Tag Archives: boundaries

Isolated systems in nature?

Is a coconut an isolated thermodynamic system?  This is a question that I have been thinking about this week.  A coconut appears to be impermeable to matter since its milk does not leak out and it might be insulated against heat transfer because its husk is used for insulation in some building products.  If you are wondering why I am pondering such matters, then it is because, once again, I am teaching thermodynamics to our first year students (see ‘Pluralistic Ignorance‘ on May 1st, 2019).  It is a class of more than 200 students and I am using a blended learning environment (post on 14th November 2018) that combines lectures with the units of the massive open online course (MOOC) that I developed some years ago (see ‘Engaging learners on-line‘ on May 25th, 2016).  However, before devotees of MOOCs get excited, I should add that the online course is neither massive nor open because we have restricted it to our university students.  In my first lecture, I talked about the concept of defining the system of interest for thermodynamic analysis by drawing boundaries (see ‘Drawing boundaries‘ on December 19th, 2012).  The choice of the system boundary has a strong influence on the answers we will obtain and the simplicity of the analysis we will need to perform.  For instance, drawing the system boundary around an electric car makes it appear carbon neutral and very efficient but including the fossil fuel power station that provides the electricity reveals substantial carbon emissions and significant reductions in efficiency.  I also talked about different types of system, for example: open systems across whose boundaries both matter and energy can move; closed systems that do not allow matter to flow across their boundaries but allow energy transfers; and, isolated systems that do not permit energy or matter to transfer across their boundaries.  It is difficult to identify closed systems in nature (see ‘Revisiting closed systems in nature‘ on October 5th, 2016); and so, once again I asked the students to suggest candidates but then I started to think about examples of isolated systems.  I suspect that completely isolated systems do not exist; however, some systems can be approximated to the concept and considering them to be so, simplifies their analysis.  However, I am happy to be corrected if anyone can think of one!

Image: https://www.flickr.com/photos/yimhafiz/4031507140 CC BY 2.0

Revisiting closed systems in nature


It is the beginning of the academic year and once again I am teaching introductory thermodynamics to engineering undergraduate students and my MOOC entitled ‘Energy: Thermodynamics in Everyday Life‘ is running in parallel.  Last week after my lecture on thermodynamic systems, a student approached me to ask whether the universe is a closed and isolated system.  It’s an interesting question and the answer is depends on the definition of universe.   In thermodynamics, we usually define a boundary to delineate the system of interest as everything inside the boundary and everything else are the surroundings.  The system and surroundings taken together are the universe (see my post ‘No beginning or end‘ on February 24th, 2016).  If the universe is defined as the system then there are no surroundings; hence the system cannot exchange energy or matter with anything which is the definition of a closed and isolated system.

Physicists often refer to the observable universe, or define the universe as everything we can observe.  We are aware that we cannot observe everything.  Hence, it is reasonable to suppose that the observable universe exchanges energy and matter with the unobservable space beyond it, in which case the observable universe is an open system.  We could also consider the concept that we are part of multiverse and our universe is only one of many, in which case it seems likely that is not isolated, i.e. it can exchange energy, and perhaps it is open, i.e. it can exchange both energy and matter with other parts of the multiverse.

This is not really thermodynamics in everyday life.  However, the occurrence of closed systems in nature appears to interest a lot of people to judge from the visits to my previous posts on this topic.  See ‘Closed Systems in Nature?‘ on  December 12th, 2012; Is Earth a closed system? Does it matter? on December 10th, 2014; and ‘No Closed Systems in Nature‘ on August 12th, 2015. For more about system boundaries, see my post entitled ‘Drawing Boundaries‘ on December 19th, 2012.

Is Earth a closed system? Does it matter?

 Earth's annual global mean energy budget,  from Kiehl and Trenberth 1997

Earth’s annual global mean energy budget, from Kiehl and Trenberth 1997

The dictionary definition of a system is ‘a set of things working together as parts of a mechanism or an interconnecting network; a complex whole’. So it is easy to see why ‘systems engineering’ has become ubiquitous: because it is difficult to design anything in engineering that is not some kind of system.  Perhaps the earliest concept of a system in post-industrial revolution engineering is the thermodynamic system, which is a well-defined quantity of matter that can exchange energy with its environment.

Engineers define thermodynamic systems by drawing arbitrary boundaries around ‘quantities of matter’ that are of interest, for instance the contents of a refrigerator or the inside of the cylinder of a diesel engine [see my post entitled ‘Drawing Boundaries‘ on December 19th, 2012].  These boundaries can be permeable to matter in which case the system is described as an ‘open system’, as in the case of an diesel engine cylinder into which fuel is injected and exhaust gases ejected. Conversely, the boundary of a ‘closed system’ is impermeable to matter, i.e. the refrigerator with the door closed.  The analysis of a closed system is usually much simpler than for an open one.  In his Gaia theory, James Lovelock proposed that the Earth was a self-regulated complex system.  Is it also a closed thermodynamic system?  It is clear that energy exchange occurs between the Earth and its surroundings as a consequence of solar radiation incident on the Earth (about 342 Watts/square meter) and radiation from the Earth as a consequence of reflection of solar radiation (about 107 Watts/square meter) and its temperature (235 Watts/square meter).  This implies that we can consider the Earth as a thermodynamic system.  The Earth’s gravitation field ensures that nothing much leaves; at the same time the vast of emptiness of space means that collisions with matter happen only very occasionally, so the inward flow of matter to Earth is negligible.  So, perhaps we could approximate Earth as a closed thermodynamic system.

Does it matter?  Yes, I believe so, because it influences how we think about our complex life support system, or spaceship Earth that sustains and protects us, as Max Tegmark describes it in his book ‘Our Mathematical Universe’.  In a closed system there is finite amount of matter that cannot be replenished, which implies that the Earth’s resources are finite.  However, our current western lifestyle is focused on consumption which is incompatible with a sustainable society in a closed system.  Even the Earth’s energy balance appears to be in equilibrium based on the data in the figure and so we should be careful about massive schemes for renewable energy that might disturb the Gaia.


Kiehl, J.T., and Trenberth, K.E., 1997, Earth’s annual global mean energy budget, Bulletin – American Meteorological Society, 78(2):197-208.

Thess, A., The Entropy Principle – Thermodynamics for the Unsatisfied, Springer-Verlag, Berlin, 2011.

Tegmark, M., Our Mathematical Universe, Penguin Books Ltd, 2014.