fine the gaps are. The finer the soil, the more slowly water can move through it. It's thought that water can take several hours to about half a day to spread through several meters of depth from the surface. This delay is what separates the peak of the rain from the peak of the danger.
HS+Also, when water in the soil's gaps is squeezed, that pressure spreads to the surrounding area. How it spreads varies greatly with how fine the soil is, and if there's a layer that water passes through poorly, water builds up above it and the pressure there rises. This is thought to be why, when a slope contains such a poorly-draining layer, the boundary there tends to become the slip surface.
Univ.Slope stability from a geotechnical-engineering perspective
In geotechnical engineering, the ratio of the "force trying to cause sliding" to the "force resisting it," measured along the slip surface, is called the factor of safety, and a value below 1 means collapse. The resisting force is expressed as the sum of effective stress multiplied by the friction coefficient, plus the soil's cohesion. Effective stress is the total force minus the pore water pressure. The "principle of effective stress" — that only what remains of the total force after subtracting water pressure generates friction — comes from Terzaghi, and the formula for the resisting force is known as the "Mohr–Coulomb failure criterion." This framework was worked out in the first half of the 20th century and still underlies the design of slopes and levees today. Methods that compute groundwater level from rainfall infiltration and track how the factor of safety changes over time are also widely used.
📖 For the derivation of the formula and further reading: Effective stress (Japanese Wikipedia)
ResearchWhat's still not fully understood
- Predicting exactly when a collapse will happen. Even when it's possible to identify which slopes are dangerous to some degree, pinpointing the exact hour and minute of a collapse in advance is still considered difficult.
- What triggers deep collapses. Collapses where tens of meters of depth move together as a mass don't always correlate clearly with rainfall, and their relationship to groundwater pathways is under study.
- How well tree roots actually work. How the soil-holding strength of roots changes with tree species, age, and years since logging — including how to even measure it — remains an active area of research.
- The link to a changing climate. As short bursts of intense rain become more common, how patterns of collapse will change is being examined region by region.
In other words, even this article only explains things as far as they're currently understood. That's exactly why it makes sense to act on a hunch of danger and get away early.
Connections to textbooks, by level
| Level | Subject / unit | Where in this article |
|---|---|---|
| MS | Science — forces at work / changes in the land | Forces acting on a slope, water soaking in |
| HS | Physics Basics (resolution of forces, friction) | Calculating sliding force and friction |
| HS+ | Advanced / sidebar (water permeability, pressure transmission) | Why the effect kicks in late, poorly-draining layers |
| Univ. | Geotechnical engineering / hydrology (effective stress, factor of safety) | Calculating slip-surface stability |
| Research | Erosion-control and disaster science (ongoing) | Predicting collapse timing, deep collapses, root effects |
| ― | Everyday relevance | How to read landslide warnings, deciding to get away from cliffs early |
- Ministry of Land, Infrastructure, Transport and Tourism (国土交通省), "Basic Knowledge on Landslide Prevention" and explanatory materials on landslide warning signs.
- Japan Meteorological Agency (気象庁), explanatory pages on the "soil water index" and "landslide warning information."
- High-school "Physics Basics" textbook sections on resolution of forces and frictional force.
- Terzaghi, K., "Theoretical Soil Mechanics", 1943. (The classic textbook that established the concept of effective stress)
※ This article is a general-audience science explainer. The figures given are approximations meant to help illustrate the mechanism. For actual evacuation decisions, follow the instructions of your municipality and the information released by the Japan Meteorological Agency and your prefecture.