Scalability of environmental solutions
Oct 15, 2010

Scalability of environmental solutions

One of the largest challenges that faces the environmental movement is scalability of its solutions. By this I mean the difficulty of taking a solution that works well when a small proportion of the population utilize it and having it work when a large proportion of the population uses it. Indeed, localization is a key aspect of many environmental proposals, however there are challenges to having a local paradigm be globally applied and widely supported. That said, I don't mean to minimize the viability of environmental solutions with scaleability problems merely to acknowledge that this is often where their largest impediment lies and one focus of our attention ought to be in resolving them.

Perhaps the clearest example is that of electric cars. When electric cars are used by a small proportion of the population, they are excellent. Especially if the regional electricity infrastructure is largely hydro, nuclear or other clean energies than the small influx of extra capacity needed can be absorbed in a clean way. However, if we imagine this solution scaled such that a majority of people used electric cars then an enormous increase in primary electricity energy sources would need to be found to satisfy the demand. Such energy sources may well be superior in cleanliness or efficiency to gas engines however the point is the majority of the problems with electric cars as a viable solution are in this scaleability problem.

Food production suffers a similar issue. The importance of buying local is heralded for environmental concerns, health and lifestyle reasons, antiglobalization arguments and the like. For a big city, this is excellent as long as a small number of people do it but suffers sole significant challenges when scaled to a large proportion of the population. When a city like Vancouver with perhaps three million people in its surrounding area the carrying capacity of its immediate "local" area (surrounded by mountains as it is) simply isn't sufficient to feed the entire population with a varied and consistent diet. As long as we are going to have a society with large population centers there will have to be delocalized food distribution networks to supply them.

There is also the issue of efficiency of food production. For all it's numerous failures, factory farming of, say, grains on the prairies is very efficient in terms of manpower and financial cost. Indeed, one of the key factors of the development of civilizations is the ability to reduce the dependency on food production and to engage in other pursuits. Our western civilization made huge gains through things like the green revolution, today requiring a mere 1.3% of people to work in the production of food. Many factors go into this, but the ability to produce food in the best areas and ship it long distances as well as the ability to centralize and corporatize the production of the food is, if nothing else, efficient. The idea of small local farms nearby major population centers often reduces the net efficiency. Likewise, some of the environmental desires like organic products,  free range meat products and the like are less efficient from a cost and manpower perspective than their conventional counterparts. The result is that when a large proportion of the population opts for the solutions of localized and environmentally consciously grown foods, significant increases in costs and decreases in efficiency may well occur.

Clean energy production is a another example. Take Ontario which has arguably North America's most competitive subsidization of environmental technology starting up through guaranteed "feed in tarrif" prices for wind and solar. Since currently these technologies are more costly than other dirty energy sources, the tax subsidy makes these other energy sources economically viable. While the total amount of such programs is still a small percentage of the total, the cost to taxpayers are low. However, should such a program be applied where wind and solar was, say, half of the total energy than the costs would be staggering at this scale.

Now I should be clear that simply because there are issues with scalability does not mean we should reject small scale solution on hand nor that they are not beneficial. It merely means that these issues must be acknowledged and address head on. In many cases, while a solution may not be able to scale it doesn't minimize the benefit it is giving. So for example, rooftop and balcony gardens are never going to be able to feed all of Toronto, but they can reduce the environmental and energy footprint of the city in a meaningful way and should be applauded nonetheless. Additionally, while there are often barriers to scaling solutions to larger levels, there can also be advantages. As solution achieve economies of scale, they become more viable and efficient. In the feed in tarrif example, it is true that the program as it is with the current costs and subsidies could not scale. However if the program jump starts investment in green technology which after time achieves economies of scale then the subsidy can be lowered and eventually eliminated. Despite the difficulties, I retain considerable optimism that localized small scale answers to many problems can achieve large scale solutions.

Thoughts on this post? Comment below!

Share this post:

Tweet It! Facebook Add Feed Reddit! Digg It! Stumble Delicious Follow

Post a Comment

Frequent Topics: