A Gulf coastal city seen through morning atmospheric haze

AIR QUALITY DATA CONFIDENCE

Know where your air quality map is supported — and where it is guessing.

A grid of observation age, satellite coverage fraction, nearby ground-station count and uncertainty flags for each map cell. Helps air quality teams distinguish well-supported estimates from sparse-observation zones before publishing public air quality maps.

AVAILABLE · GET ENDPOINTSee how it works ↓

Illustrative solar and atmospheric imagery

THE DECISION IT SUPPORTS

Which cells in the air quality map can we publish confidently — and which need a caution flag?

The confidence grid returns a machine-readable table of quality indicators per cell. Monitoring platforms can use these flags to automatically suppress or caveat low-confidence cells, protecting map users from acting on poorly supported estimates.

Built for

Air quality monitoring platforms
Public health and environment agencies
Smart city and urban planning teams
Research teams requiring provenance metadata

WHY TEAMS USE IT

Turn broad data into a clearer next step.

Each result is designed to narrow a decision, show the inputs behind it, and surface what must still be checked locally.

Flag cells before publishing

Each cell carries an overall_confidence level — high, medium or low — that can drive automatic display logic.

Expose observation age

Time-since-last-observation distinguishes real-time data from hours-old backfill.

Show satellite coverage quality

Cloud obscuration and orbit gaps affect satellite-derived estimates in predictable ways.

Surface ground-station support

Cells with no nearby stations cannot be validated and are marked explicitly.

FROM REGION TO SITE

See where each signal fits.

Images illustrate the operational setting. They are not live observations; use source timestamps and local measurements for decisions.

Flag cells with stale measurements
Cells with obs > 6h are marked low-confidence on regional maps.

Coverage fraction per grid cell
Cells with < 70% satellite coverage receive automatic uncertainty flags.

How many stations support each cell?
Zero-station cells cannot be validated and are clearly marked.

THE INFORMATION STACK

What goes in—and what it helps explain.

Availability is stated for every input. Missing site telemetry, irradiance readings, or atmospheric observations are identified rather than inferred.

INFORMATIONSOURCEHOW IT IS USEDSTATUS

01resolution_km (10–100)Query parameterGrid cell size. Default 25 km.OPTIONAL

02lat_min, lat_maxQuery parameterBounding box latitude filter.OPTIONAL

03lon_min, lon_maxQuery parameterBounding box longitude filter.OPTIONAL

A PRACTICAL WORKFLOW

From input to reviewable decision.

The workflow keeps model context, field evidence and human approval in their proper places.

01
Call GET with region or bbox
Omit bbox for the full Middle East grid.
02
Read cell confidence levels
overall_confidence = high / medium / low drives display logic.
03
Check observation age
obs_age_hours > 6 triggers automatic medium or low downgrade.
04
Suppress or caveat low cells
Remove or add visual uncertainty indicators before publishing.
05
Track coverage trends over time
Archive confidence grids to monitor sensor network health.

HOW INFORMATION BECOMES USEFUL

Examples of decisions it can inform.

These are screening patterns, not automatic instructions. Operational thresholds and local advisories remain authoritative.

SCENARIO 01

Cell with no ground stations nearby

Signal: overall_confidence = low, pm25_uncertainty_flag = high.
Useful action: Do not publish a PM2.5 value. Show a data-gap indicator.
Check before acting: Satellite-only estimates can still identify broad event boundaries.

SCENARIO 02

Observation age > 6 hours after a dust event

Signal: medium confidence despite good coverage.
Useful action: Mark the cell as "updating" and withhold PM2.5 display.
Check before acting: Check whether satellite coverage was cloud-blocked during the event.

SCENARIO 03

Multiple high-confidence cells around an industrial zone

Signal: Strong station network and recent satellite pass.
Useful action: Publish with full confidence metadata visible to users.
Check before acting: Still validate PM2.5 against calibrated sensor readings.

FOR PRODUCT TEAMS

A transparent API contract.

The response exposes gaps and limitations so an application can explain confidence instead of hiding it.

POST/api/v1/air-quality/middle-east/confidence-grid

Example request

GET /api/v1/air-quality/middle-east/confidence-grid
  ?resolution_km=25
  &lat_min=23
  &lat_max=26
  &lon_min=54
  &lon_max=56

JSONdecision response

Example response

{
  "cell_count": 4,
  "cells": [{
    "cell_id": "ME_N1",
    "label": "Abu Dhabi & Coast",
    "obs_age_hours": 2.1,
    "satellite_coverage_percent": 94,
    "ground_stations_within_200km": 5,
    "pm25_uncertainty_flag": "low",
    "overall_confidence": "high"
  }]
}

CAPABILITY BOUNDARIES

Useful now. Honest about what comes next.

Supported now

Observation age per cell
Satellite coverage percentage
Ground-station proximity count
PM2.5 uncertainty flag
Overall confidence tier (high/medium/low)
Bounding box filter

Additional data required

Calibrated ground PM2.5 sensors for validation
CAMS credentialed aerosol retrievals
Station metadata with calibration dates
Vertical profile data for boundary-layer correction

BUILD RESPONSIBLY

Use the result to focus attention—not replace field judgment.

Pair AtmosGrid’s regional weather context with local irradiance sensors, site telemetry and operator review.

Read the API documentation ↗Open the Middle East viewer ↗Explore all solar products ↗