An example of the social good is addressing areas like climate change and global heating. This latest edition of Essential Science looks into three examples of where big data analytics have been harnessed for environmental protection.
Before diving into the green examples, it is useful to consider what exactly big data is.
What is big data?
The volumes of data make up the totality of digital information for a system comes under what has been designated ‘Big Data‘. This can represent anything, provided it is something that contains a larger (sometimes unimaginably large) data set. Here data about individuals, groups and periods of time are taken and combined into bigger groups or longer periods of time. From this amass of information, new and more in-depth inferences are attempted to be drawn.
How good is big data?
Big data is everywhere, and advocates say analyzing masses of information holds the answers to almost any problem in the world. This grandiose claim does not fully-account for the fact that simply amassing huge quantities of data is not the same as understanding it; and to understand big data good analytical tools and artificial intelligence algorithms are required. For example, the quality of the analysis is co-dependent upon the quality of the mathematical tools used to extract meaning from vast data sets.
To improve the accuracy of big data analytics the mathematical approach of mutual information has been used to construct big data analysis tools. This approach also helps to reveal unexpected reveal patterns in large lists of numbers.
For example, findings patterns in data sets on the numerous bacterial species that help with biotechnology and drug discovery. Using big data for this type of research is important, as the objective is to identify all possible types of patterns within the data without reliance upon any prior assumptions.
When handling big data that uses personal details, especially health-related data, it is important to have an ethics policy in place.
Improved gas membranes
With the first environmental related example, scientists have developed a method that brings together big data and machine learning to design gas-filtering polymer membranes. These membranes via the development of advanced gas separation membranes.
Plastic films are used as membranes to separate mixtures of gases, like carbon dioxide and nitrogen. It is possible to use this approach to separate out carbon dioxide from other gases for natural gas purification and to facilitate carbon capture.
Because there are hundreds of thousands of plastics that can be produced the task of selecting the appropriate materials is very complex, based on significant variations in chemical structure. To aid this task big data analytics and machine learning was used to select the most appropriate materials.
The outcome of the process and details of the most suitable membrane have been published in the journal Science Advances. See: “Designing exceptional gas-separation polymer membranes using machine learning.”
Big data to improve the environment
The second example is more of a general point, about using big data analyses to help to develop environmental policy and management. To develop this, there needs to be a commitment on behalf of companies towards the opening up of data streams and with making as much data as possible as open to all.
As an example of the power of big data, it has been revealed that 2.3 million square kilometers of forest has been lost between 2000 to 2012. This has been gained from a big data analysis of over 700,000 satellite images. A similar assessment of the seas has revealed that the Earth has lost more than 20,000 square kilometers of tidal flats since 1984.
The application of this form of big data has been outlined in the journal Nature Communications, in a study headed “Opportunities for big data in conservation and sustainability.”
Big data and energy efficiency
A new inquiry has found that magnetic wires, spaced a certain way, can lead to up to a thirty-fold reduction in the amount of energy needed to run neural network training algorithms. This will result in significant energy savings and a reduction in heat. This was drawn from big data analysis.
The application is explained in the journal Nanotechnology, where the research is titled “Maximized Lateral Inhibition in Paired Magnetic Domain Wall Racetracks for Neuromorphic Computing.”
Essential Science
This article forms part of Digital Journal’s long running Essential Science column, where a topical science subject is examined each week by Dr. Tim Sandle.
Last week the topic was why, with COVID-19, many people experience the loss of smell (anosmia). New research has the answer to why this happens.
The week before we learned that astrophysicists have proposed a novel method designed to find black holes in the outer reaches of the Solar System. This may also answer the question as to whether the hypothesized Planet Nine really exists
