Marine debris poses a growing threat to the environment. Monitoring from remote sensing is promising but there are limitations due to the diversity of marine debris. The open-access MADLib collection, with 24889 hyperspectral reflectances from 3032 diverse debris samples, is anticipated to support algorithm and sensor development based on well-curated representative spectra of marine debris. We also discuss gaps and propose improved metadata schemes for expanding the living MADLIB collection.

These high-quality data document a harmful algal bloom dominated by Nodularia spumigena, a cyanobacterium that has been recurring in waters around the world, using advanced water observation technologies. We also showcase the benefits of experiments of opportunity and the issues with obtaining synoptic spatio-temporal data for monitoring water quality. The dataset can be leveraged to gain more knowledge on related blooms, develop detection algorithms and optimize future monitoring efforts.

This paper describes a dataset consisting of 47 hyperspectral-reflectance measurements of plastic litter samples. The plastic litter samples include virgin and real samples. They were measured in dry conditions, and a selection of the samples were also measured in wet conditions and submerged in a water tank. The dataset can be used to better understand the effect of water absorption on the plastics and develop algorithms to detect and characterize marine plastics.

Technologies to support detection and tracking of plastic litter in aquatic environments capable of repeated observations at a wide-area scale have been getting increased interest from scientists and stakeholders. We report findings about thermal infrared optical properties of naturally dried samples of algae, sands, sea shells and synthetic plastics obtained in Chile. Diagnostic features of the dataset are foreseen to contribute towards research relevant in thermal infrared sensing of plastics.