Jacobson and team have just released a new study covering 100% renewables for 143 countries representing 99.7% of fossil fuel CO2 emissions. It’s an update and maintains the mix of technologies, omitting nuclear and CCS. Expect more pushback from people who don’t accept the empirical realities related to those technologies. My assessment of various aspects of the report will be broken down into three chunks of roughly equal size around specific subsets of the topic.
The first covered the global costs and massive savings associated with the transition. This piece covers off some interesting edges of the study related to storage and transmission, highlighting the inherently conservative choices Jacobson and team are making in their assessments. The third covers off the interesting question of how people will perceive the study’s explicit support and expansion globally of the Green New Deal’s core electrical generation and electrification targets.
For the record, I’ve done my own, less sophisticated assessments multiple times of nuclear vs renewables and multiple carbon capture approaches and agree completely with Jacobson and team that they are unnecessary, uneconomic, and in the case of CCS almost entirely harmful to progress. My net is broader, however, as I also look at industrial and land use processes in addition to energy, and see at least one place in industry where it will be hard to avoid some carbon capture. But for energy generation, there is no merit to CCS vs simply eliminating fossil fuel use.
“… only 9% more generator nameplate capacity is needed, in the 143-country average, to meet time-dependent load than to meet annually averaged load. Storage is also needed to meet time-dependent load”
This ties back to the much smaller amount of energy we need when we don’t throw away the majority of it, but I wanted to draw out a different point, that related to storage. I’ve know for years based on my reading, analysis, and discussions that we need much less storage than most people realize (and certainly much less than detractors assert) and that it won’t be an unreasonable cost. I also know that it’s more of an end game requirement for decarbonization, not a must-have at the beginning of the transition.
Let’s start with the selection of storage technologies per Table S13 from the report from Jacobson et al.. It’s incredibly conservative and skews to the expensive types. When I looked at it, a few things leapt out. First, that it ignored a couple of obvious technologies: flow batteries and the use of vehicle to grid (V2G) rolling batteries on wheels we call cars and trucks. I’ve been going deep on flow batteries (expect a series on the subject) and know that they have complementary characteristics to lithium technologies at a lower cost, making both useful. Similarly, I’ve calculated that there will be roughly 20 TWh of batteries in light vehicles with perhaps a quarter of that easily available for V2G applications in addition to their obvious demand shifting models.
The second thing that leapt out was the mix of storage Jacobson and team posited.
For the purposes of this discussion, I assembled the US-specific table above from the much larger Table S13. Recently, I’ve been on a pumped hydro tear, triggered by the Australia National University study on global resources. It showed that the US has 250x the resource capacity as is required. Further, it’s much cheaper than lithium battery technology and cheaper than flow batteries. And, of course, it’s rock solid, proven technology, having first been built in the 1890s.
My emerging assertion is that lithium technologies will be very useful for in-day storage for rapid grid balancing and to shift daytime solar into the evening, roughly 4-12 hours of storage. Flow batteries will overlap with 8-24 hours of storage and be very useful for day ahead reserve markets. Finally, pumped storage will be useful a bit for day ahead, but especially for the longer term storage, so its storage will be measured in days and weeks.
And to that point, Jacobson only has the highest cost solution. My hypothesis was that for simplicity of modeling, he and his team selected a minimal set to prove the point, and that they weren’t being prescriptive of specific technologies in this space. Jacobson confirmed this to me through email, saying:
“There are, in fact, many solutions rather than the one we showed. We just wanted to show one solution that was defensible. Including more PHS, V2G, and other battery types will make the problem even easier to solve.”
Storage is a solved problem. Even the most expensive and conservative projections as used by Jacobson are much, much cheaper than business as usual, and there are many more solutions in play.
“… sensitivity tests were run in Figure S13 of Jacobson et al. to check the impact on cost of different fractions of wind and solar power produced subject to long-distance transmission. The result was that, if congestion is an issue at the baseline level of long distance transmission, increasing the transmission capacity will relieve congestion with only a modest increase in cost.”
One of the key aspects for an all-renewables grid is transmission, especially continent-scale grids. At the scale of North America or Europe, there is sufficient offshore and onshore wind, solar and hydro to balance the large majority of needs the large majority of the time. Storage makes the remainder manageable. However, one of the many talking points of those who claim that renewables won’t suffice or will be too expensive is that grid congestion is impossible to overcome or that new transmission cannot be built economically.
Jacobson’s study lays both of these to rest. First, it models the requirements and finds that congestion is a lower concern than stated, and second that just as with storage, additional transmission requirements are much smaller than many anticipate.
“Costs are highest in small countries with high population densities (Taiwan, Cuba, South Korea, Mauritius, and Israel). Nevertheless, the 2050 WWS private energy cost per year in all five regions is 43% to 65% that of BAU”
But continent-scale grids don’t solve the problem for nations which are isolated by oceans or politics from easily trading electricity with their neighbors. Israel’s independence is hard fought for in an often hostile region and a region subject to fairly regular conflict which destabilizes nearby countries.
South Korea is isolated from the rest of Asia by North Korea. Cuba is both close to the US geographically and far from it politically. Japan is missing from the short list, but undoubtedly has similar challenges, resting as it does 200 kilometers from the differently isolated South Korea, and about 800 kilometers from the nearest point of China.
Jacobson’s study doesn’t downplay this, but once again leans into the conservative costing, one which finds that even sticking to grid isolation, the direct and indirect costs of a WWS grid are lower than business as usual.
And similar to the storage question, this is also a case where existing technologies and proposals are in hand. In this case, for many of these places HVDC can assist in substantially lowering the costs of 100% renewables. As I pointed out a couple of years ago in my assessment of the technology, it traverses bodies of waters with much lower costs than high-voltage AC, typically achieving payback at 50 kilometers and gaining advantages with each additional kilometer. China is engaged in building out an Asian HVDC grid to share renewables with neighboring countries including South Korea and Japan. They’ve proposed a trans-continental polar grid to share wind and solar across the northern hemisphere.
The trend in Jacobson’s study as each point is looked at is that he and his team aren’t suggesting anything radical, but a very conservative approach that is possible to massively improve upon.