River plume and estuarine productivity mapping for coastal fisheries

Riverine discharge plumes concentrate nutrients that drive coastal fish and invertebrate productivity, but their boundaries shift daily. Ocean-colour sensors can map chlorophyll and suspended sediment in these optically complex waters, within the limits of cloud cover and overpass timing.

Sensors

Sentinel-3 OLCI: 21 spectral bands from 400 to 1020 nm at 300 m spatial resolution, with a 2-day revisit at the equator. The band configuration and dedicated Case-2 water algorithms in the Copernicus ground segment make it the primary workhorse for estuarine chlorophyll and total suspended matter retrieval.
MODIS Aqua/Terra: 250 m to 1 km resolution depending on band, with combined morning and afternoon overpasses giving up to two looks per day. The archive runs from 2000 (Terra) and 2002 (Aqua), making it the deepest continuous ocean-colour record available for trend analysis in estuarine systems.
GOCI-II (Geostationary Ocean Color Imager-II): Operated by the Korea Institute of Ocean Science and Technology, GOCI-II acquires up to 10 images per day over East Asia at 250 m resolution. Geostationary repeat is the only current way to track intra-day plume dynamics; coverage is limited to its fixed disk centred on 130°E.
Landsat 8/9 OLI: 30 m multispectral resolution with a 16-day single-satellite revisit, halved to 8 days with both satellites operating together. At this resolution, small tidal creeks and narrow estuarine channels that are invisible to MODIS and OLCI become mappable, though the slow revisit is a constraint in rapidly evolving plume events.

Why the plume boundary is the fishing boundary

River discharge carries dissolved inorganic nitrogen, phosphorus and silica into coastal waters. Where that nutrient pulse meets sufficient light, phytoplankton bloom. Zooplankton follow. Juvenile fish and invertebrates aggregate in the resulting productive fringe, and so do the fishers who know where to look. The spatial position of that fringe is not fixed: it shifts with river flow, wind, tidal stage and coastal current direction, sometimes by tens of kilometres over a few days.

Satellite ocean-colour mapping does not replace a skipper's local knowledge, but it does provide a synoptic view no vessel survey can match. A single Sentinel-3 overpass covers an entire estuarine system in one acquisition, resolving the chlorophyll gradient from the turbid inner plume to the productive outer margin where light and nutrients are both available.

Case-2 waters make the optics hard

Open-ocean algorithms assume that the water-leaving radiance signal is dominated by phytoplankton pigments. Estuaries break that assumption. Suspended sediment, coloured dissolved organic matter (CDOM) from terrestrial runoff, and phytoplankton all absorb and scatter light simultaneously, and their optical contributions overlap. These are classified as Case-2 waters in the oceanographic literature, and they require dedicated retrieval algorithms, typically semi-analytical or neural-network approaches trained on in-situ optical data from comparable systems.

Sentinel-3 OLCI ships a Case-2 Regional processor (C2RCC) in the standard Copernicus toolbox, and MODIS data have been processed with similar approaches in published regional studies. Even so, retrieval uncertainty in highly turbid inner-estuarine zones can exceed 50% for chlorophyll-a concentration. The outer plume margin, where turbidity drops and the phytoplankton signal becomes cleaner, is generally retrieved with much higher confidence. Honest interpretation of these products requires knowing which part of the plume you are looking at.

What temporal resolution you actually get

Polar-orbiting sensors pass over a given estuary once or twice per day at best. Cloud cover in humid coastal environments can reduce usable acquisitions to a handful per month in wet seasons, precisely when river discharge, and therefore plume productivity, is at its peak. MODIS Aqua and Terra together offer morning and afternoon passes, which helps, but a single cloud deck can blank both.

GOCI-II is the only operational sensor that changes this equation, providing up to 10 acquisitions per day over East Asian coastal waters. For estuaries outside its disk, compositing strategies, typically 3-day or 8-day medians, are the practical answer. They smooth out tidal and meteorological noise but also obscure the episodic discharge events that often drive the most significant productivity pulses. Buyers should decide whether they need event-scale or seasonal-scale intelligence before specifying a monitoring programme.

Spatial resolution sets what you can see

MODIS at 1 km cannot resolve a 500 m wide tidal channel. Sentinel-3 OLCI at 300 m does better but still misses fine-scale features in complex delta systems. Landsat OLI at 30 m can map individual creek mouths and narrow estuarine arms, revealing productivity gradients that coarser sensors average away. The trade-off is the 8 to 16 day revisit, which means Landsat is better suited to mapping structural plume geometry over a season than to tracking a discharge event as it evolves.

For many fisheries management applications, this is a reasonable trade. Understanding which estuarine arms consistently produce the highest chlorophyll in the post-flood period, for example, is actionable information for gear deployment decisions even if it is derived from monthly composites rather than daily snapshots.

From spectral signal to fisheries intelligence

The analytic chain runs from atmospherically corrected reflectance to bio-optical retrieval to spatial product. Chlorophyll-a concentration maps, total suspended matter maps and CDOM absorption maps are the primary outputs. From these, secondary products can be derived: plume area and volume flux estimates, productivity anomaly indices relative to a multi-year baseline, and front detection identifying the productive outer plume boundary.

Satellize runs this chain on open Sentinel-3 and MODIS archives, applying regional bio-optical algorithms where published calibration data exist for the target system. The Tonga crop-estimation programme demonstrated that open-data analytics, properly calibrated to local conditions, can produce decision-grade outputs for resource-constrained clients. The same principle applies here: the satellite data are free; the value is in the processing, calibration and interpretation.

Fisheries agencies typically want outputs as GIS layers ingested into existing fleet management platforms, or as periodic reports correlating plume extent with catch-per-unit-effort data from logbooks. Both are achievable. What is not achievable, at present, is a cloud-free daily map of any specific estuary from polar-orbiting sensors alone.

Archive depth and what it enables

The MODIS Terra archive begins in February 2000, giving over two decades of estuarine ocean-colour data for trend analysis. Landsat adds further depth, with OLI data from 2013 and the broader Landsat archive extending to 1972 for structural change context. Sentinel-3 OLCI data are available from 2016.

Multi-year archives support questions that single-season monitoring cannot answer: Is the productive plume zone expanding or contracting as upstream land use changes? Are post-flood chlorophyll peaks arriving earlier or later than they did a decade ago? Has a new dam altered the seasonal discharge pulse that historically sustained a nearshore fishery? These are the questions that turn a monitoring programme into a management tool.

Typical figures

Spatial resolution (ocean colour)	300 m (Sentinel-3 OLCI); 250–1000 m (MODIS); 30 m (Landsat OLI)
Revisit cadence	1–2 days (MODIS Aqua+Terra combined); 2 days at equator (Sentinel-3); 8–16 days (Landsat 8+9); up to 10×/day (GOCI-II, East Asia only)
Usable acquisition rate (cloud-affected tropics)	Typically 3–8 cloud-free scenes per month in wet season; higher in dry season
Key spectral bands	Blue (443–490 nm), green (555–560 nm), red (665–670 nm), near-infrared (865 nm) for atmospheric correction; OLCI adds red-edge bands at 681 and 709 nm for Case-2 retrieval
Chlorophyll-a retrieval range	Approximately 0.1–100 mg m⁻³; uncertainty highest (>50%) in highly turbid inner-estuarine zones
Total suspended matter detection	Detectable from ~1 g m⁻³ in outer plume; saturation effects possible above ~100 g m⁻³ in near-field discharge zones
Archive depth	MODIS from 2000; Landsat OLI from 2013 (broader Landsat archive from 1972); Sentinel-3 OLCI from 2016
Typical data latency (near-real-time products)	3–6 hours for MODIS NRT; same-day for Sentinel-3 via Copernicus Online Data Access
Delivery formats	GeoTIFF, NetCDF, OGC-compliant WMS/WFS, tabular CSV summaries

Analytics Satellize can run

Chlorophyll-a concentration map	Semi-analytical or neural-network Case-2 retrieval (e.g. C2RCC) applied to atmospherically corrected Sentinel-3 OLCI or MODIS reflectance	GeoTIFF layer per acquisition or composited over 3/8/30-day windows, with uncertainty band estimate
Total suspended matter map	Empirical or semi-analytical retrieval using red and near-infrared bands calibrated to regional in-situ data where available	GeoTIFF layer showing suspended sediment concentration gradient across the plume
Plume extent delineation	Threshold or gradient-based classification on turbidity and chlorophyll retrievals to define inner, middle and outer plume zones	Polygon shapefile or GeoJSON boundary layer per scene, with area statistics in tabular report
Productive outer-plume front detection	Spatial gradient analysis on chlorophyll fields to locate the transition zone between light-limited inner plume and nutrient-limited offshore waters	Polyline GIS layer marking front position, suitable for overlay with vessel track data
Seasonal productivity anomaly index	Comparison of current chlorophyll composite against multi-year MODIS or Sentinel-3 climatological baseline for the same calendar period	Standardised anomaly map (z-score or percentile rank) delivered as GeoTIFF with accompanying narrative summary
Long-term plume trend analysis	Time-series regression on annual or seasonal chlorophyll and plume-area statistics extracted from the MODIS archive (2000 to present)	PDF report with trend charts, spatial maps of statistically significant change, and interpretation for fisheries management context
Discharge-event tracking composite	Near-real-time MODIS NRT processing triggered by river gauge threshold or precipitation event, producing rapid composites over 1–3 days post-event	Automated GeoTIFF delivery to client FTP or cloud bucket within 24 hours of trigger

Sources

Who does the work

We can get this done for you. Satellize runs its own analyst desk and a strong science team. You do not buy a data feed and work out what it means; our people source the imagery, run the analysis described on this page, and hand you the answer with its confidence limits stated. Discuss this requirement.