Imperialcollege blog #1
TWIGA at AGU 2018

Photo credit: Gary Wagner, AGU               

Of Science and Sensors: TWIGA at AGU


The Fall Meeting of the American Geophysical Union (AGU) is each year one of the highlights of many environmental scientists. The conference is probably the biggest gathering of earth and environmental sciences in the world, with around 25,000 attendees.  This year the event landed in the cavernous convention center of Washington DC. Several TWIGA members attended this year’s fall meeting, including our principal investigator Nick van de Giesen. Nick gave an invited presentation on the use of sensors in TWIGA in the “MacGyver session”, which is a tong-in-cheek reference to the television series that aired the 1980s.


The MacGyver session is a classic of AGU. It is organized by Nick’s colleague Rolf Hut and gives a platform to scientists who are creative with technology and try to come up with novel or unintended ways of applying new technologies, such as the use of umbrellas to collect rain data.


Our group from Imperial College London hosted a technology-oriented session as well, on the scientific design and application of sensor networks, which is one type of technology that we are keen to explore as part of the TWIGA project.  We invited Rolf to speak in our session, and he managed to give a live demonstration to the audience of how you can use a smartphone to send data from a sensor directly to a server on the internet.


Other exciting talks in our session explained how to use cameras to measure river flow, and cosmic rays to measure soil moisture.


Rolf Hut hosting the MacGyver sessions at AGU 2018

The truth is, if you look from the technical side of things then the possibilities for new sensors and data collection are almost endless. Many of the technologies are also becoming very affordable, with sensor costs as low as a few dollars. That begs the question as to why it takes such as long time to develop and deploy new sensors in the field, and use them to improve water supply, inform farmers, make communities resilient to natural disaster, and more. That is, of course, one of the main questions that TWIGA aims to address.


The reason is that the problem is much more complex than just the technical challenge of building a new sensor. Of course, you can’t just install a sensor and expect that a farmer or a water manager will make better decisions. The raw data have to be processed and made accessible to the end-user in a form that makes them “actionable” for an end user. For example, data produced by a rain gauge are interesting to know how much it has rained, but to make them really useful to a farmer, it is better to convert them into a weather forecast. Similarly, data from a soil moisture sensor can feed into an indicator of the crop status or a warning that it is time to irrigate. Such processing typically requires complex computational models, and often also the integration of different data from different sources (such as satellite observation). And of course, the end-user needs to be able to access the information and be able to interpret it. This often requires training and follow-up, especially in a context where end-users lack formal education or are not familiar with new technologies.

Imperial College presents at AGU 20

Presentation by Wouter Buytaert at AGU 2018

In short, the generated information has to “useful, usable, and used”. Making sure that a technical innovation ticks those three boxes is a big challenge that requires many disciplines to work together. This includes experts in fields as diverse as psychology, user interface design, and information theory.


The scientific community is slowly waking up to this need for interdisciplinarity, and several sessions at the AGU Fall Meeting are making an effort to stimulate interaction between disciplines. In the same meeting, the Imperial team had the opportunity to present some of our work on working with social scientists to use low-cost sensors to support community-based flood warning and resilience building in Nepal, where we work with the NGO Practical Action. This gives us useful insights into how we can apply citizen science in the context of building resilience to natural hazards.


The challenge is now to put that experience to good use when we start designing and testing new sensors for TWIGA!


**Written by Wouter Buytaert


This project has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under grant agreement No.776691. The opinions expressed on the web page are of the authors only and no way reflect the European Commission’s opinions. The European Union is not liable for any use that may be made of the information.