The J.P. Morgan Annual Energy Paper for 2015 is an excellent short resource “A Brave New World: Deep De-Carbonization of Electricity Grids”. They have packed a lot of data and analysis into 28 pages. The focus is Energiewende and Caliwende (the California version of Germany’s Energiewende). The high quality of this report is due at least in part to guidance from Armond Cohen, Executive Director and co-founder of the Clean Air Task Force. And of course Vaclav Smil. Chief Investment Officer Michael Cembalest closes with this:
Deep de-carbonization of the electricity grid via renewable energy and without nuclear power can be done, but we should not underestimate the cost or speed of doing so in many parts of the world. At the minimum, the costs involved suggest that efforts to solve the nuclear cost-safety puzzle could yield large dividends in a post-carbon world. Such is the belief of the scientists, academics and environmentalists who still see a substantial role for nuclear power in the future (see Appendix V). See you next year.
The report gives enough detail that you can see why Germany’s nuclear ban leads to a shocking cost of avoidance of $300. I’ve circled in green the baseline Energiewende result estimated to cost $300/mt CO2. J.P. Morgan modeled a balanced deep decarbonization strategy, which using 35% nuclear, costs only $84/mt CO2.Note that the $300 is a bare-bones estimate – none of the cost of the additional transmission infrastructure required by high-renewables is included in the analysis. Even so the baseline Energiewende plan will double already second-highest in Europe current costs from $108 to $203/MWhr.
What about Caliwende? The cost to consumers is lower than for Energiewende but the CO2 avoidance cost of the baseline plan is $477/mt CO2 — even worse than Germany because California has already done more CO2 avoidance. Happily, if California implemented a balanced plan (35% nuclear) that drops the CO2 avoidance cost to $174/mt CO2. That is still unnecessarily expensive because of the high-renewables ideology.
Well, at least both plans have closed a lot of those nasty fossil plants, right? Actually not. Because of the intermittency, at least all of the current thermal generation is required to cover the demand gaps. These charts show just how big those gaps are. This is the largest single source of the high CO2 avoidance costs. All that mostly-idle thermal capacity is still required by the ideology of high-renewables. That means a very small capacity factor so the capital has to be amortized over too-few generation hours.
What does it all mean? Back to Michael Cembalest (his emphasis):
- Intermittency greatly reduces the importance of wind and solar levelized cost when assessing high- renewable grids. The cost of backup thermal capacity and storage is an inextricable part of any analysis of a high renewable system. Academic and industry research has reached similar conclusions. A 2015 paper from the Potsdam Institute for Climate Impact Research notes that integration costs in systems with high levels of renewable energy can be up to 50% of generation costs, and that the largest single factor is the additional cost of backup thermal power.
- Energy storage reduces CO2 emissions but its cost, utilization rate and energy loss must be accounted for. Even when assuming continued learning curves, storage adds to net system cost
- Even in California, there are uncertainties to this Brave New World: California’s Independent System Operator gave a presentation in 2014 highlighting how the impacts from increasing renewable energy on the grid are still not fully understood. They mentioned voltage fluctuation due to upward/ downward ramps, high voltage issues on distribution circuits, voltage/power regulation control issues, the greater number of operations and increased maintenance on voltage control, etc.
I appreciated the final paragraphs which concisely dispense with some of the common tooth-fairy stories. There are also appendices backing up these points:
We often hear people referring to other what-ifs regarding high-renewable grids. Many rely on highly uncertain assumptions and conjecture, while others neglect related costs.
- Could cross-border integration of high-renewable grids reduce the need for backup power and its corresponding cost? That’s the next wave of renewable energy research. It would cost money to build these interconnections, but in theory, if wind and solar patterns are more divergent the larger the geographic area covered, the problem of renewable intermittency could simply be diversified away. Unfortunately, new research on wind suggests that this theory has major limitations. This remains a premise best proven empirically rather than by assumption.
- What about over-building renewable energy and storage so that the need for and cost of backup power is eliminated? The good news: it’s an emission-less system. The problem is that incremental solar, wind and energy storage costs would dwarf foregone costs of backup thermal power. Our models determined that a system in California with enough wind, solar and storage to eliminate backup power entirely would cost $280-$600 per MWh, which is 2.5x – 5.0x more expensive than Caliwende (depending on assumed storage system properties and costs). Bottom line: a renewable energy storage version of the Temple Granaries looks to be prohibitively expensive.
- Why not draw on electricity stored in electric car batteries (“car-to-grid”) to reduce storage costs? Another theoretical possibility that’s only worth discussing when we can determine the penetration rate of plug-in vehicles, the participation rate of drivers willing to share their battery with the grid and how much of it they would share, the cost of interconnections, and the cost of incentives required by drivers to have their expensive car batteries cycled more frequently. See Appendix VII.
- What about “demand management”? If demand could (somehow) be reconfigured to match up with variable renewable generation, unused surpluses and demand gaps would be smaller and system costs could decline. However, demand management is meant to deal with intraday supply-demand issues, not intermittency issues which span weeks and months. See Appendix VIII.