“Our satellite data is helping Kenyan farmers”
Andrew Bowell, a third-year MPhys (Astrophysics) student, is using satellite data to help Kenya’s Maasai farmers locate pastures for cattle grazing.
Last summer, I was looking into different internships and I came across the Junior Research Associate (JRA) scheme here at Sussex. A few friends of mine had research placements and they said it was really rewarding, so I decided to apply.
DISCUS (Data Intensive Science Centre) at Sussex had quite a few projects and they all caught my eye as they were all data science and coding based. I worked on two machine-learning projects. SimFarm2030 uses past weather and crop yield data to try to predict future crop yields across the UK, whereas AstroCast uses satellite data to predict areas in Kenya at risk of drought.
After my eight-week JRA period, I was approached about carrying on full time for another four weeks until the start of term. I was then asked if I would like to continue developing the forecasting software for the AstroCast project, which I am doing for a few hours every week. It’s great to feel that the work I am doing is benefiting agricultural communities in this worrying time of climate change.
I went to Kenya to train stakeholders for the AstroCast project. It truly was a once-in-a-lifetime experience. The stakeholders included the National Drought Management Authority, Kenya Red Cross, Kenya Met Department, World Food Programme, and the Regional Centre for Mapping Resources for Development.
I was training them on how to use the model developed from the AstroCast project to forecast the vegetation condition index (VCI), which is an agriculture drought index derived from satellite remote sensing images of earth. It looks at how sunlight reflects off vegetation and measures how green/healthy the plants are. If it isn’t doing well, the region is said to be in drought or at risk of a drought.
Using a mathematical model called Gaussian Processes, we are able to analyse how the VCI has changed throughout the past and then create skilful forecasts of how it will change up to 12 weeks in advance. The goal was to get the stakeholders able to process satellite data, create a VCI time series, perform forecasts and generate reports. The training gave me a taste of the teaching world. I learned how to give presentations, create workshop materials and generally teach people how to code in regard to data science.
Before the training we went on a three-day safari out to the Maasai Mara game reserve. It was a really great experience; we saw pretty much every animal you could as well as going into a Maasai village and meeting the local people. These were the people the project hopes to help as, when implemented, forecasts of how the vegetation in different grazing blocks will change over time will be available for agro-pastoralists to view and decide where to take their cattle next to graze.
I applied to Sussex as it was one of the best universities in the country as well as being fairly close to my family home in Essex. I visited a fair few across the country, but when I visited the Physics and Astronomy department at Sussex there was a really good level of enthusiasm. The staff and students on the open day seemed a lot happier, giving long and detailed answers to questions. They enjoyed doing their degree here, and that’s exactly what I was looking for.
I have really enjoyed my time at the uni so far and don’t want to leave! The Physics department is quite tightly knit, and everyone seems to know each other. The study areas always have people able and willing to help you out if you’re stuck with a problem. It’s just a happy and positive environment. This was obvious during the summer months, when everyone was out on the grassy areas eating and drinking.
I’ve also enjoyed going out in Brighton with my mates for a drink or a bite to eat. We always seemed to end up in The King and Queen pub, playing pool. I live by the beach so I like to jog along the promenade every other day with my girlfriend. This has become especially important at the moment! It’s nice to get outside and try to keep sane.
Working for DISCUS has opened my eyes to the possibility of a PhD in data science or physics. I’ve enjoyed being in a research setting. The idea of doing ground-breaking research fascinates me and I would love to continue aiming for that. I definitely know that whatever I do will involve machine learning and programming, whether at Sussex or for a company somewhere else.
Sussex has been good at offering support through the Physics department. I have always had amazing help when making applications for different internships, or even just writing a CV for a part-time job. We even had a dedicated careers module that prepared us for going into the outside world. The DISCUS team has helped out too, offering me amazing experiences that have accelerated and diversified my learning. It’s been a great combination.
Interview by Jacqui Bealing
This profile is part of our This Sussex Life series.
Visit the University of Sussex website