Emailing "The Problem with Carbon Counting"
With all the discussion about carbon cap and trade dominating the regulatory environment, it's easy to lose sight of a vexing question -- how do you count carbon? After all, carbon itself is simply a molecule, an invisible and pervasive element that is a part of virtually all human activity. Counting dollars and sense, or gallons, or yards, or meters, or kilograms, are all understandable, but what about carbon? And even if you can settle on how to measure carbon, there's the related question of how you measure carbon output for any given commercial activity.
Take, for example, the problem with
milk:
Several studies in various countries have already sought to tally the impact of milk from its production on a farm to the disposal of its carton. In between, the studies try to measure such intricacies as the energy used to make the fertilizer to grow feed for the cows, to fuel trucks delivering the milk, and to power refrigerators cooling it in kitchens.
It isn't surprising that each of these studies sizes milk's footprint differently, in large part because each varies in the way it counts one or more of those factors.
Supermarkets around the world are attempting to quantify what sort of carbon impact each gallon of milk produces. Factors that affect that number include the type of machinery used in farms, what sort of feed the cow eats (which in turn affects bovine belching), the weight of the cow, and the relative economic value of the milk versus other parts of the cow such as meat and leather.
The biggest problem, in my mind, is the lack of standard way of counting. The article linked to above mentions at least four different methods of counting carbon in producing milk: a private supermarket, a British NGO, the U.S. dairy industry, and the International Organization for Standardization.
Without standards, comparison is impossible. A gallon of milk with "abc" carbon emission listed on its label simply can't be compared with another gallon of milk from a different producer with "xyz" carbon emission. And without some sort of standardization, government regulation to clear up the confusion seems all but inevitable.
The exact same problems appear to affect Green IT as well. Not much carbon counting is happening today, but when it comes, how will carbon be counted? What variables, from producing server equipment to network traffic miles traveled to trucks and machinery used to lay down equipment to electrical operating requirements, will be counted? The problems with milk point to much confusion for Green IT carbon counting in the months and years ahead.