Remote Sensing NPP

• Page authors: PRIMO
• Responsible curator: Hagen Buck-Wiese

---

Remote sensing NPP
Approach: satellite ocean colour and NPP models
Context: remote sensing (satellite)
Spatial scale: 0.5–4 km pixel; global coverage
Temporal scale: days to years
Units: mg C m-2 d-1
Community captured: bulk
Co-measurements: remote sensing reflectance (Rrs), PAR, sea surface temperature, mixed layer depth

Method Overview

Satellite ocean colour sensors (e.g., MODIS, SeaWiFS, OLCI) measure the spectral radiance leaving the ocean surface in multiple visible and near-infrared bands. Bio-optical algorithms convert remote sensing reflectance (Rrs) into chlorophyll-a concentration and, in some models, the backscattering coefficient as a proxy for phytoplankton carbon (C_phyto). NPP models then combine these biological proxies with satellite-derived PAR, sea surface temperature, and mixed layer depth estimates to calculate depth-integrated NPP.

Major NPP model classes include the Vertically Generalised Production Model (VGPM; chlorophyll-based), the Carbon-based Productivity Model (CbPM; phytoplankton carbon-based), and absorption-based models. Each uses a different formulation of the relationship between biomass, physiology, and carbon fixation.

Scale of measurement

Satellite pixels are typically 0.5–4 km; global coverage is achieved every 1–4 days depending on orbit and cloud cover. The optical signal originates from the surface euphotic layer only (approximately the first attenuation length). Depth integration of NPP is achieved through models rather than direct measurement.

Data generated

Depth-integrated NPP in mg C m^-2 d^-1; time series enable seasonal and interannual trend analysis at regional to global scales.

Units & currency

Units are mg C m^-2 d^-1. The currency is carbon.

Sample size

No sample collection required; the satellite measures upwelling radiance over a pixel footprint of km².

Repositories & databases

Global satellite NPP products are distributed by NASA Ocean Color: oceancolor.gsfc.nasa.gov.

Limitations

Satellite reflectance is limited to the optically accessible surface layer and cannot directly sense NPP below the first attenuation depth. Cloud cover creates temporal gaps requiring interpolation. Algorithm uncertainties in Rrs-to-chlorophyll conversion, the chlorophyll-to-carbon fixation relationship, and NPP model structure propagate into NPP estimates. Cross-mission biases require careful inter-calibration. Validation against direct measurements (e.g., ¹⁴C) shows large regional uncertainties in models.

Example Applications & Protocols

Classic examples

• Westberry et al. (2023) Gross and net primary production in the global ocean: an ocean color remote sensing perspective

Common calculations/conversions

• NPP (mg C m^-2 d^-1) = ∫ C_phyto(z) × µ(z) dz; where µ is the phytoplankton-specific growth rate modelled from temperature and irradiance.