How to Read PAGASA's 5-Day Track Forecast (A Walkthrough)
A five-day track forecast contains four separate claims, and most people read it as though it contains one.
There is a line, a shaded area around the line, a set of labels on the line, and a set of coloured areas on land. They are produced by different processes, they carry different levels of confidence, and the one people fixate on — the line — is the least reliable thing on the map.
The four things on the map
The line is the forecast position of the centre at each future time step, joined up. It is the single most likely path of one point. It is not a route, it is not a corridor, and the storm is not steered along it.
The cone is the area around the line inside which the centre is expected to fall most of the time, based on how wrong forecasts of that lead time have been historically. It is a statement about forecast error, not about the size of the storm. This distinction is the single most consequential misreading in the whole product, and this site has a longer piece on why the cone is not the storm and the line is not the forecast.
The intensity labels on each forecast position are the weakest part of the map. Track forecasting has improved enormously over thirty years; intensity forecasting has improved far less, and rapid changes in strength remain the thing forecasters are worst at.
The wind signals shaded over provinces are a different product entirely. They are not derived from the cone. They are a statement about expected wind over a named area within a stated period, and they are what the school and work closure rules are built on. The wind signal numbers are explained separately, and the important point here is that a province can be inside the cone with no signal, or under a signal while outside the cone.
A worked example, and what it shows
Take a system this site followed in August — Typhoon Saudel, which tracked toward the Ryukyu Islands and was covered here at the time. The official figures then ran roughly as follows: 65 kt at 03:00 on 22 August, 90 kt twenty-four hours later, with central pressure falling from 970 hPa to 935 hPa over that day, and a forecast peak near 925 hPa on 24–25 August before easing to about 950 hPa and 80 kt near the islands on 26 August.
The Saudel figures above are the Japan Meteorological Agency forecast values as reported in this site’s coverage at the time. PAGASA’s archived bulletins and forecast cones are not available in a machine-readable form, so this page contains no computed table and uses two decision tools instead. Checked 26 September 2026.
Two things in that sequence are worth extracting, because they generalise.
First, the intensity forecast moved in steps of 25 kt in a day and then flattened. A traveller reading only the peak number — 925 hPa, a serious storm — would have prepared for the wrong thing on the wrong day. The number that mattered for a specific island was the one attached to the time the centre was nearest, not the peak anywhere on the track.
Second, the forecast included a weakening on approach. Never plan around a forecast weakening. Published intensity error at four and five days runs around 20 kt in each direction, which is the width of a whole category. A storm forecast to arrive at 80 kt can plausibly arrive at 100.
Checklist one: reading a track in order
Work through the map in this sequence. It takes about ninety seconds and it prevents nearly every common misreading.
- Find the issue time first. A five-day track is stale within six hours. If you cannot see when it was issued, do not use it.
- Look at the cone, not the line. Ask whether your location is inside the shaded area at any point, and ignore the line entirely on the first pass.
- Check whether you are near the edge or the middle. Near the edge means the situation can change completely with the next issue; near the middle means it probably will not.
- Read the time labels, not just the shape. The useful question is what day the centre is nearest you, because that is when your flights and ferries are affected.
- Now read the intensity at that specific time step — not the peak somewhere else on the track.
- Then look at the wind signals separately. This is the closure decision and it is not visible in the cone.
- Finally, look for the rainfall product. It is issued separately, it is frequently the actual hazard, and it is not on the track map at all.
- Compare with the previous issue. A cone that has shifted consistently in one direction over three issues is telling you something a single map cannot.
Checklist two: what each reading should make you do
| What the track shows | What it means | Decision |
|---|---|---|
| You are outside the cone at every time step | Low probability of the centre reaching you — but not of rain | Carry on. Check the rainfall product, not the track. |
| You are at the edge of the cone on day 4 or 5 | Very low confidence at that range | Do not cancel yet. Re-read in twelve hours; this is the range that changes most. |
| You are inside the cone on day 2 | Meaningful likelihood | Move flexible bookings now, while waivers are open and seats exist. |
| A signal is raised over your province | Closures follow from this, not from the cone | Assume schools and government work stop. Plan for the day to be lost. |
| Cone missing you but a red rainfall warning is up | The dangerous case. Wind maps look clear | Do not travel within the city. This is what does most of the damage. |
| Forecast shows weakening before it reaches you | Intensity error is around 20 kt | Prepare for the stronger case anyway. Weakening forecasts miss. |
| Cone has shifted the same way on three consecutive issues | A trend, not noise | Act on the trend rather than waiting for the next map. |
| You have an inter-island ferry booked | Sea state lags the storm | Assume two extra days. The track map says nothing about this. |
Two maps that disagree
A storm approaching the Philippines is covered by more than one agency, and their maps do not match. This confuses people into thinking one of them must be wrong.
PAGASA issues the Philippine bulletin, with Philippine wind signals and a local name for the storm. The Japan Meteorological Agency issues regional guidance for the whole western Pacific with the international name. The US Joint Typhoon Warning Center issues its own track. All three are competent, and they will differ — on position by tens of kilometres, on intensity by more than that, and sometimes on the shape of the recurve.
The intensity gap is usually not a disagreement about the storm at all. American agencies use a one-minute sustained wind and most of the rest of the world uses a ten-minute average, which puts the same system roughly ten to fifteen per cent apart on paper. A number that looks alarming against another number is often just a different stopwatch.
The rule for deciding which to follow is simple and it is not about which looks best. Over Philippine territory, PAGASA is the authority — it issues the signals, and the closures are keyed to those signals rather than to anybody else’s map. Use the others for a second opinion on the track at long range, where more views genuinely help, and use PAGASA for anything that turns into a decision.
What the product deliberately does not tell you
Three absences are worth naming, because people go looking for them on the track map and conclude the forecast is incomplete.
It does not show storm surge. Surge depends on the angle of approach, the shape of the coast and the state of the tide, and it is carried in its own bulletin. On the Philippine eastern seaboard it is frequently the hazard that kills, and it is invisible on a track map.
It does not show rainfall. A slow, weak system over mountains does more damage than a fast strong one over open water, and the track map cannot distinguish them. This site has covered how the rainfall warning colours work and why they are local for exactly this reason.
It does not show timing precision beyond the issue schedule. The centre's arrival time carries error in the same way its position does, and a twelve-hour slip at day three is ordinary. Plans built on an exact hour three days out are built on something the product does not claim to provide.
Official Philippine forecasts, bulletins and the five-day track are published by PAGASA. Regional track guidance in English is published by the Japan Meteorological Agency, and the two do not always agree — when they differ over Philippine territory, PAGASA is the authority.
Comments
Post a Comment