Ocean Colour from Space: How Satellites Read the Sea

Look at the ocean from orbit in the right wavelengths, and it stops being anonymous blue. Swirls of emerald trace phytoplankton blooms the size of countries; milky turquoise patches mark coccolithophore armour drifting in the current; brown tongues of sediment pour from river mouths; and along temperate coasts, thin dark fringes appear that are, remarkably, kelp forest canopies. This is ocean colour remote sensing — the science of turning the faint light that leaves the sea into maps of life, chemistry and habitat. It underpins everything from global chlorophyll estimates to Tidewright's own Canopy Ledger monitoring. Here is how it works.

What "ocean colour" actually is

A colour satellite doesn't photograph the ocean the way your phone does. It measures spectral radiance: how much light, in narrow wavelength bands, arrives at the sensor from each patch of the planet. Of the photons reaching a sensor above the ocean, the overwhelming majority — typically around ninety percent — never touched the water at all. They were scattered by air molecules, haze, aerosols and clouds. The signal scientists want is the minority: water-leaving radiance, light that entered the sea, scattered off things in it, and came back out.

Extracting that sliver is called atmospheric correction, and it is one of the hardest quiet triumphs of remote sensing. Subtract the atmosphere accurately and the ocean's true colour appears; get it slightly wrong and you map smog and call it plankton.

Once you have water-leaving radiance across several wavelengths, the colour tells a story, because different things in the water absorb and scatter light differently:

• Phytoplankton pigments, chiefly chlorophyll-a, absorb blue and red light and scatter green — so where cells are dense, blue drops and green rises ("green water").
• CDOM (coloured dissolved organic matter — the tannins and humic tea of rivers and wetlands) absorbs blue strongly, darkening coastal water.
• Suspended sediment scatters broadly, brightening and whitening the water across the spectrum.
• Coccolithophores, shedding their calcium-carbonate plates, scatter light into that distinctive milky turquoise.
• Kelp canopies at the surface behave like vegetation: they strongly reflect near-infrared light, which water barely reflects at all — the key to mapping them from orbit.

From colour to chlorophyll: the algorithms

The workhorse product of ocean colour science is surface chlorophyll-a concentration, used as a proxy for phytoplankton biomass. The classic approach is a band-ratio algorithm: compare reflectance in a blue band (around 443 nm) with a green band (around 550–555 nm). The blue-to-green ratio falls as chlorophyll rises, and decades of paired satellite-and-ship measurements have turned that relationship into global algorithms.

The limits matter as much as the power:

• Case 1 vs Case 2 waters. In the open ocean ("Case 1"), what colours the water is almost entirely phytoplankton and their debris, so algorithms behave well. In optically complex coastal water ("Case 2") — which includes nearly every kelp coast — CDOM and sediment also colour the water, and naive chlorophyll algorithms can be badly fooled. Coastal products use alternative bands (red-edge, fluorescence) and regional tuning.
• It sees only the first optical depth. Satellite colour samples the upper metres of the water column. A deep chlorophyll maximum at forty metres is invisible from space. That is why satellites pair with profiling tools like Argo floats.
• Clouds, glint and low light. Clouds block the view; sun glint contaminates it; at high latitudes in winter the sun angle is too low to work at all. Satellite chlorophyll is a fair-weather, daytime measurement stitched into long records.

Newer hyperspectral sensors — measuring dozens to hundreds of narrow bands instead of a handful — can go further, distinguishing pigment groups and even hinting at which phytoplankton communities are blooming, from iron-hungry diatoms to calcifiers.

Mapping kelp from orbit — the near-infrared trick

Kelp is a different problem from plankton. A giant kelp canopy floats at the surface in a thin, discontinuous fringe along the coast — far too fine for one-kilometre ocean colour sensors. Mapping it uses high-resolution multispectral satellites (Landsat-class and better, at tens of metres per pixel) and a beautiful quirk of physics: liquid water absorbs near-infrared almost completely, while living canopy tissue — like terrestrial leaves — reflects it strongly. A pixel that glows in the near-infrared in the middle of a dark sea is very likely kelp at the surface.

Using the four-decade Landsat archive, researchers have reconstructed canopy extent along coasts such as California's for decades back, revealing how forests pulse with El Niño, marine heatwaves and recovery. It is now one of the foundational records of kelp change worldwide — we unpack the method in how satellites see kelp.

The caveats are real: only canopy that reaches the surface is visible; submerged understory kelp, young plants and deep forests on clear reefs can vanish from the satellite record entirely. Tide state, waves, sun angle and image frequency all matter, and canopy area says little about biomass or health beneath the surface. That is why Canopy Ledger fuses satellite canopies with drone surveys, in-water transects and sonar — space for coverage, divers for truth.

A short history of ocean colour sensors

The field runs on a relay of instruments:

• CZCS (1978–1986) — the proof of concept. It showed that phytoplankton blooms, eddies and fronts were visible from space in pigment, transforming biological oceanography.
• SeaWiFS (1997–2010) — the workhorse that built a continuous global chlorophyll record.
• MODIS and VIIRS — broader-spectrum sensors adding fluorescence, sediment and sea-surface temperature on shared platforms.
• MERIS and OLCI (Sentinel-3) — European sensors with bands tuned for coastal and Case 2 waters.
• PACE (launched 2024) — the first global hyperspectral ocean colour mission, opening the door to mapping plankton community composition from space.

Meanwhile, high-resolution multispectral constellations (Landsat, Sentinel-2) carry the coastal-canopy and habitat mapping load, and commercial fleets now image coasts daily.

Validation: why ships still matter

Satellite ocean colour is a model of the ocean, not a direct census, and it is kept honest by in-situ validation: ships and moored buoys measuring chlorophyll, absorption and scattering in the water at the same time a satellite passes overhead ("match-ups"), plus dedicated marine optical buoys that continuously measure water-leaving radiance as a reference. Algorithm updates and sensor cross-calibration — stitching CZCS to SeaWiFS to MODIS to OLCI without artificial jumps — are a discipline of their own.

What ocean colour is for

• Tracking productivity: chlorophyll feeds primary-production models that estimate how much carbon marine photosynthesis fixes each year — context for every blue-carbon calculation.
• Blooms and hazards: monitoring harmful algal blooms, coccolithophore events and river plumes after storms.
• Fisheries: locating fronts and productive water where fish aggregate.
• Habitat mapping: canopy kelp extents, seagrass in clear shallow water, and coastal change over decades.
• Restoration monitoring: exactly the stack Canopy Ledger uses — satellite canopies for hectares of coverage, drones for metre-scale detail, eDNA and diver transects for ecological truth.

Key takeaways

• Ocean colour satellites measure water-leaving radiance — about ten percent of the signal at the sensor — by subtracting atmospheric scatter.
• Band-ratio algorithms convert blue-to-green colour shifts into surface chlorophyll-a, a proxy for phytoplankton biomass.
• Coastal "Case 2" waters and the fact that satellites see only the upper metres mean coastal products need regional tuning and ship-based validation.
• Kelp canopies are mapped with high-resolution sensors exploiting strong near-infrared reflectance — but only surface canopy is visible, so in-water surveys remain essential.
• New hyperspectral missions are moving ocean colour from "how much plankton" toward "which plankton".
• Ocean colour, fused with drones, sonar and biology, is the backbone of how Tidewright's Canopy Ledger monitors restoration at scale.

FAQ

Can satellites see kelp forests from space?

Yes — where kelp forms a surface canopy (like giant kelp and bull kelp), high-resolution satellites detect it through its strong near-infrared reflectance against dark water. Submerged or understory kelp can't be seen, so satellite maps are combined with drones, sonar and diver surveys for a full picture.

How do satellites measure chlorophyll?

Chlorophyll pigments absorb blue light and scatter green, so algorithms compare blue and green reflectance bands: a lower blue-to-green ratio means more chlorophyll. Coastal waters with sediment and dissolved organic matter need special coastal algorithms to avoid confusion.

What is atmospheric correction?

Roughly ninety percent of the light reaching an ocean-viewing sensor comes from the atmosphere, not the sea. Atmospheric correction models and removes that haze of scattered light to recover the small water-leaving signal — it is the single most delicate step in ocean colour science.

Why can't ocean colour satellites see the whole water column?

Light is absorbed and scattered rapidly by water, so the signal from space comes only from the top few metres to tens of metres — the "first optical depth". Deep chlorophyll maxima, deep scattering layers and bottom habitats are invisible from orbit and need profiling floats, gliders and acoustics.

What was the PACE mission?

Launched in 2024, PACE carries the first global hyperspectral ocean colour instrument, measuring hundreds of narrow wavelengths. Where older sensors estimated bulk chlorophyll, hyperspectral data can begin to distinguish plankton types — diatoms from coccolithophores, for example — from orbit.