Marine Debris Research at the Great Pacific Garbage Patch

Featured Project

Marine Debris Research at the Great Pacific Garbage Patch

In 2021 and 2022, ForeSight Drones partnered with Oceans Unmanned, 501(c)(3) based in Santa Barbara, CA. The mission was to fly a ship-launched 6’ wingspan fixed-wing amphibious UAV to collect marine debris data at the Great Pacific Garbage Patch. The goal was to illuminate the distribution and concentration of marine debris to benefit the clean-up efforts of The Ocean Cleanup, NGO based in Rotterdam, The Netherlands.

Results

Between two 6-week campaigns, 34 flights were conducted, collecting ~27,000 images, and detecting 3000 individual pieces of marine debris. Images were post-processed using a machine-learning based object detection model, and mapped in GIS to analyze size classification and debris densities. This novel approach to aerial marine debris mapping proved effective for collecting large amounts of data to fill the scale-gap between remotely sensed satellite imagery, and trawl-based sampling, while also illuminating the challenges present in operating hand-launched, amphibious belly-landing UAS from an off-shore work vessel.

Services

Aerial Mapping,
GIS Analysis

The Ocean Cleanup,
Oceans Unmanned

Client

2021 & 2022

Project Date

A large white fixed-wing drone (UAV) sits on the deck of a ship in the ocean at the great pacific garbage patch.

Two drone pilots prepare to launch a large fixed-wing drone (UAV) over the railing of a ship at sea at the great pacific garbage patch.

Aerial image taken from a fixed-wing drone (UAV/UAS) showing the underside of the drone and looking backwards down towards two ships in the ocean.

Screenshot of Mission Planner ground control station (GCS) software of an in-progress flight showing the HUD, key parameters, and the flight plan of the drone.

An aerial shot looking straight down at the blue ocean with the bow of a large ship visible at the edge of the photo.

Three crew in a small workboat, one of which is holding a large white fixed-wing drone (UAV/UAS) are suspended from a cable alongside a ship at sea.

A collection of 12 screenshots of individual plastic debris detections that have been identified and placed in a bounding box by a machine learning convolutional neural network (CNN, AI).

A large pile of plastic debris sits on the deck of a ship after being hauled out of the great pacific garbage patch (GPGP).

A map of one flight from the Great Pacific Garbage Patch showing red dots representing plastic debris detections found in aerial images collected with a drone (UAV/UAS).

Looking out at a ship near the horizon on the ocean at the great pacific garbage patch at dusk with an orange and pink glow on the cloudy sky.

Technical Presentation

ForeSight founder, Nick Wagner, presenting at the PNAMP Remote Sensing Forum, December 2021.

Advance to t=46:00 to begin presentation.

“The UAV campaign results provide a dataset that sparks new insights into the plastic density in the North Pacific subtropical gyre, and that can aid in refining advection models specifically for floating macroplastic (> 50 cm).“

Robin de Vries, Geospatial Analyst, The Ocean Cleanup

Short video available on Twitter/X.

FEATURED CASE STUDY