You can slow, divert, or even stop a lava flow, but it takes enormous resources and favorable conditions. Humans have successfully defended towns from advancing lava using three main approaches: cooling it with massive amounts of water, building earthen barriers to redirect it, and in rare cases, bombing lava tubes to disrupt supply channels. No method guarantees success, and each one works best against slow-moving flows that give responders days or weeks to act.
Cooling Lava With Seawater
The most famous example of stopping lava with water happened in 1973 on the Icelandic island of Heimaey, when the Eldfell volcano erupted and sent lava creeping toward the town’s harbor. Residents and emergency crews pumped seawater directly onto the advancing flow, starting with about 100 liters per second and eventually scaling up to 1,000 liters per second with enough pressure to shoot water 100 meters into the air. After 15 days of continuous pumping using nearly half a million cubic meters of seawater, a solid barrier of cooled rock formed across the front of the flow and stopped it. By the time operations ended in July, crews had sprayed roughly 6 million cubic meters of seawater onto the lava.
The results were dramatic but uneven. Early attempts were less effective because water ran off the surface before it could absorb enough heat to turn to steam. The real breakthroughs came when crews concentrated water on specific sections. At one point, a sustained flow of 400 to 500 liters per second onto a single area caused the lava to solidify not just where the jets hit, but 100 to 200 meters beyond, as the water traveled along the flow front and cooled it rapidly. The lava at that point was already moving slowly, just 1 to 2 meters per hour, which gave the cooling effort time to work.
This is a critical detail: water cooling works best when the flow is already sluggish. A fast-moving river of lava will simply overwhelm any volume of water you can deliver. The Heimaey success depended on timing, proximity to an unlimited ocean water supply, and a flow that was losing momentum on its own.
Earthen Barriers and Diversion Walls
Rather than stopping lava in place, the more common strategy is to steer it away from whatever you’re trying to protect. This means building massive walls of earth, rock, and volcanic debris in the lava’s path to redirect it toward less populated areas.
During the 1991 to 1992 eruption of Mount Etna in Sicily, the village of Zafferana Etnea sat directly in the path of an advancing flow. Engineers built an earthen barrier 234 meters long and 21 meters high using 370,000 cubic meters of earth, loose volcanic rock, and stones piled up by mechanical excavators. That wall held the lava back for about a month before being overtopped on April 9, 1992. Three additional smaller barriers, ranging from 90 to 160 meters long and 6 to 12 meters high, were then built to buy more time while the eruption wound down. The village survived.
Iceland has taken this approach further than any other country. Following a series of eruptions on the Reykjanes Peninsula starting in late 2023, Icelandic authorities built an extensive network of lava barriers to protect the town of Grindavík and the nearby Svartsengi geothermal power plant. The system now stretches roughly 13.5 kilometers (about 8.4 miles), built from around 2 million cubic meters of material. The barriers average about 7 meters (23 feet) tall, though sections where lava has piled up against the walls have been raised as high as 21 meters (69 feet). These defenses successfully redirected lava during multiple eruption events.
Building barriers requires heavy equipment, available material, and enough warning time. Lava rarely arrives without notice. Flows typically advance at walking speed or slower, giving communities hours to weeks to prepare. The key engineering challenge is reading the terrain correctly to predict where lava will go and placing walls where gravity will carry the diverted flow somewhere safe.
Bombing Lava Tubes
A more aggressive tactic involves using explosives to break open the underground tubes or surface channels that feed a lava flow. If you can collapse or rupture the supply system, the advancing front loses its source of fresh, hot lava and begins to cool and stall on its own.
This was attempted in 1935 when Mauna Loa in Hawaii sent a flow toward the headwaters of the Wailuku River, which supplied water to the town of Hilo. Volcanologist Thomas Jaggar called in the U.S. Army, which dropped twenty 600-pound bombs (each loaded with 300 pounds of TNT) on the lava flow’s source area on December 27, 1935. The flow stopped about six days later.
Whether the bombing actually caused the flow to stop remains genuinely controversial. Jaggar believed it worked, noting that the flow’s speed dropped from about one mile per day before the bombing to roughly 1,000 feet per day afterward. He reported that the supply channel had been “broken up” and fresh lava was spilling over the sides rather than feeding the front. But geologist Harold Stearns, who inspected the site, estimated the tube walls were 25 to 50 feet thick and doubted the bombs could have broken through. He believed the flow was already dying naturally. “I am sure it was a coincidence,” Stearns wrote decades later. Modern volcanologists mostly side with Stearns, though the question has never been definitively settled.
The bombing approach has never been proven to work reliably, and it carries risks: explosives could redirect lava unpredictably or create new flow paths toward populated areas.
Predicting Where Lava Will Go
Any defense strategy depends on knowing where the lava is headed. Volcanologists use simulation software that models how lava will flow across real terrain. One widely used tool, Q-LavHA, runs as a free plugin inside open-source mapping software. It combines probabilistic models with calculations of how lava cools and thickens as it moves, using high-resolution elevation data to predict which areas are most likely to be inundated. Users can simulate flows from a single vent or across a broad area, making it useful for both long-term hazard planning and real-time crisis response.
These models help engineers decide where to place barriers and how tall they need to be. Lava follows gravity, so the terrain’s shape is the most important variable. Even small ridges or valleys can determine whether a flow threatens a town or drains harmlessly into open land. Accurate, up-to-date elevation maps are essential, especially during ongoing eruptions where earlier flows reshape the landscape.
What Happens When Lava Meets the Ocean
When lava reaches the sea on its own, the interaction creates hazards beyond the flow itself. The lava releases sulfur dioxide and hydrochloric acid into the water, making it more acidic and generating brown hydrothermal plumes filled with tiny glass particles. Water temperatures in these plumes can reach 120°F. Near the coast, these conditions are devastating: researchers have found zones where all phytoplankton and algae disappeared entirely, with no detectable chlorophyll from the surface down to 15 feet.
Farther offshore, the story reverses. Iron and phosphorus from the lava act as fertilizer, triggering massive algal blooms. During Hawaii’s 2018 Kilauea eruption, satellite images revealed a bloom stretching about 100 miles long and 1 mile wide, with chlorophyll concentrations nearly 10 times higher than normal. So while lava sterilizes the nearshore environment, it feeds the base of the marine food web farther out. Large-scale water cooling operations, like those used in Iceland, would introduce some of these same chemistry changes to coastal waters, though at a smaller scale than a full ocean entry.
Why Stopping Lava Is So Difficult
Every successful defense has involved lava that was already moving slowly and an eruption that was winding down. No human intervention has ever stopped a large, fast-moving flow at full eruption intensity. The volumes involved are staggering: a single eruption can produce millions of cubic meters of molten rock, and the heat energy in a lava flow dwarfs anything humans can counteract directly. Cooling one cubic meter of lava from eruption temperature to solid rock requires removing an enormous amount of thermal energy, which is why the Heimaey effort needed 6 million cubic meters of water for a relatively small flow.
The practical reality is that stopping lava is more about buying time and steering damage than overpowering a geological force. Barriers get overtopped. Cooled surfaces crack as fresh lava pushes from behind. Bombing results remain ambiguous after nearly 90 years of debate. The most reliable protection is still choosing where to build in the first place, guided by hazard maps that show which areas are most likely to see lava in the coming decades. When an eruption does threaten infrastructure, the combination of barriers, water cooling, and accurate flow modeling gives communities their best chance of saving what matters most.