Detecting Microplastics Pollution in World Oceans Using Sar Remote Sensing

Abstract

Plastic pollution in the world's oceans is estimated to have reached 270.000 tones, or 5.25 trillion pieces. This plastic is now ubiquitous, however due to ocean circulation patterns, it accumulates in the ocean gyres, creating "garbage patches". This plastic debris is colonized by microbes and microorganisms that create unique ecosystems of sea-slicks and microbial bio-films. Microbial colonization is the first step towards disintegration and degradation of plastic materials: a process that releases metabolic byproducts from energy synthesis. These by-products include the release of short-chain and more complex carbon molecules in the form of surfactants, which we hypothesize will affect the fluid dynamic properties of waves (change in viscosity and surface tension) and make them detectable by SAR sensor. In this study we used Sentinel-1A and COSMO-SkyMed SAR images in selected sites of both the North Pacific and North Atlantic oceans, close to ocean gyres and away from coastal interference. Together with SAR processing we conducted contextual analysis, using ocean geophysical products of the sea surface temperature, surface wind, chlorophyll, wave heights and wave spectrum of the ocean surface. In addition, we started experiments under controlled conditions to test the behavior of microbes colonizing the two most common pollutants, polyethylene (PE) and polyethylene terephthalate (PET) microplastics. The analysis of SAR images had shown that the speed of surface wind supplemented by accumulation mechanism of Langmuir cells is found to be a main meteorological factor influencing the distinct appearance of the sea-slicks and microbial biofilms. The preliminary conclusion of our study is that SAR remote sensing may be able to detect plastic pollution in the open oceans and this method can be extended to other areas.

Investigating Detection of Floating Plastic Litter from Space Using Sentinel-2 Imagery

“…Other approximations estimate that the oceans may already contain more than 150 million tons of plastic [22]; around 250,000 tons of these contaminants is fragmented into 5 trillion plastic pieces, which may be floating on the oceans' surface [23]. It has also been calculated that every year, between 4.8 and 12.7 million tons of plastic find their way into the ocean from coastal populations worldwide [16], while the Ellen Macarthur sensing methods to identify plastics in the sea [29,[68][69][70][71][72][73]. Satellite images can be used to identify plastics in the water, such as Sentinel-1A and COSMO-Sky-Med Sar images [74], C-Band Radarsat-1 SAR images [75] as well as Landsat TM and EMT+ satellite images [76][77][78].…”

Indoor spectroradiometric characterization of plastic litters commonly polluting the Mediterranean Sea: toward the application of multispectral imagery

“…The plastics litter release the surfactants in the sea during their degradation. These change the fluid-dynamic propriety of the sea that could be detected from the COSMO-SkyMed SAR image 27 but, before to effectively use remote sensing technology for monitoring, it is necessary to perform further investigation. Currently, it is possible to identify with satellite image the large plastics litter but is difficult to identify smaller plastics (debris) such as microplastics or nanoplastics.…”

Critical Review of Processing and Classification Techniques for Images and Spectra in Microplastic Research

“…The machine learning-based detection of micro-and nano-plastic particles in various types of images from optical to electron microscopies is in its infancy. Davaasuren et al 170 used a machine learning model to interpret synthetic aperture radar image data based on experimentally derived proxies of microplastic hotspots. However, it is unlikely that microplastics will be detectable using satellite-based remote sensing products soon because of low spatial resolution.…”