Video monitoring framework in support of sargassum management,

Abstract

The waters off Caribbean islands have seen large amounts of Sargassum seaweed during the last years (Wang and Hu, 2016). This record-breaking events of algae blooms and mass stranding started in earnest in 2011, then 2015 saw the next large-scale event and in January 2018, unusually large aggregations of Sargassum have been spotted on satellite imagery (Optical Oceanography Laboratory, n.d.). Normally, when floating offshore Sargassum provide important habitat and refuge to a large diversity of animals. However, while weeds approach nearshore and beach, due to currents and winds actions, in such massive quantities, they start to be deathtrap for many animals and contribute to the degradation of important coastal habitats, threatening coastal activities and ecosystems. The decomposing mass, which can be several meters high, creates oily slicks in its wake and releases a foul odor, damaging tourism activities since the sight and smell left beaches highly unappealing. Both satellite and modelled surface current data point to the North Equatorial Recirculation Region as the origin of recent mass blooms – north of the mouth of the Amazon, between Brazil and west Africa, in an area not previously associated with Sargassum growth (Gower et al., 2013). A number of factors including nutrients, rising sea temperatures and Sahara dust storms have been put forward as potential causes (Louime et al., 2017). Mathematical models developed to analyze satellite imagery and detect floating algae, the Floating Algae Index (FAI) (Hu, 2009), reveal that it is only in recent years that the area has seen the mass proliferation of Sargassum – satellite imagery from before 2011 shows the area to be ‘largely free of seaweed’. A number of strategies have been planned to deal with the large accumulations of algae resulting from the mass blooms. Removal and burial of the algae as soon as it gets stranded at beaches has been widely recommended (CRFM, 2016), although it appears to harm the environment (turtle habitat) and more specifically the sediments loss at beach. Nowadays, the information retrieved from remote sensing satellite system are very useful in order to estimate qualitatively and quantitatively the presence and motions of such macroalgae offshore. Unfortunately, operational warning devices able to anticipate algae washing ashore still have disadvantages related to the inadequate sampling and temporal frequency (MODIS observations e.g., Wang and Hu, 2017) and the interposing obstacles such as cloud shadows, sun glint constitute important issues; actually, there is little room for improvement as these are natural phenomena.Aside the need of anticipation, there is a need for the local government to deal with this severe issue and increase the abilities of quantifying Sargassum onshore on a seasonal basis. There is a special need for management planning in order to an increased resilience and benefit from Sargassum influxes (Cox, S., 2019). Location and amounts of Sargassum ashore can be then accurately evaluated to requirements. The purpose of this study is to implement a video-based framework approach in order to classify the coastal morphologies and specifically detect floating and beached algae in a very efficient and replicable manner with a moderate cost. This is conducted by means of a semi-supervised superpixel-based Deep Convolutional Neural Network (DCNN) classification of ground-sensed images. The implementation of a warning system for algae’ quantification with a relative thresholding approach will be described and a discussion on extension for not-instrumented coastal sites will be presented.