News

Monitoring marine life and plastics in the Mediterranean with XCAM

 

We’ve started the year with a project that’s quite different to the norm for XCAM. Instead of surveying towns and cities, we’ve been surveying the sea.

It’s all part of a wider project that is trying to find the best methods for identifying sea life and floating objects in the Mediterranean Sea.

 

Background

This project was commissioned by the University of Valencia and the University of Barcelona who are part of a consortium of organisations (Medsealitter). Their aim is to find solutions that will ultimately reduce the amount of plastics that enter the sea and adversely affect the flora and fauna. 

This is an important piece of work. The Mediterranean Sea is one of the world’s biodiversity hotspots but it is also one of the most polluted seas worldwide. Plastics, rubber and macro debris can be ingested by marine life and cause them serious harm. This particularly affects fish, dolphins, whales and marine turtles which has a knock-on effect on the rest of the food chain.

Turtle with plastic stuck in mouth

 

The Project

The aim of this project was to test the use of XCAM in its ability to record the position of marine life and litter and for the images to be clear enough to allow identification of the objects. The Universities knew that aerial survey was going to be the most effective method for this and had previously experimented with the use of drones/UAV with limited success due to their low range, payload and flexibility in operating over the sea.

 

The Method

To test whether XCAM could see certain objects we needed to carry out a controlled test. To do this, the University team deployed a small boat into a sheltered part of the waters around the delta and trailed behind them a series of plastic objects and mimics of marine life, fixed to a line of course so it was easy to collect after the survey.

We then conducted an aerial survey using a Grup Air-Med Partenavia P68 with an XCAM RGB/NIR/Thermal fitted within the survey hatch. The XCAM settings were optimised for the flight over the sea and was set to run in a ‘simulation mode’ which took a set of images every 2 seconds. 

The project site was the Delta Del Ebro on the East coast of Spain, about 200km to the south of Barcelona. The aircraft flew to the site, which was about 20 minutes away and then flew over the boat 8 times from different angles at two altitudes, 750ft and 1000ft. This gave us enough imagery to see what could be seen from the air.

Misty morning over the project site
Misty morning over the project site

 

The Survey Plan

For this survey the XCAM was set into its ‘Simulation Mode’ so it would continuously capture and GPS tag images as soon as it was started. As we needed to be flexible with where we would capture the data, a pre-defined flight plan would not have been useful.

Grup Air-Med P68 being prepared for survey
Grup Air-Med P68 being prepared for survey

 

Flying at 1000ft at a speed of 110 knots we were achieving just 10% overlap between images and a GSD of 3.3cm. The overlap was very low compared to a normal survey where we would have over 60% along-track overlap, but for this project it was enough.

 

The Results

The initial results showed that the boat and floating objects could be seen clearly. The images could be reviewed frame-by-frame and can also be added into a GIS to display the photo points for each survey.

Here are some sample images from the survey:

Close up of dolphin and litter

The inset image above shows a few of the planted floating objects including a mock-up of a dolphin and two turtle shells. All of the samples showed up in the imagery.

The interactive slider above shows the difference between the RGB image and the Near Infrared (NIR) image. The NIR shows a positive response but not particularly strong compared to the RGB.

The thermal response, at approximately 8x lower resolution than the RGB and NIR, unsurprisingly doesn’t show very much, even after an image enhancement. It does seem to pick out the two people onboard the boat (top left) and the two dolphin mock-ups.

As well as the original targets, we also picked up a number of floating or moving objects. Could it be a Ray or a plastic box? We’ll leave it to the experts at Medsealitter to identify this one.

Close-up of possible Ray

The next example is maybe a bit easier to identify. At over 20 degrees Celcius and calm waters, it was a great day to be out kayaking!

With a bit of simple processing we can also enhance the images to display only features in the image that give a high NIR response. This is easier for visual analysis, but will still need interpretation to find out if it's genuine objects or sea reflectance.

Enhanced NIR image with white showing high reflectance

This method of using a manned aircraft with XCAM worked very well for this monitoring project. The survey crew could cover large areas in a relatively short time, giving a greater sample area for analysis compared with a drone/UAV.

 

Future Work

The University group will now conduct further surveys using the XCAM along the Spanish coast with the hope that this can be rolled out to other groups around the Mediterranean to collect a larger record of plastics and marine life.

This work will aid future efforts to reduce plastic waste and littering through policy changes and engaging businesses and individuals to change their habits and reduce their plastic usage.

GeoXphere would like to thank the University of Valencia, University of Barcelona and Grup Air-Med for involving us in the project. We hope that the work goes to improve the science and make a positive change.

 

XCAM can be used for a variety of mapping and monitoring projects. Contact the GeoXphere team to discuss your project.