How we score a house reef.
One method, applied to every reef in the country the same way, from data anyone can check. Where we can’t measure something, we say so instead of guessing.
Why there are two names.
They are not the same thing, and we would rather that were obvious than tidy.
Reef Conditions
The three things a satellite can honestly measure: reef cover, reef structure and shore access. Complete within its own scope, comparable across the whole country from day one, and every number shows its measurement.
The Maldives Index
The full picture — Reef Conditions plus fish life and big-animal encounters, confirmed in the water. No property has earned it yet. It appears only when there is real in-water or first-party evidence behind it, and never before.
The temptation is to publish one number and let people assume it means everything. We would rather show a smaller number we can defend, and be plain that the bigger one has to be earned.
Three things we can measure.
Reef Conditions is the mean of three sub-scores, each 0–10, computed for 297 reef sites covering all 1,037 registered properties. There are no hidden weights: the three count equally. The national median is 6.
| Sub-score | What it measures | Measured from |
|---|---|---|
| Reef cover (modelled) | Share of the mapped reef classed as Coral/Algae, reduced for heat stress accrued since the imagery was taken. | Allen Coral Atlas benthic map + NOAA Coral Reef Watch |
| Reef structure | Reef-slope share (the wall), whether a reef crest is present, and the reef’s overall extent. | Allen Coral Atlas geomorphic map |
| Accessibility & ease | Metres from the property to the reef slope, and how much shallow reef flat must be crossed to reach it. | Allen Coral Atlas geomorphic map + property coordinates |
Every number on a reef page shows the measurement behind it — the actual coral/algae percentage, the reef-slope area, the metres to the drop-off. You should not have to take our word for any of it.
Two things we refuse to guess.
Most reef ratings quietly invent these. We would rather publish a shorter, truer score — so they read Pending until real in-water or first-party evidence exists.
| Not scored | Why not |
|---|---|
| Fish diversity & abundance | A satellite cannot count fish. Open species records (GBIF, OBIS, iNaturalist) track where divers point cameras, not where fish live — a 5 km box around one island held 1 record, another 1,595, with no relationship to reef quality. Using that would rank resorts by whether someone visited with an app open. |
| Big encounters | Same problem, worse: manta and whale-shark records cluster at two famous aggregation sites, so a proximity score would just re-encode “is this near Hanifaru or South Ari”. |
They will be filled by the only routes we consider legitimate: reviews submitted directly to us by visitors, logged sightings from dive operators, and in-water surveys. We do not mine other platforms’ reviews to build a rating — their terms forbid it, and a rating built that way could be switched off by someone else’s lawyer.
What this score is not.
It is not a measurement of live coral
The Allen Coral Atlas benthic class is “Coral/Algae” combined — a satellite cannot separate live coral from algae. A thriving reef and an algae-smothered dead one can look identical from orbit. So “reef cover” is modelled hard-bottom cover, not coral health.
The imagery predates the 2024 bleaching
Allen Coral Atlas imagery dates from 2018-2021. The 2024 global bleaching event came after it. We correct for this by subtracting a penalty based on heat stress accrued since — but a penalty is an estimate, not a new photograph of the reef.
Nobody has been in the water
Every score here is Modelled. No reef in this index has been surveyed in person. When one is, it will carry a date and a named surveyor — and until then we will not imply otherwise.
We score reefs, not doorsteps
Resorts are one island per property, so their score is their house reef. Guesthouses cluster on inhabited islands and genuinely share one reef — so they share one score, and we say so rather than invent a private reef for each.
The trend arrow.
Not a guess about the future — a real trend from NOAA’s thermal record. We compare mean peak annual heat stress (Degree Heating Weeks) over the last five years against the five before it. Rising heat means a reef under pressure, and it shows here before it shows in the cover.
▲ Improving → Stable ▼ Declining
Reading the headline.
| Score | Band |
|---|---|
| 9.0+ | Outstanding |
| 8.0–8.9 | Exceptional |
| 7.0–7.9 | Excellent |
| 6.0–6.9 | Very Good |
| 5.0–5.9 | Good |
| 4.0–4.9 | Average |
| < 4.0 | Below Average |
Where the data comes from.
Two open sources, both free to use commercially, both citable by you as easily as by us.
- Allen Coral Atlas — benthic habitat and geomorphic zonation maps, 5 m resolution, licensed CC BY 4.0. Allen Coral Atlas (2020). Imagery, maps and monitoring of the world’s tropical coral reefs. Zenodo. doi:10.5281/zenodo.3833242
- NOAA Coral Reef Watch — CoralTemp v3.1 daily 5 km heat-stress record, US Government public domain. NOAA Coral Reef Watch. 2019, updated daily. NOAA Coral Reef Watch Version 3.1 Daily 5km Satellite Coral Bleaching Heat Stress Product Suite. College Park, Maryland, USA.
- Maldives Ministry of Tourism registry — the property list itself: every registered resort, hotel and guesthouse island.
Read the raw dataset, with the evidence behind every number (JSON) →
Generated 2026-07-18. Ratings are never for sale; we earn referrals only, and a referral has never moved a score.