Death rates are measured based on deaths from accidents and air pollution per terawatt-hour (TWh) of electricity.
First post on the fediverse. Hopefully it auto loads the link photo but if not I'll put it as the first comment. Sorry for anything incorrect in handling this.
The oil and gas industry switched to the other side of that battle decades ago when they realised there was nowhere near enough high concentration U235 to make a difference and the alternative involved wind and solar thermal eating at least half their business (as wind and PV did, very nearly immediately, within 12 years of finally getting a tiny fraction of thebsubsidies nuclear recieved).
Coal power barons were on the other side from the second they started scamming tax payers to buy nuclear plants for them to have a monopoly on.
Even better they convinced a bunch of people that all other green ideas are stupid, due to greens being anti nuclear. Hence they get both, no nuclear and no renwables, evs, public transport and the other sensible ideas of the greens.
For nuclear it does, depening how many cancer deaths have been caused by mainly Chernobyl. That is however massivly disputed, as it thankfully only ever happend once and well the Soviets had no intresst in actually being honest about that. Depending on the measures that can be thousands of deaths, due to an accident.
The "accidents" (or rather intentional genocide) involving native americans, uzbeks, nigerians, and lower caste indians in the upstream of the fuel cycle dwarf chernobyl by at least an order of magnitude. Still not rivalling the coal or oil industry, but not far off.
Can people stop linking to this fractally garbage 15 year out of date source already? The sovacool study is garbage (anything from sovacool is so methodologically garbage it discredits his point even though he's often coincidentally right for unrelated reasons). Arbitrarily rejecting its numbers for one particular power source and switching to one with another methodology is worse. The renewable technology referenced is nothing like what is used now. It completely ignores the primary source of ongoing harm for everything except fossil fuels (and those too in the long run).
Do you have a better source for death rate or similar relative to the actual amount of energy generated? Happy to compare if you do. Not trying to push this as a be all data source and happy to replace it with a better one if available.
What you're after is an LCA of a specific instance of new technology (which attempts to measure the sources of harm in context) rather than a context-stripped summary of a complex subject drawing arbitrary lines designed to create a bad faith talking point. Anyone reducing it to a single number is making a bad faith propaganda point one way or the other. You won't find any credible version of it because it's not a credible exercise.
The uranium that feeds the canadian project (excluding the historical tens of thousands of native deaths from intentional waterway poisoning) is harmless compared to what is happening in Arlit or Adapa (the harms of which are just beginning and are actively covered up). All estimates (by anti-nuclear advocates or by nuclear shills) of chernobyl are poor. Arbitrarily excluding santa susanna or windscale or mayak is done without reason. The risk profile of a 200kW wind turbine is vastly different to a 3MW one or a 15MW offshore one. Utility solar is nothing like rooftop. Countries with mandatory working at height safety equipment have vastly different risk profiles than those without.
Acknowledge that the harms are low for the bottom four options if they are done properly, then actually enforce doing them properly rather than using it as ammo to justify the horiffic (and rising) pollution from uranium milling, mining and plutonium extraction whilst exaggerating having one person die in an entire country's utility solar program. And also treat mining for rare earths for either magnets or obsolete USA-based thin film solar or burnable neutron poison the same way holding both to the exact same standards per unit of energy.
Treat all waste from all options the same way (non-recycled waste must be only a few dozen kg per lifetime of energy and it must be permanently dealt with before profit is disbursed).
Treat danger from all options the same way. Disaster cleanup (be it oil spills or radiation or a dam burst) must be fully collaterised with no liability limit from assets that won't crash if something happens.