Typhoon Dolphin Turns West: Reading Japan’s Most Dangerous Forecast

In less than two days, Dolphin’s central pressure plunged 70 hPa and its sustained winds rose to the Japan Meteorological Agency’s highest strength class. The official five-day track brought it toward the Ogasawara region. What came after remained the question: a turn away, a run toward Honshu, or a farther-west route toward a Japan already managing heat and earthquake damage.

Forecast, not outcome: At 9:00 a.m. JST on July 30, JMA analyzed Dolphin near Wake Island at 910 hPa, with 55 m/s maximum sustained winds and 80 m/s gusts. It forecast 905 hPa, 60 m/s winds and 85 m/s gusts by Friday morning, then a westward track into the Ogasawara region by August 4. JMA had not forecast a western-Japan landfall at the reporting cutoff. Longer-range model scenarios showing such a risk were reasons to prepare and watch—not proof of a route.

The most important thing on the typhoon map Thursday morning was not the line. It was the bend that had not yet happened.

Far out in the western Pacific, near Wake Island, Typhoon No. 13 had become “violent,” the highest of the Japan Meteorological Agency’s strength classes. At 9:00 a.m., the agency placed its center at 16.4 degrees north, 165.8 degrees east, moving west-northwest at 20 kilometers an hour. Central pressure was 910 hPa. Maximum sustained winds, measured as a 10-minute average under JMA practice, were 55 meters per second; gusts were estimated at 80.

The storm’s name, Dolphin, sounds gentle. It comes from Hong Kong’s proposal to the regional Typhoon Committee and refers to the white dolphin, one of the territory’s emblematic animals. The atmosphere had produced the opposite impression. At 3:00 p.m. Tuesday, Dolphin was at 980 hPa with 35 m/s winds. Forty-two hours later, its pressure was 70 hPa lower and its sustained wind 20 m/s stronger.

That transformation was already real. Its eventual relationship to Japan was not. The five-day JMA forecast carried Dolphin west-northwest past the Wake Island region, toward Minamitorishima and then into the Ogasawara area. Beyond that horizon, computer guidance offered a fan of futures. Some brought the cyclone north or northeast. Others maintained a farther-west track, creating a possible threat to the main islands. The responsible way to read that fan was neither dismissal nor certainty. It was time.

910 hPaJMA central-pressure analysis at 9:00 a.m. July 30

55 m/sMaximum sustained wind; JMA’s “violent” threshold begins at 54 m/s

150 kmRadius of winds of at least 25 m/s around the center

August 4Day-five forecast placed the center in the Ogasawara region

A storm deepening faster than its destination could settle

Tropical cyclones are heat engines. Warm ocean water supplies water vapor. As that vapor rises and condenses inside deep thunderstorms, it releases heat, lowers pressure and accelerates the circulation. JMA notes that typhoons generally form over water of about 26–27 degrees Celsius or warmer, but temperature alone is not enough. Moist air, low disruptive wind shear and an efficient outflow high above the storm all matter.

Dolphin crossed an environment unusually favorable for organization. Its thunderstorms wrapped more tightly around a developing eye, pressure fell and the wind field intensified. The 70 hPa fall from Tuesday afternoon to Thursday morning is a startling measure of the change, although formal “rapid intensification” definitions often use different wind-averaging periods and 24-hour windows. The safer description is also the obvious one: Dolphin intensified extraordinarily quickly.

The forecast at 10:00 a.m. showed that process continuing. JMA projected a 905 hPa center with 60 m/s sustained winds and 85 m/s gusts at 9:00 a.m. Friday. It then anticipated gradual weakening—910 hPa on August 1, 920 on August 2, 925 on August 3 and 940 on August 4—while the cyclone moved steadily west. Weakening at sea, however, is not the same as harmlessness near land. The day-five forecast still called for a “very strong” typhoon with 45 m/s sustained winds and 65 m/s gusts.

Central pressure is an indicator, not an impact forecast. Damage depends on the actual wind that reaches a place, the size and structure of the circulation, forward speed, rainfall, terrain, tide and the vulnerability of buildings and infrastructure. The 85 m/s gust in the forecast referred to the storm’s maximum over open water near the core. It was not a prediction that every place inside a circle would experience that wind.

What “violent” means—and what it does not

JMA classifies typhoon strength using maximum 10-minute sustained wind. “Strong” begins at 33 m/s, “very strong” at 44 m/s and “violent” at 54 m/s. Dolphin’s 55 m/s analysis put it just inside the top category. The agency also analyzed a 150-kilometer radius of storm-force winds of at least 25 m/s and a larger gale-force area of at least 15 m/s, extending 330 kilometers north of the center and 220 kilometers south.

The typhoon had no official size label. That does not mean it was tiny. JMA reserves “large” for a 15 m/s wind radius of at least 500 kilometers and “very large” for at least 800 kilometers; storms below that threshold are shown without a size adjective. A compact, intense cyclone can create an especially sharp transition from manageable weather to destructive wind near its eyewall.

JMA’s wind-impact guide gives the numbers human meaning. At average winds of 20–25 m/s, people may be unable to stand without holding on and flying objects can injure. Around 25–30 m/s, inadequately secured scaffolding can collapse and moving trucks can overturn. Above 35–40 m/s, many trees can fall, utility poles can be damaged, exterior materials can scatter widely and some buildings can collapse or deform. Those are generalized effects, not a promise of identical damage in every location.

The forecast line is not a railway track. It is one summary through a widening field of plausible positions, and the wind is wider than the line.

The cone is uncertainty, not the size of the storm

A JMA forecast circle is designed so that the cyclone’s center has roughly a 70 percent probability of being inside it at the valid time. The line connecting circle centers is not a guarantee that the storm will follow that exact route. Nor does the circle show the physical width of rain or wind. The separate storm-warning area estimates where violent winds could occur if the center follows any position inside the forecast circle.

At the cutoff, the day-five forecast for 9:00 a.m. August 4 placed the center at 24.1 north, 142.6 east, with a 280-kilometer forecast-circle radius. The storm-warning area extended 480 kilometers around the projected center. That was an Ogasawara-region forecast, not a precision appointment with Chichijima or Hahajima.

A figure that appeared reassuring required the same care. The 9:00 a.m. probability table gave Ogasawara a zero-percent chance of entering Dolphin’s storm-force wind area during the first 72 hours, through the morning of August 2. That did not cancel the day-five risk. It said the system was still too far away for such winds during the shorter window. A zero in one time interval is not an all-clear for the days after it.

Map element	What it means	Common misreading
Center line	Connects the centers of successive forecast circles.	“The typhoon will travel exactly on this line.”
Forecast circle	Area with roughly a 70% chance of containing the center at that time.	“This is the size of the typhoon.”
Storm-warning area	Area that could enter winds of at least 25 m/s, allowing for track uncertainty.	“Every point inside will experience the maximum wind.”
Entry probability	Chance a locality enters the storm-force wind area within a defined period.	“A low early number means no later danger.”

Ogasawara: where distance reduces options

On a national map, the Ogasawara Islands are part of Tokyo. Operationally, they live on a different clock. Chichijima lies about 1,000 kilometers south of the capital. There is no civilian airport linking the inhabited islands to the mainland; passenger and much cargo access depend on the long sea route. When waves build and sailings change, the consequences include food, fuel, mail, construction supplies, tourism and the timing of outside assistance.

That isolation makes early preparation rational even while the center track remains uncertain. Island households and businesses cannot assume that a missed ferry can be replaced the next morning. Boats need safe harbor decisions before conditions deteriorate. Loose objects, shutters, water, medicine, backup power and communications have to be addressed while outdoor work is still possible.

History gives the islands no reason to treat a distant storm casually. In November 1983, Typhoon No. 17 passed close to Chichijima. The station recorded a sea-level pressure of 932.5 hPa and a 58.6 m/s gust. Contemporary damage summaries list six injuries, 30 buildings destroyed, 24 half-destroyed and 73 partly damaged. In April 2022, Typhoon Malakas brought a 46.5 m/s gust and an almost island-wide power failure on Chichijima.

Those events are analogues of vulnerability, not forecasts for Dolphin. Track angle, storm size, tide and structure differ from cyclone to cyclone. Their lesson is narrower and more useful: a remote island can lose power, transport and repair capacity at the same moment. Preparedness has to account for the days after the wind, not only the hours during it.

Why western Japan appeared in the conversation

Low-latitude typhoons in the western North Pacific are often steered westward along the southern edge of the subtropical Pacific high. Farther north, a weakness in that high or an approaching trough can allow a cyclone to turn poleward and eventually northeast into the mid-latitude westerlies. The timing and location of that turn—recurvature—can separate an offshore passage from a direct approach to Japan.

Dolphin’s five-day official track had not yet reached that decision point. Forecast models looking farther ahead disagreed about the shape and strength of the high and the influence of weather systems to the north. A stronger ridge could keep the cyclone moving farther west. An earlier weakness could draw it north toward the Ogasawara chain or curve it away. Small differences accumulated across several days into very different outcomes near the main islands.

That is why reports could responsibly say western Japan faced a possibility while it would have been irresponsible to announce a landfall. At 10:12 a.m. Thursday, no official JMA forecast circle reached Kyushu, Shikoku, Chugoku or Kansai. There was no basis for naming a landfall prefecture or time. There was a basis for emergency managers, utilities, transport companies and households to review plans before the track became more certain and the preparation window became shorter.

A typhoon approaching an earthquake zone would be a compound disaster

The western-Japan scenario mattered more than usual because Kumamoto was already in crisis after the July 28 earthquake produced the maximum 7 on Japan’s seismic-intensity scale. Roofs had been damaged, people were in shelters, utilities were under strain and rescue operations continued in extreme heat. Even a glancing rain event could carry more consequence than it would have a week earlier.

JMA and the Ministry of Land, Infrastructure, Transport and Tourism explicitly recognized that vulnerability. Because shaking can loosen slopes, they lowered landslide-warning thresholds in heavily affected municipalities. The temporary criteria were set at 70 percent of normal in 15 Kumamoto municipalities and at 80 percent in additional communities in Kumamoto, Nagasaki and Kagoshima.

That is an administrative expression of a physical fact. Cracked soil, disturbed retaining walls, blocked drainage and unstable rock can fail under less rain than an intact landscape. Blue tarps protecting roofs can become dangerous projectiles. Temporary power and water systems can be interrupted. Evacuation routes already narrowed by debris or inspections can close again.

If the forecast shifts toward Japan

Follow official timing: use JMA warnings, local government evacuation information and storm-force wind probabilities—not a single social-media model image.
Finish outdoor work early: secure tarps, scaffolding, signs, tanks, construction material and household objects before wind begins.
Plan for outages: charge devices, protect medicines, confirm water, food, lighting and safe generator use.
Respect earthquake damage: avoid cracked slopes, retaining walls and buildings; use the lowered landslide thresholds and local hazard maps.
Move before conditions close routes: once violent wind begins, outdoor evacuation can become more dangerous than sheltering.

Japan’s typhoon memory was built by different kinds of loss

The 1959 Ise Bay Typhoon remains the origin point of modern Japanese typhoon policy. It killed or left missing 5,098 people, largely through catastrophic storm surge. The disaster helped drive the Basic Act on Disaster Management and a national framework for warning, evacuation and coordinated response. It also left an enduring lesson: a reasonably forecast track does not guarantee that society has understood the scale of the impact.

Typhoon Jebi in 2018 demonstrated the modern infrastructure version of that lesson. At Kansai Airport, the maximum gust reached 58.1 m/s. Record storm surge flooded runways, and a drifting tanker struck the access bridge. Fourteen people died across the affected region, and the peak power outage in Kansai reached about 2.247 million households and businesses. A cyclone became an airport, port, road, rail, communications and energy emergency at once.

The 2019 Boso Peninsula Typhoon showed how a relatively compact windstorm could concentrate damage. Chiba recorded a 57.5 m/s gust; two transmission towers and nearly 2,000 utility poles were damaged or destroyed, contributing to a peak outage of about 935,000 customers. Heat and delayed restoration turned the electrical failure into a prolonged public-health problem.

Dolphin was not destined to repeat any of these disasters. The point of history is not to force a new storm into an old track. It is to identify the systems that fail in recognizable combinations: surge and evacuation, wind and power, transport and isolation, earthquake damage and later rain.

An active season, and a warming background

Dolphin became the fifth typhoon of July and the thirteenth of 2026. JMA’s preliminary count showed at least one formation in every month from January through July: one each in the first four months, two in May, two in June and five in July. July’s climatological average is 3.7. Weathernews noted that only 1965, 2015 and 2026 had produced a typhoon in every month from January through July since records began in 1951.

A busy seven months does not by itself establish a climate trend, and one storm’s rapid strengthening cannot be attributed to climate change from a forecast bulletin. Seasonal variability, atmospheric circulation and ocean conditions all matter. The IPCC’s broader assessment is different in scale: with additional warming, average peak tropical-cyclone winds and the global proportion of the most intense storms are expected to increase, while cyclone rainfall rates also rise.

The distinction matters. Climate change is not a shortcut for explaining every wobble or every pressure fall. It is a shift in the background conditions against which Japan must design warnings, drainage, power networks, shelters and recovery plans. Dolphin’s immediate intensity came from the ocean and atmosphere it encountered. The public decision remained simpler: use the forecast window before the forecast becomes an emergency.

Watch the bend, not the drama

In the next advisories, four changes would matter most. First, whether Dolphin held its extraordinary intensity or began a faster weakening cycle. Second, whether the five-day circles shifted north, south, east or west around Ogasawara. Third, whether the spread between longer-range models narrowed around a recurvature point. Fourth, whether JMA’s local storm-force wind probabilities and early-warning information began rising for specific islands or prefectures.

Pressure headlines would continue to attract attention. Track and size would determine who experienced the weather. A 940 hPa cyclone passing far south can be less damaging to Japan than a weaker storm moving slowly over a vulnerable watershed. An offshore center can still create high waves and dangerous coastal conditions. Rain can begin ahead of the core as moisture interacts with terrain or a front.

At the reporting cutoff, the most accurate sentence was also the least sensational: Dolphin was a violent typhoon with an official path toward the Ogasawara region, and its route after that remained uncertain. Ogasawara had a reason to prepare first. Western Japan had a reason to review its plans. Kumamoto had a reason to treat even a low-probability later approach seriously.

The line on the map would move again. The useful forecast was not the one that looked most frightening at one moment. It was the sequence—position, probability, wind field, timing and local warning—read without turning uncertainty into either panic or permission to wait.

Reporting Notes and Sources

Information was checked through 10:12 a.m. JST on July 30, 2026. Typhoon position, intensity and track forecasts change with every advisory. Numbers in this article describe JMA’s analysis and forecast available at the cutoff, not the storm’s later verified path. Longer-range western-Japan scenarios are identified as uncertain model guidance, not an official landfall forecast.