Engineering Sustainability: an Ethics No-brainer

David Wagner 2007/04/22 23:47

This paper considers sustainability as an ethical issue for civil engineers. What are the implications of unsustainable practices? How can engineers evaluate the sustainability of alternatives under consideration? What can engineers do to make projects more sustainable? This report focuses on three ethical issues of sustainability.

  • The appropriate use of the terms sustainable and sustainability by engineers
  • Conservative assumptions, “worst case” scenarios, graceful failure, and redundancy
  • Implications for project service life and suitability for purpose

Just what is Sustainability, Precisely?

sustainable, adj.

“being able to continue into the future.”
Definitions of Anthropological Terms, Oregon State University

“able to continue over a period of time”
Cambridge Advanced Learner's Dictionary

Sustainability is often presented as an imprecise term, there are dozens of varying definitions,1) and it may seem a term more suitable for a marketing strategy than something engineers can quantify and analyze. But an energy and mass balance model is suggested by observing how the definitions of sustainability center around a basic theme; if it’s not sustainable, it will come to an end.

This suggests a steady-state condition, with quantifiable energy and mass flows into and out of storage. For steady-state conditions, the flow of resources out of a system (Qout) approaches the flow of the energy and material into the system (Qin), and the change in storage (ΔS) of these resources approaches zero.

This drawing shows a sustainability model as a mass or energy balance box with Qin flowing into storage S, and Qout flowing out while ΔS approaches 0. In this paper, storage is essentially the biosphere, the volume near the surface of the earth humans can access (or some defined portion of it) excluding those locations where the quantity of interest is currently in use. For example, consider timber. The mass of suitable trees on the planet is timber in storage along with all the suitable wood in lumberyards, reclamation facilities, disused structures, and other locations where the timber not actively in use. Although it is possible to evaluate the sustainability of timber or other resource use worldwide, it is often more convenient to consider a smaller region to analyze the local self-sustainability of resource use.

This drawing shows an example mass balance for timber use in an area.  Qin is annual timber production in the area and Qout is annual timber used in the area.  S is the quantity of live trees and  ΔS approaches 0. Thus, an engineer could consider the area occupied by a community and determine the annual amount of timber (Qout) needed to repair and periodically rebuild its wooden structures. In most cases it will be safe to assume the amount of timber in storage is represented by the amount of live trees, and the rest is negligible. Now note timber is not a conservative quantity. The annual amount of timber produced in the area (Qin) is proportional to the amount of land used as productive forest, and thus it is relatively easy to estimate whether enough of the area is forested to be self-sustainable in timber.

Similar analyses can be done for all the significant material and energy needs of a community, or for any other project, to determine whether it can be considered sustainable overall. Although this description only outlines the process, it is clear sustainability can be analyzed quantitatively and it would be irresponsible for engineers to use the term without some kind of formal justification.

Some Conservative Assumptions

Conservative Assumptions

  • Worst Case Scenarios
  • Graceful Failure
  • Redundancy

Standard engineering practice often requires designing a project to withstand reasonable worst case conditions using conservative, even pessimistic assumptions, and to ensure it will fail gracefully under more extreme conditions. Thus a building might be designed to withstand an earthquake of 7.8 on the Richter Scale and to fail in such a way as to still allow egress for survivors if a greater quake were to strike. A stormwater drainage system may be designed to completely contain a 10-year flood, and for greater floods to overflow into parks and other nonresidential areas. But most of the worst case conditions considered are simply extreme values of design quantities such as applied forces or rainfall rates.

To illustrate, transportation infrastructure is often designed to meet estimates of future vehicular demand based on current trends. The conventional assumption is that the greatest reasonable future vehicular transportation demand is the worst case, and designing a transportation system to meet these needs is a conservative strategy most likely to meet whatever the transportation needs actually become. But this design strategy unreasonably assumes a high-quality energy source will continue to be available to utilize and maintain this infrastructure and to fuel the distribution of vehicle types currently in use. 2)

Thermodynamic Laws

  1. Energy cannot be created or destroyed.
  2. Energy systems have a tendency to increase their entropy.

A completely different, and perhaps far worse case is to assume the laws of thermodynamics continue to hold and we must adapt our transportation infrastructure to provide adequate service using substantially less energy and lower-quality fuels. In other words, within the next decade or two we must be able to provide much more energy-efficient transportation than is currently in use. A transportation system designed in such a way as to be readily adapted to more fuel-efficient modes of transportation such as rail would thus be expected to fail gracefully under this reasonable worst case scenario.

An even better and more conservative practice is to prepare for the likely shifts in transportation modes by introducing redundancy now. Relying on a single source of energy and single mode of transportation introduces two critical points of failure. Introducing one or more parallel transportation modes using different energy sources (electric-powered rail being the most obvious) will not only prepare for future transportation needs, it will also alleviate many of the current problems with the fragile oil-powered tree and hub-and-spoke topology used for modern road systems. Although redundant alternative modes may not be used at their full capacity right away, they will provide immediate relief from current and predicted level-of-service failures as well as relief from the growing failure of private vehicles on public roads to provide efficient and economical transportation service.3)

In short, it is reasonable to assume, conservatively, that abundant, cheap, high-quality energy may not be available over the expected service life of a civil engineering project. It would be irresponsible for engineers not to consider this as a potential cause of failure and to design for it appropriately.

Service Life and Suitability

Sustainable Energy Estimate for a Commercial Building

E_offices = {{e*I_solar*A*n_offices}/{n_A}} * {5{d/wk}}*{{8h}/{24h}}

  • Eoffices :: Sustainable energy consumption of a commercial building (kW⋅h/wk)
  • e :: Overall efficiency ≈10%4)
  • Isolar :: Regional solar insolation (kW⋅h/m²⋅d)
  • A :: Area of region served (m²)
  • noffices :: Number of workers in the building
  • nA :: Population of the area served

As another example, consider any inhabited structure such as a commercial office building. If the expected service life is longer than five or ten years, what will be the effect of high-quality energy prices more than quadrupling?5) Will the structure even be inhabitable; could it maintain a tolerable internal environment if its total energy consumption was limited to a sustainable level? Is it ethical for an engineer to guarantee a commercial or residential building will have a service life greater than twenty years when it is almost certain it will not be possible to supply enough energy over this time period to keep it cool enough to inhabit? Can one even claim such a structure is even suitable for its purpose?

It is considered good engineering practice to design for the worst conceivable conditions likely to occur within the next 25, 50, or even 100 years. There is no rational reason to believe our energy consumption can continue to exceed sustainable levels over this time period, and it is reasonable to assume this will happen even sooner6) and within the expected service life of most civil engineering projects being constructed today. Responsible civil engineers will conservatively assume no more than a sustainable level of energy and other resources will be available to operate and maintain the systems they design.

Consider, for example, a related situation. Suppose a farmer hired an engineer to design an irrigation system to pump a permitted amount of water from an existing but as yet untapped water well. If groundwater modeling showed the system was likely to start pulling brackish water within five years of being pumped at the specified level, it would be unethical (and in this case perhaps illegal) to claim a longer service life for the system no matter how long the mechanical and structural components are expected to last.7)

Conservative design assumptions must include the very real possibility of a considerable decrease in the amount of resources available to service and maintain civil engineering works. Engineers cannot claim an expected service life longer than a few years without addressing how the design will continue to function at sustainable levels of resource consumption, or at the very least will fail gracefully. Engineers working to address broader issues of infrastructure and regional planning must determine how to achieve life-cycle sustainability and avoid the possibility of catastrophic failures as available resources decline over the coming decades.

Designing projects likely to fail under the reasonable conservative assumption of declining resources is contrary to engineering standards of practice. Every work of civil engineering can and should be designed to consume only a sustainable amount of resources over its expected service life. Engineers should never claim, neither implicitly nor explicitly, that a project is sustainable without having done a quantitative analysis to support such a claim. This is not tree-hugging rhetoric; it's sound engineering practice.

2) It is beyond the scope of this paper to show how abundant supplies of cheap, high-quality (low-entropy) energy sources are no longer available and how within two decades, and likely much sooner, this will force a radical redesign of our entire transportation infrastructure to operate on sustainable sources. But even without proof shown here, the conservative assumption is to assume this will happen sooner rather than later.
3) San Antonio public bus ridership increased 9% from 2005 to 2006.—American Public Transportation Association, Largest Bus Agencies Transit Ridership Report, 2007-03-12, accessed 2007-04-22,
4) This assumes about 50% coverage and 20% energy efficiency.
5) A five percent decline in the supply of limited resource typically quadruples the price. See
6) Current estimates of fossil fuel supplies are about 10-30* years of oil, 30-40 years of natural gas, and 100-150 of coal, without curtailing consumption to mitigate global warming or to save some for later. Also note most predictions assume constant growth in the consumption of each fuel until depletion and do not consider the transfer of demand from one fuel to another and the efficiency losses of replacing a lower-entropy energy source with a higher-entropy source. See Energy Trends and Implications for U.S. Army Installations, U.S. Army Engineer Research and Development Center, September 2005, accessed 2007-04-28; and The History and Future of World Energy, Loren Cobb, The Quaker Economist, accessed 2007-04-28. *It is difficult to get accurate estimates of oil reserves since it is in the interest of most countries and corporations to overstate their reserves.
7) If the modeling showed it was also likely to render permanently useless other wells nearby, use of this system for its designed purpose would be worse than failure, and the engineer responsible morally, if not also legally.

Personal Tools