Typhoon Dolphin Update: Japan on August 7, Then Jiangsu on August 9

Updated 1 August 2026

Yesterday the forecast for Dolphin spanned three countries and nobody would commit. Today there is a timeline, and it is specific enough to plan around.

Before the details, one piece of housekeeping. On 31 July this site said the cone should compress once the recurve resolved, and that Dolphin would likely weaken well below its peak before reaching land. The first half held. The second is still open — the storm is stronger today than the guidance suggested it would be.

Where it is now

The Joint Typhoon Warning Center placed Dolphin roughly 181 nautical miles south-southwest of Wake Island, tracking west-northwest at about 13 mph with sustained winds near 165 mph.

Satellite presentation is symmetrical with a clean eye. The storm appears to be working through another eyewall replacement, which explains the fluctuation in the intensity numbers between advisories.

JTWC expects a peak around 167 mph within 36 to 48 hours, followed by gradual weakening. Even as the peak winds come down, the wind field stays broad and the seas stay very high — a detail that matters more for shipping and coastal surge than the headline category does.

The timeline that emerged today

A longer-range projection now puts the sequence like this:

7 August — landfall in Japan, with very heavy rain into Kumamoto Prefecture, Fukuoka and the surrounding area
9 August — a second landfall in Jiangsu province, north of Shanghai
11 August — the remnant circulation moving on toward Henan

That is a far more concrete picture than yesterday's three-country spread. It is also still a projection at six days out, and the agencies are careful to say the route and any landfalls remain subject to change.

Where to check this yourself

Do not take any single blog's word for a storm, including this one. These are the agencies that actually issue the warnings, and they are the sources the numbers in this article come from.

Japan Meteorological Agency (JMA) — the official authority for Japanese waters, warnings and landfall.
Korea Meteorological Administration (KMA) — Korean track guidance and warnings.
Joint Typhoon Warning Center (JTWC) — the US centre behind most Category-equivalent figures quoted for western Pacific storms.
Hong Kong Observatory — clear track maps, and useful because it shows several agencies side by side.
China National Meteorological Center — the authority for any Chinese landfall.

If those five disagree with each other, that disagreement is the story. It is the same spread this article is describing, and no blog resolves it faster than the agencies do.

What changed to make this possible

High pressure is building to the northwest of the storm. JTWC has Dolphin holding a northwest heading for about 72 hours, then bending more westward as that ridge asserts itself.

That westward bend is the whole ballgame. It is what takes the storm from an open-ocean track into the Japan-then-China sequence above. Yesterday the models could not agree on whether the ridge would hold; today they mostly do.

If you are travelling

Kyushu, 5 to 8 August. This is now the highest-exposure window in the highest-exposure region. Kumamoto and Fukuoka are named directly in the rainfall projection. If your dates sit inside it, this is the point at which watching turns into planning.

Shanghai and Jiangsu, 8 to 10 August. Second in line. Airports in the region tend to cancel in blocks rather than individually once warnings go up.

Tokyo. Still outside the projected track. Rain and wind on the fringe are plausible; a direct hit is not currently indicated.

Airlines generally publish travel waivers 48 to 72 hours ahead. For a 7 August landfall that means watching for announcements around 4 or 5 August.

Why a six-day forecast is still worth reading

There is a reasonable objection to everything above. If the agencies say the track can still change, why read a six-day projection at all?

Because the value is not in the date. It is in knowing which decisions are now cheap and which are about to get expensive.

Six days out, changing a flight usually costs nothing or close to it. Two days out, when the waiver is issued and everybody moves at once, the seats you wanted are gone. The long-range forecast does not tell you what will happen. It tells you when to start paying attention, and that is worth more than a false precision would be.

It also tells you what to rule out. Tokyo sitting outside the projected track for six consecutive model runs is genuinely useful information, even though the landfall point itself is unsettled. Knowing where the storm is not going is often the more actionable half.

The mistake is treating a six-day line as a promise and booking around it as though it were. The right use is the opposite: assume the specific date will move, and use the window to decide whether your plans have any flexibility left in them. If they do not, that is the thing to fix now, while fixing it is still free.

The number to stop looking at

165 mph is today's figure over open ocean near Wake Island. It is not what arrives in Kyushu.

Dolphin has thousands of kilometres of travel ahead, and the East China Sea offers cooler water and more shear than the open Pacific. Weakening before landfall is the expected path. Rain totals and the size of the wind field will matter more to people on the ground than the peak intensity ever will.

This page will be updated as the 2 and 3 August model runs come in.

More Lao and Thai food

A cyclone's spiral bands well away from land. Library image for illustration. It is not live satellite imagery of the storm discussed here — for current imagery use the agency links above.

이 블로그 검색

Coinkoink