February 2021: On sensors, measurements and conventions

In the latest of CSIC’s monthly Smart Infrastructure Blogs, Dr Miguel Bravo Haro, CSIC Research Associate, brings focus to the important role of sensors, measurement and, in particular, conventions.

On the morning of the 1 November 1755 the city of Lisbon, the fourth-largest city in Europe after London, Paris, and Vienna, was struck by a disastrous earthquake followed by a tsunami. In Lisbon alone, historical estimates place the death toll at between 30,000 and 50,000 people. Based on the slip of the fault, along with the vastness of the damaged area, we know today that the magnitude of such an earthquake had to be between 8.5 and 9.0 in the Moment Magnitude scale [1].

The NASA investigation that followed revealed that one of the engineering teams had applied the traditional American units, while another had used metric units, resulting in a trajectory error of 60 miles – or 95.6 kilometres. The spacecraft encountered Mars at a lower than anticipated altitude, and was either destroyed in the Martian atmosphere or left it entering the heliocentric space for ever. The cost of this failed mission was $327.6 million. Dr Miguel Bravo Haro

The event attracted the attention of commentators far and wide. Among the well-known intellectuals of the time who responded to the natural disaster and devastation was the French Enlightenment writer Voltaire, who claimed to have lost faith in his deistic philosophy and declared that there is evil in the world and it is futile to suppose otherwise. As for the German philosopher Immanuel Kant, while less inclined to blame a divinity, he hastened to postulate the existence of hidden underground cavities across the Earth, where materials of the likes of iron and sulphur were running through and, “fermenting” on contact with water, leading to the uprising of the Earth’s crust [2]. What an imagination one needed to be among the most towering minds of Western history. However, today we know that earthquakes are produced by the drift of tectonic plates, a theory put forth in 1912, but not widely accepted until the 1950s and 1960s, when data collected by modern sensors provided all sorts of relevant information, from ground acceleration to dating of sediment cores from the seabed. This explanation was based on data-driven observation – an early example of data science [3] – which has bestowed the theory the status of scientific revolution [4].

The accelerating advancement of modern science has been powered in part by the development of sensors that have enabled scientists and technicians to verify observations or measurements to support their theories [5]. Meticulous balances for weight measurement allowed Henry Cavendish and Antoine Lavoisier to precisely measure the volumes and weights of gases. Mechanical and pendulum clocks became accurate enough to show that the duration of the length of the day was not constant. The torsion balance developed by Charles-Agustine de Coulomb to precisely measure force was key to the progress of electrostatics, and enabled Cavendish to measure the forces between terrestrial objects. The invention of the reliable mercury thermometer by Daniel Gabriel Fahrenheit was pivotal for countless discoveries in chemistry, physics and biology [6].

The list is endless and fascinating, but there is another important issue that arises as a consequence of being able to measure the world around us: the need for convention. We measure to understand and interpret, rarely alone but exchanging and sharing observations and measures with our peers. In this process of communication, the choice of convention is vital. Among the most iconic examples are the rivalry between Celsius and Fahrenheit, and the never-ending dispute for metric or imperial systems [6,7]. For many, these choices might appear as fictitious quarrels over simple units of measure which in the end are arbitrarily imposed. But these are not solely battles about the size of a given quantity; the choice of unit and scale are bounded to the selected method of measurement, normally based on substantial assumptions and investigations. The choice of convention is not a capricious imposition, but a race for efficiency, simplicity and practicality. For instance, the decimal clock and the 10-day week lost the race, after being put forward during the French Revolution, in the middle of the invention of the metric system [6,8]. The original temperature scale of Anders Celsius was upside-down, with the 0º indicating the boiling-point of water [6]; imagine a world where the freezing-point of water is 100º!

Without convention or a general standard the worst can happen. In 1999, communication between NASA and the Mars Climate Orbiter, a robotic space probe launched nearly a year before to study the Martian climate, was lost. The NASA investigation that followed revealed that one of the engineering teams had applied the traditional American units, while another had used metric units, resulting in a trajectory error of 60 miles – or 95.6 kilometres. The spacecraft encountered Mars at a lower than anticipated altitude, and was either destroyed in the Martian atmosphere or left it entering the heliocentric space for ever. The cost of this failed mission was $327.6 million [8,9].

Here at CSIC, our work will be part of the history of sensors and instruments and data and analytics – and the way these are contributing to a better understanding of the critical role of our built and natural environment in shaping the society we live in. One of our goals is to lead the transformation of how people perceive and think about our civil infrastructure monitored by the accurate measurement of sensors to deliver insights to inform better decision-making. This metamorphosis is in its very early stage, and transformation involves more than the deployment of sensors and interpreting measures for better data-driven asset management. CSIC is in the race for the choice of conventions, for setting standards and protocols based on the best-tested procedures, methodologies and technologies which will hopefully become the pervasive language in this field.

• Dr Miguel Bravo Haro's work is supported by the Centre for Digital Built Britain (CDBB)/the Construction Innovation Hub (CIH) funded by UK Research and Innovation through the Industrial Strategy Fund.