Pollution data for the world's oceans exists — but it's scattered across a dozen agencies, in different formats, at different scales. Nobody had stitched it together into a single comparable score per water body.
This project does exactly that. Twelve public datasets, cleaned and merged, combined into pollution factors weighted by scientific relevance, and turned into an index you can compare — and forecast to 2100.
35
Water bodies scored
12
Pollution factors
2100
Forecast horizon
The index covers major oceans split by hemisphere, regional seas (Mediterranean, Black Sea, Red Sea, South China Sea), and large lakes (Great Lakes, Caspian Sea, Lake Victoria, Lake Baikal). Every region gets a score from 0–100 built from real measurements.
Data Sources
What goes into the score
Every dataset is publicly available and free to use. Where regional data was missing from a source, values were filled using published peer-reviewed literature — documented per region in the codebase.
Marine Microplastics
NOAA NCEI
The most direct measure of plastic pollution in the water column
18%
River Plastic Input
Our World in Data
Plastic entering the ocean via rivers — the primary land-to-ocean pathway
15%
Dissolved Oxygen
NOAA World Ocean Atlas
Low oxygen marks dead zones where marine life cannot survive
10%
World Port Index
NGIA
Industrial coastal pressure and shipping activity
9%
Coastal Population
Copernicus / Zenodo
Population within 10km of shore — proxy for waste pressure
8%
Oil Spills
NOAA IncidentNews
Frequency and volume of recorded oil spill incidents per region
7%
Ocean pH
Copernicus Marine
Acidification from CO₂ absorption
7%
Sea Surface Temperature
NOAA World Ocean Atlas
Thermal stress driving bleaching and ecosystem collapse
7%
Wastewater & Runoff
FAO AQUASTAT
Municipal sewage and agricultural runoff per region
5%
Clean Water Score
Ocean Health Index
Independent measure of chemical, nutrient and pathogen contamination
5%
Biodiversity Score
Ocean Health Index
Health of marine species and habitats — the ultimate impact of pollution
5%
Plastic Mismanagement
OECD / Our World in Data
Share of plastic waste that is badly disposed of
4%
Methodology
How the index is calculated
Each of the twelve factors is normalised to a 0–100 scale using min-max scaling across all 35 regions, then combined using a weighted average:
Microplastic concentration
18%
River plastic input
15%
Dissolved oxygen depletion
10%
Port & shipping pressure
9%
Coastal population
8%
Oil spill pressure
7%
Ocean pH (acidification)
7%
Sea surface temperature
7%
Wastewater & runoff
5%
Clean water (OHI)
5%
Biodiversity (OHI)
5%
Plastic mismanagement
4%
Forecasting to 2100
Rather than applying one growth rate to every region, each water body is forecast using its own growth rate, blended from three independent real-world data sources:
55%
Microplastic trend
How fast plastic has actually risen in each region over 50 years of NOAA measurements (1972–2023), via log-linear regression.
25%
Population growth
The UN's real population projection for each region to 2100. Some regions grow; others shrink, reducing future pressure.
20%
Ocean warming
The IPCC's CMIP6 temperature projection (SSP2-4.5) for each ocean region. The Arctic warms fastest of all.
This means regions with rising plastic, growing population, and rapid warming climb steeply — while regions like the Mediterranean, with a declining measured plastic trend and shrinking coastal population, are actually projected to improve by 2100. The dashboard's policy slider then lets you explore how those trajectories bend under different levels of global action.
Limitations — stated honestly:
The factor weights are chosen by reasoning about relevance, not derived statistically — a common approach for composite indices, but a subjective one
Microplastic units vary across studies (items/m³, items/kg, items/km²) so normalisation across measurement methods is approximate
Oil spill data is strongest for US coastal waters (NOAA), so some regions are likely under-represented
Forecasts assume current trends continue with no major policy intervention — they are trajectories, not predictions
Some regions rely on published literature values rather than direct local measurements. Rather than hide this, every region carries a data confidence score — the share of its factors that come from direct measurement. Click any region on the dashboard to see it. Regions with little measured data (many lakes and smaller seas) are flagged as low-confidence, so their scores should be read as indicative rather than precise.
The Builder
Who made this
S
Sercan Emiroglu
BSc Computer Science · City St George's, University of London
I'm a Computer Science graduate based in London with a focus on data science, machine learning, and building things that are actually useful. I used to swim competitively, and the sea has always been my thing — so this project is personal as much as technical.
There's real data here telling a real story about where the world's oceans are headed, and I wanted to make it something anyone could actually see and understand.