In the beginning I asked whether the program is really capable of doing what it says it can, namely removing the CO2 that results from powering Dell PCs. I identified four areas of concern, and we’ve touched on two of them. Today’s focus is on the scalability of the program, or, in other words, if it’s successful, can it really keep up with the volume of CO2 that is generated.
If you recall, in an earlier post we did a spreadsheet to look at the amount of CO2 that would have to be sequestered each year. That analysis showed that, assuming no growth in Dell’s business, we’d need to sequester about 50M tons of CO2 per year in order to cover all of their PCs. Furthermore, Dell is a little under 20% of the PC market, so we’d need to do around 250M tons to cover all PCs (assuming Dell’s systems are a good estimate of the rest of the PC industry).
So first, lets start with the simple calculation of land required to sequester the CO2. I found lots of different estimates for CO2 per acre of forests. I’ll use the high end of the EPA number of 2.6 metric tons of carbon per acre per year, or approximately 10 tons of CO2 per acre per year. This means we need 5 million more acres of forest to handle the CO2 from Dell’s PCs, or about 25 million acres for all PCs (note that with no growth, we only need the new forest once - it will sequester enough per year to keep up for a long time). At 640 acres per square mile, we’d need around 10 thousand square miles of new forest, or approximately the land are of Maryland or Vermont.
According to this FAQ, all of the US today sequesters about 840M metric tons, covering roughly 15% of the total CO2 emitted each year. So we’re talking about adding another 30% capability to handle this. Of course the rest of the world can help sequester as well. Putting this all together, the numbers look big, but not impossible.
However, as I researched this topic, I ran across a slightly different issue. According to the EPA, the amount of carbon that can be sequestered by a forest is greatly effected by how much money is set aside to manage that forest. Ultimately, in an undermanaged forest the carbon is re-released into the atmosphere, negating the original intent. So the question we need to be asking isn’t whether CO2 sequestration in trees can scale, but whether it can scale at $6 per ton per the Dell program.
From Table 4-5 of the EPA report the role we care about is ‘Afforestation’, aka ‘planting trees’. This table confirms our results above, namely that we could plant enough new trees in the US to sequester the worldwide PC energy for all vendors. However, it also says that at $6 per ton, we can only cover less than 10% of the CO2 emissions of just Dell PCs. And while this analysis is only counting sequestration opportunities in the US, the numbers are still way off, and remember that this doesn’t include any growth for Dell’s business or PC power.
So to summarize, I have to conclude that this program does not scale as currently designed. There appears to be the land capacity to support the program, but the program is currently not setting aside enough money in order to pay for the land management required to maintain sufficient sequestration capacity.