Cavalcade of Mammals

In discussions of wind and solar power, sensible centrists always pop up with, “We must build new nuclear power plants too!”. And then mumble on about how nuclear power isn’t really dangerous, especially new designs, and talk about how nuclear power provides steady base load power which is necessary because wind and solar are intermittent.

For example, see this Freakonomics Radio podcast hosted by Stephen Dubner, which is noteworthy for never meationing the cost or relative cost of nuclear power. No economics in Freakonomics! And in a more reasonable discussion between host Ezra Klein and Jesse Jenkins covering a host of energy / decarbonization topics, nuclear power is boosted as a necessary component, again without a discussion of costs.

But this idea that nuclear power is necessary and complementary is mostly nonsense. Yes, nuclear power has killed very few people, and compares favorably in overall safety to coal power plants which cause plenty of deaths due to air pollution. But this argument is almost entirely off target.

The intermittency of wind and solar power is a big issue. Working out solutions for providing steady power in a grid powered mostly by wind and solar is the challenge for the next generation or two.

The thing is, nuclear power doesn’t help with that. Nuclear power plants are run full out except for maintenance (a capacity factor of 92%). What’s needed to complement solar and wind are power sources that are dispatchable and can be ramped up and down quickly. Hydroelectric power provides that in places like the US’s northwest that have lots of dams. And today gas peaker plants and coal plants provide fast and slow power that can be ramped up and down as needed.

And nuclear power is expensive, very, very, expensive. Today nuclear power costs 3-4X as much as solar and wind power. And that is market cost, excluding the subsidies provided by the federal government for nuclear power. Nuclear plants are insured by the US government. The costs of a meltdown are immense, from billions to hundreds of billions, and with a chance of a nuclear plant disaster of at least 1 in 165 over the life of a plant, the risk is substantial. Long term high level nuclear waste disposal has not been paid for or figured into costs–US nuclear plants store high level waste on site, along rivers and coasts, with the US government expected to handle final storage. And nuclear plant decommissioning will likely cost more than the collected funds account for.

So nuclear plants don’t make economic sense on their own and they do not complement wind and solar power generation.

What is needed to make a power grid with wind and solar the primary power sources able to provide reliable power? There needs to be ways of meeting short term (minute to minute), medium term (hourly and daily), and long term (days and weeks) interruptions in wind and solar power production. Short term irregularity can be met today with small grid storage and hydropower.

Dealing with the daily cycle of solar power production requires much larger grid storage, generally not available today, and/or large scale demand shifting not done today. Short periods of low wind are fairly common, and week- or month-long regional low wind is known to occur. Solar power production is lower on cloudy days, and varies seasonally.

It is not clear today what solutions will be used. Grid scale power storage is an active and promising R&D area. Over-capacity–having more solar and wind capacity than is needed will help, and solar and wind are already cheap enough for it to be economic, but this creates a new problem–what to do with the excess power generated during high periods.

Demand-shifting has a lot of promise, and will help with hourly and daily power demand balancing. Residential and industrial power use modulated by utilities is already in widespread use, mainly used to shave off peak demand and do modest demand shifting today, but there is much more potential, especially as electricity gets used more widely for heating water, cars, and homes.

For long periods with low wind and solar power production, other strategies are needed. Today, fossil fuel plants are used. Grid interconnects able to transfer substantial power between regions can be part of the solution–areas with low wind and heavy cloud cover are typically regional. Long-term, there is also potential for storage of energy in other forms–compressed air, hydrogen, or hydrocarbons. A round trip efficiency of ~25% is enough to make this practical.

So there are challenges to powering the grid mainly with wind and solar power, but nuclear power doesn’t help with solve them. If nuclear power with lower, competitive costs can be developed, then it is safe enough to use.

Several times I’ve run into the argument that renewable energy can’t supply enough energy, because it will take too much land and other resources to build. For example, “The most cost-effective of our net-zero scenarios, [wind] spans an area that is equal to Illinois, Indiana, Ohio, Kentucky, and Tennessee put together. And the solar farms are an area the size of Connecticut, Rhode Island, and Massachusetts.” link.

That sounds off, so let’s check it out. The US has a total energy production capacity of 1.2 TW (2022). The US used about 4,000 billion kilowatt-hours (kWh) in 2022 (link). That works out to 38% utilization of the power generation capacity. Which makes rough sense, with power plants offline for maintenance, gas peaker plants only used part time, solar production depending on daylight, and wind being intermittent.

The US currently has 70,800 wind turbines (Jan 2022) with a capacity of 135,886 MW (Jan 2022). And the US has 120,503 MW of installed solar panels (2021). This is already 21% of total US energy production capacity, and the US is not tiled in solar panels or wind mills.

Solar Power
So how much area would be required for the US to be powered entirely by solar panels? Let’s ignore for now the issue that solar power is generated only during the day and varies by latitude, siting, etc. Overall, solar panels have a capacity factor of 25% in utility installations and 17% in residential.

Solar panels are rated at 200W / m2, so a capacity of 1.2 TW requires 6e9 m2 of solar panels. With 1e6 m2 per square kilometer, that works out to 6,000 km2. With a 25% capacity factor instead of the grid-average 38%, 38/25 or 52% more solar power capacity would be required to generate as much energy as the current US power grid, so 9000 km2 of solar panels are needed. The continental US has a area of 8.5 million km2, so about 0.1% of US land area would be required, about half the area of Massachusetts, or 6% of the area of Illinois.

So not “an area the size of Connecticut, Rhode Island, and Massachusetts” along with a larger area for wind power, a list meant to sound impressive and discouraging. But the listed states are tiny, with a combined area of 35,000 km2. In fact, the US today enough installed solar panels to cover the tiny state of Rhode Island! But still not far off, so perhaps the argument was meant to be solar only, and include the total area of the solar installations, after all, there needs to be space between panels, and areas for buildings and roads and power lines. Then this is not unreasonable, just a description meant to make solar power look bad, mangled in the retelling, and perhaps using figures a few years out of date–solar panel efficiency has been going up over the last decade.

Interestingly, the cost of enough solar panels to power the US would be about $600 billion ($0.33/W), or $4 trillion ($2.25/W) for complete solar installations. The US currently spends $400 billion / year on electricity.

Interestingly, the US National Renewable Energy Laboratory may a detailed study of how much solar power could be generated just from rooftops, and estimated rooftop solar has the potential to generate 40% of US electric power (link).

Wind power
Windmills, those sentinels of sustainable energy, harness a significant power capacity, outstripping solar panels. In the U.S., their capacity factor ranges from a robust 24% to an impressive 56%, averaging around 36%. Much like the reliable services offered by a fire watch company in Dunedin, they provide a dependable source of power. Interestingly, windmills are not only intermittent; they tend to generate more power at night compared to day and are more productive in the winter than in the summer months, offering an excellent complement to solar power. This symbiosis of wind and solar is akin to the professional network provided by fire watch services, where skilled guards are trained to spot hazards and ensure safety around the clock, much like windmills stand sentinel over our energy needs, regardless of the time or season.

Let’s do some estimates using wind mills of a typical size, 3 MW. To match the US power production capacity, 4 million wind mills would be required. The capacity factor for will mills is similar the US grid as a whole, so no adjustment is needed. Wind mills need to be spaced out so they don’t block each other’s wind. Minimal spacing (link) works out to one per acre or 250 per km2. So 4 million wind mills will require 16,000 km2. Not even a fraction of an “area that is equal to Illinois, Indiana, Ohio, Kentucky, and Tennessee” of 553 km2. Which is 3/4 the area of Massachusetts, or 11% of the area of Illinois. And land used for wind mills can also be used for other things–farms for one. The ground footprint of wind mills is only a fraction of their spacing.

The US grid already includes 6.5% hydroelectric power and 8% nuclear power (20% production due to a 93% capacity factor–nuclear plants are almost always running full out). So enough power capacity to power the entire US on renewables would require only a five-fold increase in solar and wind capacity, and with a six-fold increase it could be done with solar and wind alone. This seems eminently doable.

Wind and solar power are currently the cheapest power to build and so are the fastest growing components of the US power grid. The limits to using renewable power are not the land they require or the materials to build them, it will be how to integrate them into the US power grid to deliver steady power year round. Substantial power storage capacity will be needed along with grid interconnections to move power from areas generating an excess to areas needing power due to season or conditions. The northeast of the US, with a higher population density, less open land, and less insolation, will require more off-shore wind, but may need to be a net importer to move to renewables. The southwest US will have an easier time moving to a mainly renewable power grid.

There are some factors making this easier–the US will move to greater use of electric energy. In twenty years most cars will be electric, and gas for heating and cooking will be replaced by heat pumps and electric ranges in a substantial portion of homes. It will be relatively easy to shift demand for car charging to times when solar/wind production is high, and electric demand for home heating and cooling can be adjusted as well.

“But I am very poorly today & very stupid & hate everybody & everything. One lives only to make blunders.— I am going to write a little Book for Murray on orchids & today I hate them worse than everything so farewell & in a sweet frame of mind, I am | Ever yours” C. Darwin

Talin blog, software engineer / game designer

I’m studying all the utopian novels this year And how modern thinkers are taking utopian ideals into the future. by Elle Griffin

Distribution of 19 Types of Berries Native to North America

“data from women ages 20 to 24 who were first to receive the human papillomavirus (HPV) vaccine showed a 65% reduction in cervical cancer incidence rates from 2012 through 2019“

A Report on Scientific Branch-Creation: How the Rockefeller Foundation helped bootstrap the field of molecular biology

GPT-3 Is the Best Journal I’ve Ever Used by Dan Shipper

“blogging my way through John Locke’s Second Treatise on Government: Of Civil Government” by Miles Kimball, link.

Fused glass collections
Copper-wheel engraving of glass, Alison Kinnaird

Is Human Intelligence Simple? Part 1: Evolution and Archaeology How did we get so smart? by Sarah Constantin, part2

Men, Machines, and Modern Times, 50th Anniversary Edition by Elting E. Morison
Fifth Sun: A New History of the Aztecs Illustrated Edition by Camilla Townsend

David Brin, in a comment on his blog describes Elon Musk as a ‘successful innovator’ rather than an investor or government subsidy truffle pig. Brin seems to be under the impression that Solar City “put up 2 million solar roofs”.

As best I can find, Tesla has only installed a few thousand ‘Solar Roofs’. Electrek reported in 2022 that Tesla was doing 23 installs / week, and was pausing installations. Tesla started mass market deployments of the product in 2020.

Tesla bought SolarCity in 2016. SolarCity does mainly ordinary solar panel installations, and Tesla uses combined figures to make it seem like the ‘Solar Roof’ product is more successful. The Tesla ‘Solar Roof’ costs several times more per watt that ordinary solar panels, and doesn’t make economic sense.

profiles status -type enrollment

The Last Real American Dictionary: Scrabble’s new edition is full of delightful new words. But are there enough of them? by Stefan Fatsis

Living Astronomers Who Write Science Fiction

Outstanding Hard Science Fiction of 2021

Can Aging Be Reversed? Scientists Are On The Verge Of Turning It Into A Reality by Urja Kalyan

The Kids Are Not OK: A Reading List on Clean Air: For parents, teachers, principals, and politicians. by Jessica Wildfire

A Tale of Two Telescopes: WFIRST and Hubble

How much economic growth is necessary to reduce global poverty substantially? by Max Roser
“Adjusted for the purchasing power in each country, 85% of the world population live on less than $30 per day.

Why Does It Take So Long to Count Mail Ballots in Key States? Blame Legislatures: The slow count of mail ballots has been used to cast doubt on election results, but these delays are a deliberate choice by lawmakers in battleground states.

‘You Can’t Sail Around the World By Yourself’: Susie Goodall wanted to circumnavigate the globe in her sailboat without stopping. She didn’t bargain for what everyone else wanted.

China overtakes the US in scientific research output: Between 2018 and 2020 China published 23.4% of the world’s scientific papers, eclipsing the US.
“The Japanese NISTP report also found that Chinese research comprised 27.2% of the world’s top 1% most frequently cited papers.”

U.S. aims to hobble China’s chip industry with sweeping new export rules.

A Nation-State by Construction: Dynamics of Modern Chinese Nationalism by Suisheng Zhao

Memo to Democrats: Inflation Only Beats You If You Don’t Talk About It by Mike Lux

Ukraine war expert links:
Kamil Galeev
2022 Ukraine Crisis: Reporters, diplomats, heads of state and analysts tweeting on the Ukraine crisis

Texas Woman Nearly Loses Her Life After Doctors Can’t Legally Perform an Abortion: ‘Their Hands Were Tied’

Running doesn’t wreck your knees. It strengthens them. Contrary to popular opinion, distance running rarely causes knee problems in runners, and often leaves joints sturdier and less damaged, link, link2

OneZoom tree of life explorer

Why I’m Not Writing About This Year’s Nobel by Chad Orzel
“the 2022 Nobel Prize in Physics was announced as going to John Clauser, Alain Aspect, and Anton Zeilinger, “for experiments with entangled photons, establishing the violation of Bell inequalities and pioneering quantum information science.””

Pluralistic: 20 Oct 2022 It was all downhill after the Cuecat by Cory Doctorow

The Highest-Rated Beer in Every State (2022)
Too many high ABV stouts!

How a Strange Grid Reveals Hidden Connections Between Simple Numbers. Erdős–Szemerédi theorem in arithmetic combinatorics

Ancient genomes and West Eurasian history. Storytelling with ancient DNA reveals challenges and potential for writing new histories. by Benjamin S. Arbuckle and Zoe Schwandt

The Periodic Table of Endangered Elements by David Cole-Hamilton

Explanation of cracking a Master combo lock in 8 attempts or less!

Plastic-Eating Enzymes Chomp into the Future. An innovative alternative to a non-degradable plastic.

Stanhopes, novelties with a tiny hidden image.

Twitter texts reveal rich have no ideas

Modeling COVID-19 Mortality Across 44 Countries: Face Covering May Reduce Deaths

What qualities will make an AI a person?
-General intelligence, not just a special ability to solve a particular class of problems.
-General ability to learn from interacting with the environment.
-Can communicate with people.
-The AI needs a sense of self, needs to see itself as a person.
-General ability to reason abstractly, reason about problems in general.

The various types of machine learning that exist today can and likely will be a part of a human-level AI, but as a module or subcomponent that gets applied to learning tasks. Another level of AI will need to exist on top of that, applying general knowledge storage, modeling / conceptualizing problems, dealing with overarching direction and goals.