Arctic Resilience Report - Real-time social sensor

Written by Marcus Carson, Roman Yangarber, and Thomas Grandell

Welcome to the ARR project's real-time social sensor page

The Arctic is a vast, sparsely populated region undergoing rapid, and sometimes dramatic change. To monitor and better understand developments in the Arctic, scientists have deployed a wide array of scientific tools, including sensors, satellite monitoring, and painstaking on-site data collection to measure and track changes in both biophysical and social parameters. While some data can be collected by sensors (primarily biophysical parameters) in what amounts to real time, other kinds of research is by nature very slow.

Data gathering for better understanding what is happening in the communities across the Arctic is almost exclusively of the on-site variety, often case study based, with the consequence that the tracking and analysis of social developments lags far behind the developments themselves. In a period of rapid change, our knowledge of changing social conditions develops too slowly to identify critical thresholds while there is still time to respond. This presents serious challenges for policymakers who seek to base policy decisions on the best and most up-to-date knowledge available.       

We see an important opportunity to shorten the turn-around time for data collection and analysis of social developments. Our hypothesis is that the communities across the Arctic themselves constitute a form of real-time sensor, as people identify small changes in the social and physical environment around them as they occur - often before they can be properly measured and interpreted by more conventional scientific means. Residents of the Arctic have traditionally been keen observers of variations in the physical and social environment. Modern media report items of local, regional and global interest which reporting constitutes an enormous source of knowledge about developments considered by local observers and other key players to be worthy of attention and concern.

We believe that if we can complement our conventionally gathered scientific data with these newly available data sources from the social domain, and gain access to this information for analysis at an early stage, we might identify crucial trends in near-real time, identifing change before it irreversibly occurs - or opportunity as it opens up.

The challenge is this: gathering and making sense of this vast body of information is not a simple task. Important locally reported events might not be picked up in global media coverage until critical thresholds have already been passed, and then it might be too late to effectively respond. Similarly, events of a global character that would influence the Arctic region and that is reported in global media may be lost in the flow of information.

To trap these weak signals from the social chatter in the news flow, continuous monitoring of these sources is needed. This requires both collecting the data and analyzing articles - potential data sources – for events that are produced every day. As the Arctic region is vast, spanning eight countries and many more languages, this would be virtually impossible to do manually - or at least extremely time and labor intensive. That is why we are exploring the use of technology to aid us.

We are currently piloting automatic information gathering, extraction and visualization with the help of the University of Helsinki and Absorber Ltd. Our pilot project is using around 2.000 global news sources in English, from which we are currently continuously capturing news related to the energy sector (specifically oil, gas and wind) concerning investments, acquisitions and accidents in the Arctic region.

To accomplish this, we use web crawlers and RSS feeds to collect the articles, and relevant information is extracted from this stream by using state-of-the-art technology in the field of natural language processing and event extraction. The identified, relevant articles are then made available for visual and textual inspection through our web tool.

When exploring this content, bear in mind that these are the results of an early-stage proof-of-concept project, i.e. we wanted to know if these technologies could be applied in a way that provides us with added value. While these preliminary results are therefore imperfect, our assessment is that they are very promising and warrant further development of this pilot effort, adding more sources, more sectors, more event types and support for more languages to expand our material and sharpen our big picture. As the project progresses, more and more information will be made available to interested parties via our web site.

But, without further ado, enjoy the fruits of our pilot efforts here (requires setting up a free user account)»

 If you want access to the tool or would like to contribute to the success of this project or technologies used, please contact

Dr. Marcus Carson
ARR Project Director
Stockholm Environment Institute
+46 73 460 4845
This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Dr. Roman Yangarber
Principal Investigator
University of Helsinki
+358 50 415 1713
This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Mr. Thomas Grandell
Managing director
Absorber Ltd
+358 40 9000 474
This e-mail address is being protected from spambots. You need JavaScript enabled to view it