The success of satellite radiometry missions and development of biogeochemical algorithms heavily rely on the availability of optical field measurements for validation. The NASA Ocean Biology Processing Group (OBPG) maintains a local repository of in-situ oceanographic and atmospheric data known as the SeaWiFS Bio-optical Archive and Storage System (SeaBASS).
Data are traditionally sourced from satellite observations and field campaigns conducted at various oceanographic stations. These data are primarily collected by NASA-funded researchers, such as the PACE Validation Science Team (PVST) voluntary submissions from the local and international scientific community, and collaborators, as well as external sources like the AErosol RObotic NETwork (AERONET) and AERONET ocean color (AERONET-OC). Unlike traditional ship-based sampling, AERONET-OC has provided a high volume of reliable, quality controlled, and systematic validation radiometric data since the 2000s.
Validation results are provided for the main ocean color and aerosol products, such as Remote Sensing Reflectance, chlorophyll-a concentrations, and aerosol optical depth. As the validation data sources increase, other data products will be added, such as phytoplankton community composition and cloud particle effective radius.
Users can download the validation match-ups and visualize their statistical metrics using Bland-Altman and scatter plots. Bland-Altman plots provide insights into the bias and precision of the satellite measurements compared to field measurements. A mean difference close to zero indicates good agreement, while the spread of differences (limits of agreement) contextualizes the bias within the variability of the field data. Additionally, a check assesses the scale dependency of the bias, identifying whether errors increase as the magnitude of the observations increases. If a scale dependency exists, the limits of agreement are replaced with a regression line showing its direction and magnitude. Scatter plots complement Bland-Altman plots by illustrating the strength of the linear relationship between the two datasets. High correlation coefficients and low RMSE values indicate strong agreement and high accuracy of the satellite-derived measurements. The scatter plots also include the baseline uncertainty requirements for the PACE mission.
Where and when were matchup data collected?
Where and when were matchup data collected?
Click any marker to learn more.