This is an in-situ dataset of estimates of particulate organic carbon (POC) based on all the current (2021-09-22) profiles of optical backscattering (BBP) collected by (BGC)Biogeochemical-Argo floats. The dataset spans from 2010-06-01 to 2021-09-22 and covers the upper 2000 dbars of the water column (continuous profiles have been binned in 41 vertical bins). The dataset was produced by first devising a new set of automatic tests to quality control the large BBP dataset available (>31M records over >130k profiles). The QCed BBP was then converted into POC using an empirical algorithm (average of the POC: BBP slopes of Cetinic et al., 2012)

The BICEP/NCEO: Monthly global Particulate Organic Carbon (POC) v5 datasets contain POC concentrations (mg m^-3) with per pixel uncertainties estimates gridded on both geographic and sinusoidal projections at 4 km spatial resolution for the period of 1997 to 2020. The POC products were generated as part of the European Space Agency (ESA) Biological Pump and Carbon Exchange Processes (BICEP) project with support from the National Centre of Earth Observation (NCEO). The POC datasets have been produced by using a modified empirical band ratio algorithm by Stramski et al. (2008): 292*Rrs(490)/Rrs(560)^-1.49. Additional variables that were used for the calculation of the POC products are also provided in the datasets, including the Remote Sensing Reflectance (Rrs) at 490 nm and 560 nm obtained from the ESA Ocean Colour Climate Change Initiative version 5 dataset (OC-CCI v5). For more details on the algorithm and its validation, please see the BICEP Algorithm Theoretical Basis Document (ATBD) and validation report (https://bicep-project.org/Home). A related dataset based on the ESA Ocean Colour Climate Change Initiative v4.2 data is also available (see link in the related records section).

The BICEP/NCEO: Monthly global Particulate Organic Carbon (POC) v4.2 datasets contain POC concentrations (mg m^-3) with per pixel uncertainties estimates gridded on both geographic and sinusoidal projections at 4 km spatial resolution for the period of 1997 to 2020. The POC products were generated as part of the European Space Agency (ESA) Biological Pump and Carbon Exchange Processes (BICEP) project with support from the National Centre of Earth Observation (NCEO). The POC concentrations were estimated using an empirical Remote Sensing Reflectance (Rrs) band ratio algorithm by Stramski et al. (2008): 203.2*Rrs(443)/Rrs(555)^-1.034. This algorithm has shown a relatively good performance in the recent global inter-comparison study conducted by Evers-King et al. (2017). Additional variables that were used for the calculation of the POC products are also provided in the datasets, including the Rrs at 443 nm and 555 nm obtained from the ESA Ocean Colour Climate Change Initiative version 4.2 dataset (OC-CCI v4.2)(Sathyendranath et al., 2020). In addition to the papers by Stramski et al. (2008) and Evers-king et al. (2017), for more details on the algorithm and its validation, please see the BICEP Algorithm Theoretical Basis Document (ATBD) and validation report (https://bicep-project.org/Home). This version of the dataset is an updated version of the previous 'NCEO: Monthly global Particulate Organic Carbon (POC) (produced from the Ocean Colour Climate Change Initiative, Version 4.2 dataset)'. A related product based on the Ocean Colour Climate Change Initiative v5.0 data is also available (see the link in the related records section).

The National Centre for Earth Observation (NCEO): Monthly global Particulate Organic Carbon (POC) dataset contains POC concentrations gridded on both sinusoidal (SIN) and geographic (GEO) grid projections at 4 km spatial resolution for 1997-2020. The POC dataset has been produced using the Ocean Colour Climate Change Initiative Remote Sensing Reflectance (Rrs) products, Version 4.2. The dataset includes the Rrs at 443 nm and 555 nm with pixel-by-pixel uncertainty estimates for each wavelength. For more details on the algorithm and its validation, please see papers by Stramski et al. (2008) and Evers-King et al. (2017). Please note that the validation of the POC algorithm is a continuing process. To increase the accuracy of POC algorithms, further in situ POC data need to be collected with high spatial and temporal resolution.