Appologies for the break in posting--the perfect storm of the birth of my second daughter and an extremely busy period at work have forced me to prioritize, and my writing didn't make the cut. However, I'm cautiously optimistic that I'll be able to get back to posting on a fairly regular (Mondays) schedule. I plan to dive right back in to where I left off on the "Diagonal Economy" series.
For now, here are the slides from my recent presentation at the Association for the Study fo Peak Oil conference in Denver, entitled "The Renewables Gap" (.pdf). If you want to watch the video of the presentation, you can purchase it at the ASPO website. I've posted a general text of what I said to accompany the slides below:
My talk is about what I’m calling the “Renewables Gap.” The basic question that I’m seeking to evaluate here is whether, and at what cost, it’s possible for our civilization to mitigate peak oil with renewable energy generation—specifically solar PV and wind power.
One important caveat before I get started—My goal is to explore the solution space of our future, not to predict exactly what will happen.
Slide 1: If we seek to mitigate peak oil with renewable energy, we need to first ask what do we need to mitigate. My answer: the decline in NET energy produced from oil, not the decline in overall production. This graph shows the decline in NET energy available from oil, taken from Dave Murphy’s previous presentation.
If, hypothetically, 20 years from now we’re producing 100 million barrels of oil per day, but it requires 100 million barrels of oil worth of energy input to do so, we have ZERO energy left for the operation of society at large. This is functionally the same as producing no oil at all.
Slide 2: What I want to quantify is the amount of net-energy that we need to replace going forward. A “classic” peak oil decline graph shows a plateau, followed by a gradually accelerating decline. Let’s consider why that’s so. What happens when we hit a plateau—as we arguably have now? The existing fields are declining at rates between 3% and 15% per year. But, because we’re scrambling to bring new production on-line, the overall level of production is buoyed for some time. We’re compensating for this underlying decline with more expensive oil—both financially and energetically. That keeps the level of OVERALL oil production steady, but the rate of NET energy production from oil is falling. That’s what this graph depicts.
For the purpose of exploring the solution space, let’s pick two numbers to evaluate: 5% and 10% annual rates of decline in NET energy production from oil. I’ll call these the “low” and “high” range scenarios. I’ll be discussing the potential to use renewable wind and solar power to mitigate this decline.
Slide 3: Specifically, I want to focus on some systemic effects of unique profile of solar and wind energy: the vast majority of the energy invested in these sources comes up front, before they ever begin to generate. Between 80% and 90% of the total energy ever required to build, operate, and maintain these systems must be invested UP FRONT.
I won’t discuss other renewables such as tidal and geothermal power, though their profiles are largely similar. I’ll also ignore biofuels and nuclear—hopefully we’ll have time to discuss these in the question period, but the bottom line is that I think they don’t significantly change the thrust of this presentation.
Slide 4: Another preliminary issue: these renewables produce ELECTRICITY, not oil. We’re talking here about using them to replace oil—let’s talk about conversion issues. How many GWh are needed to replace 1 mbpd of oil production?
A straight BTU-to-BTU conversion: replacing 1 million barrels of oil per day production, or 365 million barrels of oil per year, equates to 70.78 Giga-Watt-Years. Clearly, however, oil and electricity are not the same thing.
Some people have suggested that you only need 1/3 this much electricity to mitigate peak oil because oil fired electricity generation can be only 33% efficient. I think that modern oil-fired electricity is actually somewhere between 50% and 66% efficient, but we need to explain the validity of using the BTU-to-BTU conversion:
First, because we need to replace oil, not electricity, and because relatively little oil is used to generate electricity, it’s incorrect to use this oil-fired electricity efficiency number.
Second, our infrastructure is currently adapted to burning oil in many applications. Therefore, to the extent we want to use renewably-generated electricity to replace this oil, we need to adapt this oil-burning infrastructure to electricity. For example, if you want to replace transportation fuel with plug-in electric cars, you need to invest in significant new infrastructure in the form of cars, batteries, charging stations, etc.
Third, any form of mitigation using renewably-generated electricity will require significant additional investment in the transmission grid to handle higher loads and to balance or store electricity due to the variable availability of renewable generation.
I don’t know if it’s possible to calculate the exact energy balance here. However, I’ll argue that, in order to mitigate peak oil with renewably-generated electricity, we’ll need to generate effectively the same number of BTUs of electricity as we’re losing in oil. Maybe slightly more, maybe slightly less, but I think the BTU-to-BTU figure of 70.78 Giga-Watt-Years per million barrels of oil per day lost is pretty close.
Slide 5: Another argument is that we don’t need to produce as much energy renewably as we lose to peak oil because conservation and improved efficiency can largely make up the difference. There’s some truth here, but it’s only ½ the equation. That’s because two factors—population growth and the desire of the world’s poor to improve their standard of living—will cancel out some or all of the gains from efficiency and conservation.
As shown in this graph, if population increases according to various UN estimates, that alone could cancel efficiency and conservation gains of as much as 40%.
Additionally, at least 5 Billion people and growing want to “improve” their level of energy consumption to Western levels. In India, car sales are up 26% over last year, to 120,000 cars per month. Admittedly, these cars tend to be more efficient than in America, but this is new demand, and far more than cancels out the fact that the Tata Nano gets 56 miles per gallon. While markets or force may deny the world’s poor access to Western levels of energy consumption, the geopolitical consequences of such this disparity will only accelerate energy scarcity.
Slide 6: The key question is: how much up-front energy input will be required to build out enough renewables to mitigate the decline in net energy from oil production? We know how much energy must be produced to meet this target, so the answer to this question is a function of the EROI and the lifespan of our renewable options.
You’ve just heard David Murphy’s excellent presentation on EROEI, which highlighted many of the same issues involved here. What I want to focus on is this concept of boundary.
We could talk about this boundary and EROEI calculation issue until we’re all blue in the face—my intent here is not to argue that some specific number is correct, but rather to point out the uncertainty and potential range. At the lower end of the range, I’ve proposed a proxy of price to account for ALL inputs and outputs. There are significant problems with this methodology, such as dealing with financing costs, but it has the distinct advantage of allowing us to account for all inputs—regressed infinitely—rather than drawing some sort of artificial boundary. IF you look at modern wind turbines using the price proxy, you get something like an EROEI of 4. I’ll call that my “low” value.
Now let’s consider more conventional calculations. Wind seems to be most promising at the moment, and I’m looking specifically at a 2009 paper by Kubiszewski, Cutler, and Endres entitled “Meta-analysis of net energy return for wind power systems.” The authors review 50 different studies of wind EROEI. In a section entitled “Difficulties in calculating EROI,” they make this statement:
“Studies using the input-output analysis [one method of calculating EROEI] have an average EROI of 12 while those using process analysis [another method] an average EROI of 24. Process analysis typically involves a greater degree of subjective decisions by the analyst in regard to system boundaries, and may be prone to the exclusion of certain indirect costs compared to input-output analysis.”
What I take away from that is that there seems to be a range of 12-24, but the authors—a highly respected group—suggest that the “24” figure fails to account for many inputs. That suggests to me that an EROEI of 12 is more accurate.
For our purposes, though, my intent is to explore the solution space, so I’ve selected what I think is an optimistic upper “high” EROI value of 20. I think this is unrealistically high—especially because this figure doesn’t even account for the intermittency, transmission, and storage energy costs that must be considered in such a large-scale societal transition—but for now let’s use 4 and 20.
Slide 7: How much energy must we invest if we want to ramp up renewable generation to keep pace with declining net energy from oil? This graph answers that question using a 5% net energy decline and a renewable EROEI of 20.
In this scenario, to mitigate the year-1 decline in net energy from oil, we’d need to invest 467 GWy of energy in year one without any production in return—that’s the equivalent of almost 7 million barrels per day. Then in year two it’s about 130 GWy more invested than cumulative production to that point, or about a 2 million barrel per day deficit. Not until year-three will the cumulative renewable generation be more than the investment deficit for that year—meaning that not until year 3 will we begin to have surplus energy available to mitigate the actual decline in oil production (which by this point leaves us 12 million barrels per day behind the peak oil decline curve.
That’s the “Renewables Gap.”
Slide 8: Here’s the pessimistic quadrant – 10% net energy decline, and a renewables EROI of 4. Here, the up-front energy investment is more than 4,600 Gwyears in year one. That’s 58 million barrels of oil per day diverted to renewable energy production. Plainly impossible. And the level of renewable energy production wouldn’t even catch up to the level of energy invested EACH YEAR until year 7.
Slide 9: Here you can see the boundaries of the Renewables Gap—the optimistic assumptions on top, pessimistic on the bottom. The lines represent, under each scenario, the net energy supplied by oil, minus the energy invested that year in building renewable energy production, plus the energy produced that year by the renewables brought on-line to date.
To be sure, we can slow the initial rate of investment in renewables in order to lessen this dramatic initial impact, but that option results in falling even further behind the net energy decline curve. We can also bootstrap the energy produced by renewables to provide the energy required for the next round of renewables—if the EROI is 20, this will work to some extent, but it will still have the effect of making us fall even further behind the decline curve. If the EROI is 4, it’s simply unworkable—we never catch up.
Is it theoretically possible to close this gap more quickly? Sure—by investing more energy up front, which actually serves to exacerbate the problem over the short term. We’ll be chasing our tail. It might be possible to catch up—to make a significant public sacrifice up front and kick start the program—IF the economy as a whole is healthy. The Renewables Gap puts us in a Catch-22 situation: using renewables to mitigate peak oil will make the situation worse before it makes it better. Our ability to absorb the up-front costs of transitioning across this gap is a function of our economic health, but to the extent that our economy remains healthy enough to do so we are unlikely to muster the political will to address the problem.
Just to provide some context for the size of this gap: Under the optimistic scenario, this is the equivalent of adding one new China to world oil demand immediately, and maintaining that for many years. Under the pessimistic scenario, this is the equivalent of adding more than 9 new China’s to world oil demand.
Now I recognize that there are energy conversion issues, there are calculation issues, there are timing issues—simply too many variables to make any definitive statements here. But what I hope I’ve highlighted here is this CONCEPT of the Renewables Gap problem, and the uncertainty of our ability to bridge that gap.
Slide 10: As a civilization, we still have a small and shrinking bank of net-energy surplus with which to build our future. We have to make tough choices about how to spend it. Perhaps our most fundamental choice will be this: do we spend it attempting to bridge the Renewables Gap—despite our uncertain ability to do so? Or, at the risk of using the phrase “Paradigm Shift” in serious conversation, do we cut our losses with the “perpetual growth project” and consider if that energy could be better spent building a fundamentally different future?
I don’t mean to make any secret about my own views here: it seems unlikely to me that we’ll be able to continue “Business as Usual” via massive investment in renewables. It seems sufficiently unlikely that I don’t think it’s the best way to spend our civilization’s limited and shrinking supply of surplus energy. I think that energy will be better invested in the infrastructure needed for communication within and transition to a much more locally-self-sufficient, topologically flat society. I’m not here to tell you that my vision of the future is somehow “right,” or that other visions are “wrong,” but I am here to suggest that ANY vision of the future predicated on a transition to renewable sources of energy to mitigate the decline in oil production must first address this Renewables Gap.
Readers may also find my litigation checklist of interest.