Staff spotlight: Aser Mata – Earth Observation Scientist

Aser has been at the laboratory for almost ten years now, having previously studied Physics at the University of Salamanca, which is located in the heart of his hometown, and is among the oldest ongoing universities in Europe, and the oldest in Spain.  

“At that time, my degree was five years long and allowed me to choose a specialization by picking topics of study during the last two years. I chose mainly topics around Quantum Mechanics, Cosmology, Particle Physics (leptons, quarks, hypercharge…) but also a number around Climatology. I felt that the ideas and the maths that built theoretical physics were beautiful, but I thought I could do something more applied with a more immediate impact for a more sustainable world.” 

“Following that, I completed a Masters in Renewable Energy at the Autonomous University of Madrid. It is worth mentioning that during my time at the university – and much earlier even – my father had a bakery and I always worked there when on holidays. It was very important for him that my sister and I learned the value of hard work from a young age, hence that was also part of my education. Right after my Masters, I came to Plymouth for love for the woman is now my wife. It was a challenge to find a job in a different country, but I had some savings and wanted to chase a job in science that also allowed me to make a positive impact on the environment. After months of unsuccessful searching, I started at PML an unpaid internship for a few months and later applied for an opportunity to work at PML as an ‘Airborne Earth Observation Data Analyst’.” 

Indeed, Aser’s wife is Silvia Pardo, who moved from Spain to England to work as an Earth Observation Scientist at PML for almost 12 years, and we are extremely lucky that this twist of fate also led Aser to the laboratory!  

Image caption: Aser pictured by ‘Smeaton’s Tower’ lighthouse on Plymouth Hoe (2016) 

He describes how his career has evolved during his time with PML: 

“Back in 2015, in my ‘Data Analyst’ position, I was working as part of the NERC Airborne Research Facility – Data Analysis Node that is now part of NEODAAS [NERC Earth Observation Data Analysis and Artificial-Intelligence Service]. I gained a lot of experience processing data and writing code for all different kinds of datasets that were collected on airborne platforms.” 

“Thanks to my Physics background, I was comfortable around optical sensors and enjoyed working on calibration and understanding the limitations when collecting data. Later on, when the opportunity arose to expand my skillset from data collected with aeroplanes, to data acquired using drones at lower altitudes; it felt somehow natural.” 

“Fast forward to 2018, PML secured our first project using a consumer grade drone through the REVIVAL project [NERC-funded ‘Rehabilitation of Vibrio-infested waters of Lake Vembanad’ project], and I learned how to pilot. I loved it, and even bought my own personal drone to practice on in my free time and create cool videos and photos.” 

Image above: Aser pictured out at sea on the PML Explorer to capture drone footage 

“It became clear very quickly that flying drones as a research institute under the regulatory frame of drone hobbyists was only a temporary solution, but not appropriate if we aimed to keep working with drones and expanse our capabilities towards the creation of a PML drone facility.” 

Image above: Aser at the laboratory with PML’s first consumer-grade drone 

Image captions: Aser pictured with PML’s first consumer-grade drone, demonstrating the new technology to students at Devonport School for Girls as part of work on ocean literacy 

“William Jay and I undertook some training and flight tests to become professional drone pilots, and I nominated myself forward to be the person looking after our drone license and capabilities. It was a responsibility but also an opportunity. I firmly believe that drones are a great tool for a lot of the research that we do either to collect data at very high resolution, or to upscale or downscale computer models, and also validate satellite products, and we will steadily use them more often in the coming years.” 

Image caption: Aser out on fieldwork with PML’s latest drone at the Lake district 

“In my role, knowing how to program is key, but understanding how sensors work and the physics laws behind them is also key to be able to create processing tools and derive outputs that make sense. One part of my job is to bring that expertise to enable others to do their own research by collecting drone data, or processing any earth observation data.” 

Image caption: Aser demonstrating to colleagues how to operate a spectrometer that was designed to be mounted on a drone for collecting hyperspectral data and derive water quality indicators. Photo credit: Dr Dan Clewley. 

“Another part of my job allows me to perform my own scientific analysis and contribute to or lead papers when my time in projects allows me. I think all of that is still part of what being a modern “scientist” is about in this line of work. My title was changed to “Earth Observation Scientist” accordingly at some point in 2021.”  

Having skillsets in so many different areas has certainly benefited both Aser’s career and our capabilities at the laboratory, and an example of this is his work in PML’s Hyperdrone project, which involved using drones for remote sensing to detect plastic marine litter over water or on the shoreline.  

“HyperDrone was the first project I led as PI a few years ago, funded by the European Space Agency (ESA) as part of the Discovery Campaign on Remote Sensing of Plastic Marine Litter. This was the first proposal I wrote, and it was amazing to have it funded successfully.” 

“The project was designed to build on findings from PML’s OPTIMAL project led by Dr Victor Martinez-Vicente, who I worked with closely. I analysed the spectral signature [a material’s unique pattern of how it reflects, absorbs, or emits electromagnetic radiation – captured by sensors – and essential for identifying and classifying objects and materials on Earth’s surface] of three different types of plastic targets deployed on Whitsand Bay, with which reflectance was collected using the hyperspectral instruments from the NERC-ARF mounted on an aeroplane flying at approx. 1 km altitude.” 

Image caption: Aser with Dr Victor Martinez-Vicente and work experience students, demonstrating how to use a Secchi disk – a simple 19th Century tool that is still useful to ocean scientists today in the age of satellites. Secchi disks can be used to monitor both ocean biomass and water quality. 

“Due to the high altitudes and the resulting pixel size, the plastic targets that we deployed needed to be quite large – around 10x10m – which made the field campaign difficult as you need quite a lot of space! And of course, logistically, to get such large plastic targets to the location, they needed to be able to be folded to fit into transport – which meant we couldn’t use hard plastics like PVC or HDPE – so we were limited to only three types of plastics.” 

“However, my goal was to collect data from a wider range of plastic types, and I could do that by exploiting the use of drones flying at lower altitudes – below 120 meters – and collect data with the hyperspectral imager available at the NERC Field Spectroscopy Facility (FSF). I worked closely with NERC-FSF, and we collected two datasets in the shorelines of Scotland in 2021 and 2022, on both sandy and rocky shorelines, respectively collecting data from approximately 15 plastic targets, of a much smaller size: 1x1m.

Image caption: Drone photo of team testing drones to help monitor water quality at the lake district in collaboration with UKCEH  

“With the help of Robbie Ramsay from NERC-FSF, we also computed uncertainties for those datasets, making it a great resource for developing radiometric algorithms for plastics detection. I made the datasets available for free to any other researchers at CEDA Archive to encourage their use [these datasets are available on the CEDA archive here and here].” 

“This project started during the pandemic, and of course, that caused delays and all sorts of logistical problems, so we were very happy with all that we achieved despite those difficulties.” 
“I also developed algorithms for plastic detection based on spectral features in the Short Wave InfraRed, and recommendations and thresholds for developing multispectral sensors for subpixel plastic detection. That lengthy work was compiled in the ESA summary, and is yet to be published, but more recently, thanks to the work led by the PhD student Ámbar Pérez-García, we published a study using the visible range of the HyperDrone dataset which proves the value of the project.”  

Image caption: Aser with Chris MacLellan working on the calibration of the NERC-ARF hyperspectral sensors. 

It’s safe to say that Aser has worked on some incredibly innovative solutions that have broadened and advanced the sector of earth observation. And, 2024 has turned out to be a busy year for Aser, he recently published his study ‘Drone imagery and Deep Learning for mapping the density of wild Pacific oysters to manage their expansion into protected areas’, which is the first time a study has explored the use of drones and deep learning to detect, count and map Pacific oysters – an invasive species with a potent harmful impact. 

Above: a cluster of Pacific oysters – Crassostrea gigas. Photographed during project fieldwork at Kingsbridge Estuary, led by Aser Mata. Pacific oysters can reach up to 10 inches in length. The shell is elongated, thick, rough and sometimes sharp. 

His study secured a high detection rate of 88% – meaning that this method of monitoring can provide a cost-effective solution covering medium to large size areas of terrain, in a small window of time, with only 12% of false positives being raised. 

Aser commented: “What makes Pacific oysters so difficult to monitor currently is the fact that they make their home in intertidal areas – which can only be accessed when the tide is out – and in tricky areas to reach like mudflats or vertical cliffs. Traditional on-foot surveys therefore may present a risk to the personnel involved and may also require suitable equipment like mudflat walkers. Such surveys are time-consuming – significantly increasing labour costs, and they are often limited to areas that are easy to access by foot, which doesn’t give us the complete picture.” 

Because of Aser’s multidisciplinary experience, he has a unique ability in being able to manage all areas of this project: from flying the drones to collect the imagery of the terrain, to using object detection and comparing different deep learning models to find the best accuracy in detecting the oysters. He describes some of his favourite moments from this piece of work:  

“It is shaping out to be a great year. The drone fieldwork for the Pacific detection was indeed very exciting, those Pacific oysters are quite large!” 

Above: Lead scientist, Aser Mata, pictured with a Pacific Oyster during fieldwork at Kingsbridge Estuary 

“If I had to pick up a favorite experience, I would say that it was sensational to see how everyone in the project was equally excited with this work. It sounds cliché but it is true nevertheless.” 

“We all worked really hard to get this paper out and had to fight to find the time and resources to do it. For example, William Jay did a great job in the fieldwork, and didn’t even hesitate when he volunteered to draw boxes around thousands of oysters manually on the screen during many, many hours.” 

Above: Lead scientist, Aser Mata, pictured with colleagues during fieldwork for the Pacific oysters study, he credits their hard work to the success of the paper. 

Image above: Model detection of Pacific Oysters in the study. 

“Dr David Moffat was always keen to have a chat and find ways to improve the model and contribute to the paper with better metrics and figures. We were also very lucky to have Silvia Almeida from MARE [Marine and Environmental Sciences Centre – Portugal] joining the field campaign in UK, and her contributions were also key to the paper in its current form… I think we all thought this small study was very innovative and wanted to get it out. Having that feeling of camaraderie that kept us moving forward is my favorite memory.” 

This groundbreaking study resulted in a new workflow, and Aser describes the potential is has to be used in other ecological management issues: 

“Indeed, this workflow is similar to that proposed by other researchers for potential standardisation of litter monitoring on the shoreline. Essentially, the outputs of the detection model (either individual Pacific oysters detected or individual items of litter) are geolocated and then transformed to UTM projection [The Universal Transverse Mercator (UTM) is a map projection system for assigning coordinates to locations on the surface of the Earth]. The analyst can select a grid of their chosen size (e.g: 1m x 1m or 5m x 5m, or more depending on the size of the area) and present the number of Pacific oysters or the number of plastic items per cell to present the data in a way that can be easily ingested.” 

“Another example could be seagrass mapping. Seagrass play an important role in intertidal ecosystems and act as carbon sinks. Their extension can be mapped and geolocated and the percentage of land covered by seagrass could be presented by cell.” 

“Perhaps even more interesting, datasets can be interrogated simultaneously for different purposes by running different detection algorithms. For example, if we had those models ready, drone data could be collected over Marine Protected Areas to detect oysters, seagrass and kelp to later present a grid map where the number of each species or is detailed by cell.” 

It’s fascinating to see this potential in environmental mapping to protect marine and coastal ecosystems. And this brings us to our next point of discussion, it was announced recently that Aser has had a new project funded – Green LiDAR! This project will use a novel sensor, designed for both topographic [relating to the arrangement of the physical features of an area on land] and bathymetric [relating to the depths of the ocean or representing the contours of the seabed] surveys, for mapping underwater environments that are hard to reach, and this will be mounted on a low carbon UAS platform. 

Aser describes what he is most looking forward to in the project as lead scientist: 

“I am very much looking forward to start collecting data with the Green LiDAR sensor, although it will not happen until spring next year at the earliest. It will also however be daunting to have that very expensive equipment flying on a drone that I will be responsible for while I pilot it remotely, but I am up for the challenge!” 

“This instrument will be able to deliver bathymetric data on shallow waters where it is very difficult or impossible to get accurate data with any other sensor. This will allow us to better map kelp and seagrass environments, as well as understand their cycle and the carbon they capture. Similarly, it will help us to better understand riverine environments. This I hope will bring new opportunities to collaborate and novel outputs.” 

“As part of this project, PML will be collaborating with NOC, BGS and UKCEH and SAMS to put to the test the sensor capabilities in three different regions that include river sections, lakes and coastal environments. We will also make that data available with our findings, therefore any researcher will be able to access showcase real data that will include benchmarking exercises of this new sensor with terrestrial data and bathymetric sensors when is possible to be deployed. I believe the teams at UKCEH and SAMS are already excited to include this capability in future projects.” 

It’s fantastic to hear that the data captured will be made available to all – collaboration and information sharing is key to the advancement of science. Stay tuned for updates! 

To finish the chat, Aser spoke about some of his hobbies and interests outside of the office. 

“I do have many hobbies and I keep switching depending on the season and mood. I love to play squash regularly and I try to get more people at PML to join me (give me a shout if you want to play!), and I really like hiking with my partner, especially when the terrain is not flat as I find it boring to see everything that lies ahead of you all the time.” 

Above: Aser out on a bike ride in Orkney during the drone fieldwork of the Talisker project. Photo credit: William Jay

Image: Aser pictured at Arashiyama Bamboo Forest” in Kyoto

“On the other hand, I consider myself an introvert – which doesn’t mean that I am shy or anything like that – but it means that I love to have downtime without social interactions that empty my batteries. I read, I play video games and board games… And probably above all, I love listening to music. And wearing something black, that is also a hobby I have that I exercise mostly every day.” 

Image caption: Aser unwinding painting Warhammer at home