09-10, 15:00–15:30 (America/Chicago), Grand H
This presentation highlights the importance of Sentinel-2 and other public remote sensing data for processing PlanetScope imagery, demonstrating the critical role of these datasets in enhancing the quality and reliability of Earth observation products.
Satellite images play a critical role in Earth observation because they provide valuable data for applications ranging from environmental monitoring to agricultural management. Dove fleet, operated by Planet Labs PBC, delivers medium-resolution, high-cadence imagery of the entire Earth’s landmass. To deliver PlanetScope imagery to their users, Planet processes the images through a pipeline built on publicly available remote sensing data. This presentation would focus on the importance of publicly available remote sensing to companies like Planet.
The spectral bands of SuperDove are intentionally designed to match those of Sentinel-2, allowing Sentinel-2 to serve as a reference for on-orbit calibration through simultaneous, intersecting crossovers. As the relative spectral responses between Sentinel 2 and SuperDove are very similar, they can collect crossover scenes from anywhere in the world to perform the on-orbit calibration and are not just limited to well characterized calibration sites where a spectral band adjustment factor can be calculated. Sentinel 2 data is vital to operational delivery of PlanetScope imagery to users.
Sentinel 2 is only one of many datasets that is used in the PlanetScope pipeline. In addition to radiometric calibration, Planet utilizes openly available data as reference data for rectification. The reference data is designed and built to support orthorectification of any type of Planet ortho scene on a global scale. This dataset covers the large majority of the global landmass with 2.5 m high resolution ALOS satellite imagery (Antarctica not covered, Greenland partially), while the continental US is still covered by very high-resolution (VHR) airborne imagery (NAIP) with resolution under 1 m. The vertical component of the reference data is derived from the Digital Elevation model with a post spacing under 30m globally. This dataset includes Copernicus 30 meter DEM, USG NED, Intermap World30 and others.
The above lists the basic mechanism in the PlanetScope pipeline to deliver a PlanetScope top of the atmosphere radiance product. Planet provides additional processing level, surface reflectance. PlanetScope surface reflectance product is derived from the standard radiance product and is processed to top-of-atmosphere reflectance and then atmospherically corrected to bottom-of-atmosphere reflectance. This product ensures consistency across localized atmospheric conditions, minimizing uncertainty in spectral response across time and location. When converting an image to surface reflectance, water vapor and ozone inputs are retrieved from MODIS near-real-time (NRT) data for same-day collects.The SuperDove fleet does not have atmospheric information on board, without MODIS data planet would not be able to deliver a surface reflectance product.
