In the next few posts I am going to go through the thoughts in my own head and explain how I’d go about developing a risk mitigation strategy around Anthropogenic CO2 Emissions.
In this post, I’m going to be assuming as true the four incontrovertible points that Skeptics and Believers both (for the most part) agree upon. If you disagree with one of those four incontrovertible points, then obviously you will disagree with my chosen mitigation strategy.
Also, these next few posts will be a response to three comments that were made in the past. One from Agellius, one from Eric, and one from Geoff. Their comments will allow me to discuss risk mitigation in more detail and explain why a risk mitigation approach is an appropriate one for even for AGW Skeptics. (Particularly if they consider themselves to not be in the throes of a religious war.)
First, let me quote Agellius’ respectful question:
What I think you neglected, and what would have been helpful to me, is to explain why we should worry about increased CO2 emissions. You say you have made the case that ‘our science can’t really connected Global Warming with Anthropogenic CO2′. Well, what can we connect it with? In other words what, specifically, do we have to worry about? Is it simply the possibility that it could be bad in some unknown way?
I believe that Eric James Stone was making a similar point when he said:
If the amount of happiness in the world were growing with no chance of cutting back without intervention, then by your definition, it would be unsustainable. Therefore, by your logic, the correct moral choice would be to immediately find ways to start cutting back on happiness, rather than waiting for a problem to show up.
Risk Mitigation Explained
So now let’s talk about how to perform Risk Mitigation. As a Project Manager, I take each project and I measure the known risks for the project. (There are also unknown risks, of course, but you can’t measure those by definition because you don’t know about them.) The typical way you measure a risk is by assessing two things about it.
1. What is the probability of the risk
2. What is the level of impact if the risk becomes an issue
For AGW, we have to consider a third factor — duration required to implement the mitigation strategy. This is because in the case of Anthropogenic CO2 mitigation, we have a potentially very long lead time for our mitigation strategy to take effect.
Now for a software project, you would probably try to estimate the probability as something like ‘high,’ ‘medium,’ and ‘low.’ But sometimes you try to get more specific. You might throw in ‘nearly impossible’ or ‘nearly a given’ for example.
Then you measure the impact to the project if the risk is realized and becomes an issue. The impact might be that the project will cost an extra $10,000 (small) or that it will cost an extra million (probably big.). Or it might be catastrophic failure for the project (really big) or even catastrophic failure for the company (really really big) or cause a loss of life (really really big.)
Once you’ve estimated the probability and the impact you “multiply” them together and create a final risk score.
For example, let’s do the following to keep it really simple:
1 = Nearly impossible (less then 1%)
2 = Low probability (1% to 25%)
3 = Medium probability (25% to 60%)
4 = High probability (60% to 99%)
5 = Nearly impossible (99% to 100%)
And let’s measure impact like this:
1 = Will cause virtually no impact beyond inconvenience or financial impact up to 1% of the project cost.
2 = Will cause financial impact that will put the project’s 10% target tolerance at risk for budget or schedule.
3 = Will cause financial impact beyond the target tolerance
4 = Will cause financial impacts significantly (30% or worse) beyond the target tolerances
5 = Will cause catastrophic failure for the project (i.e. will never be useful), loss of life, or damage to the companies ability to do competitive business.
Now in real life, the 1 to 5 scheme is kind of hokey. For starters, if we had a 0.9% risk that was going to cause loss of life, we’d want it to have a much higher risk score than a mere 5. So we’d probably allow for a better way to weight impacts. But you get the idea even with this overtly simplified example.
We’d then write down each risk, assign it a value on probability vs. impact, do the math, and come up with a final risk score.
All risks must have an assigned action. Sometimes we’re allowed to put ‘no action’ if the score is particularly low. But any action other than ‘no action’ will have a direct or indirect cost to the project budget. In fact, we do a risk mitigation session before we come up with a project budget in part because we need it to create the correct project budget. The final risk score helps determine how much money we are going to spend on mitigating the risk.
For example, if possible loss of life is involved, we’ll probably have 10 testers to every programmer, or something like that. But if we only have $10,000 at risk and the probability is medium to low, it probably only makes sense to spend up to $1,000 to avoid risking it. Anything beyond that lacks a good ROI.
The Problem of Making an Estimate
Now one thing that has always bugged me about project risk mitigation is that the ‘probabilities’ I assign to things are really just a gut feel. Most of the time there is no way to know what the probability of a certain risk is. Furthermore, I always disliked that I’m often using ‘probability’ to mean something that will or won’t happen in the future – I just don’t know which it is.
Nevertheless, even if you can’t really measure the probability, the important thing is that we make our best gut feel and do our best to try to react to a given threat to a project. Beyond that, there is nothing else you can do.
My Own Risk Rating for CO2 Emissions
I’ll go into more detail in my next post, but for now, here is how I rate Anthropogenic CO2 Emissions as a risk.
Probability: In the short run, very low. I am an AGW Skeptic. I do not believe Anthropogenic CO2 Emissions has currently caused any problems and I do not believe any sort of massive problem is imminent. 
However, in the long run, I rate this probability as 100% if given conditions (i.e. continual growth of Anthropogenic CO2) don’t change. I know this to be the case because CO2 is known to be a greenhouse gas. This is not in dispute. CO2 has, in the past, played a significant role in heating the earth up. So run away growth in CO2 <i>will cause a problem with global warming at some point.</i> What is being debated is what that point is. The scientific consensus is that we’re close or even past the disaster line. I am assuming they are wrong about this for my risk analysis. But I know they are right at some future point.
Because CO2 is a greenhouse gas, eventually it will cause out of control global warming. I just don’t know how far off that problem is. It might be, for all I know, millennia (though I doubt that). Or it might be much closer than that, say five to ten decades.  I’m going to go with my gut feel that we are no less than five decades away from a problem if we do not curb current growth and probably no more than two centuries. 
Impact:This one is easy. Catastrophic global failure. Also, other bad things will happen before that point. Acidic oceans, for instance, or distruption of the food chain.
Time to Implement: If you don’t mind destroying the economy or civilization as we know it, time to implement a strategy to stop CO2 Growth can be very fast. I consider that unacceptable, so I’m not even going to consider it. If you don’t care about CO2 Growth, but only global warming, geo-engineering can be very fast. But we are right to fear that strategy, so it’s going to end up being only my contingency plan. (More on that later.)
If I am looking only at Anthropogenic CO2 Strategies that do not destroy the economy or civilization as we know it then we are looking at very long lead times. Based on my reading of Stephen Leeb’s The Coming Economic Collapse I’m going to guess that it would take no less than five decades to implement a CO2 Curbing strategy without an economic collapse. Probably longer than that. 
Conclusion: Results of My Risk Analysis
So where does this put me personally? 100% chance of global disaster, but not for a long while yet. I’m guess between 50 and 100 years before anything truly disastrous happens, given current CO2 Growth rates. However, we need a lead time of at least 50 years, probably more like 100 years. Therefore my mitigation strategy calls for immediate action today.
I also assert that unless I can be assured that there is either no impact, no probability, or nothing I can do about it anyhow, that I should always have a risk mitigation action in place. If I don’t, I’m not doing risk mitigation competently.
Everyone do your own mitigation strategy and see what you come up with. Put the results in your comments.
 Of course I might be wrong and doom might be imminent. Since I believe our science isn’t matured enough on this to tell us one way or the other with high reliability, I can’t discount that possibility. However, a risk mitigation is what I believe is going to happen. So for my mitigation strategy, I’m going to go with my sincerely held belief that we are about a century away from any global disaster. When you do your own risk mitigation strategy, you can decide for yourself what you believe.
 The current issue (Sept 2010) of Scientific American, who are staunch AGW Believers, list a 1 in 2 chance (they are assuming we continue to act against it) of run away AGW in two centuries. So I’m allowing a two century upper limit on the grounds that an overwhelming AGW Believer rag, that obviously feels it’s actually much closer, put that as their upper limit. I am, if you will, assuming the best case scenario from the AGW Believer stand point.
 The Coming Economic Collapse. I read a book called The Coming Economic Collapseby Stephen Leeb. I do not agree with this book’s conclusions. This is a book about how, in the author’s opinion, we’ve reach peek oil and because of that the world economy is going to collapse because we can’t possibly meet demand for oil until we switch our entire global energy infrastructure over to use alternative energy sources. It is not a book about Global Warming or CO2 Emissions at all.
The reason I disagree with Stephen is because I think his evidence for peek oil is questionable. I do not doubt his evidence for how long and difficult it will be to switch over to alternative energy sources.
According to Stephen’s research, once we reach an economic disaster due to the price of oil spiraling out of control, it will take over five decades (maybe a lot more than that) to switch over to the point where oil prices come under control again.
So I’m going to propose that if you want to switch over to alternative energy sources without an economic disaster that Stephen’s five decades (and remember he’s assuming an economic disaster) is probably a good starting assumption.
If you think its 30 years instead of 50, this changes virtually nothing. Probably anything more than 10 years works out to still be a pretty good case for taking immediate action now to start the ball rolling since a 10 year economic project is just too far out to be trust worthy. Deciding that we have, say, 30 years before we have to worry, so we can wait 20 years before starting to take actions has so many levels of unknown that it’s still scary. That’s why I advocate a 50 year plus plan for switch over. Better to be safe than sorry on this. Plus the longer I plan to take, the less the economic impact will be. So I’d rather have a very long lead time.