since chernobyl there have been a few very major civilian reactor meltdowns, and zero deaths caused by them
We are capable to focus on more than one thing at a time. Of course it doesn't make sense to instantly shut down all of nuclear power stations, just like it doesn't make sense to retire every fossil fuel powered engine at the same time or shut down coal from one day to the next. It's a long and slow process.
"The incidence of thyroid cancer in children, ordinarily very rare, had begun to increase markedly following the accident. Scientists have estimated up to 5,000 cases of thyroid cancer among residents who were children at the time of the accident"
"Plus an statistical excess of 9,000 cancer cases (leukemia + all solid tumors combined excluding thyroid) in the areas affected, and 5,000 people killed in this category".
"The study demonstrated increases in breast cancer incidence in all areas following the Chernobyl accident"
"the morbidity rate in Belarus for cervix cancer is as high as 36%, compared with 21% in Germany"
(Several sources, ten seconds search in ddg)
NRC denial of petition on LNT: https://www.regulations.gov/document/NRC-2015-0057-0671
(q) No adverse health effects among Fukushima residents have been documented that are directly attributable to radiation exposure from the FDNPS accident. The Committee’s revised estimates of dose are such that future radiation-associated health effects are unlikely to be discernible. The Committee believes that, on the balance of available evidence, the large increase, relative to that expected, in the number of thyroid cancers detected among exposed children is not the result of radiation exposure. Rather, they are the result of ultrasensitive screening procedures that have revealed the prevalence of thyroid abnormalities in the population not previously recognized. An increase in the incidence of cancers is unlikely to be discernible in workers for leukaemia, total solid cancers or thyroid cancer. The Committee has insufficient information to reach an informed judgement on the risk of cataracts;
i would also say that unscear is a more credible organization than the nrc when it comes to measuring and predicting effects of radiation, and that their conclusion ought to be the canonical one, as should be their position on lnt
> Also, the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) has recently published a review of cancer risk due to low dose rate radiation from environmental sources. (116) UNSCEAR concluded that “the results of the studies of cancer risk due to radiation exposure at low dose rates from environmental radiation do not provide strong evidence for materially lower risks per unit exposure than in studies of high radiation doses and dose rates.” (117)
So, no, UNSCEAR has not rejected LNT. UNSCEAR is in fact asserting that the evidence does not support rejection of LNT.
unscear as an organization have rejected lnt. it was resolved in assembly about a decade ago that it's just plain unreliable for extrapolating to low and even medium doses. if you look in reports you'll find that conclusions are never based on lnt, and when sources using lnt are incorporated they're always reevaluated against other models and often specifically discredited as overestimating low dose health effects, like this little bit from the 2019 report:
472. Due to the limited statistical power, it is generally difficult to detect a departure from linearity in the dose–response relationship for circulatory diseases. As an alternative to the standard approach based on a single preferred model, Schöllnberger et al. [S2] applied a multi-model inference approach to the dose–response analysis for mortality from cardiovascular disease in the LSS cohort. A combined set of plausible models fitted equally well. With this combined set of models, they found a dose response that was about one third weaker than that with the LNT model at doses below 0.6 Gy. The dose response at higher doses was stronger. More recently, the multi-model inference approach was applied to mortality in the LSS cohort from cerebrovascular and heart disease [S3]. For cerebrovascular disease, the dose–response curve estimated by multi-model inference was below the LNT model at low to medium doses (0–1.4 Gy) while at higher doses, a higher risk was estimated compared to the LNT fit. Similarly, a sublinear dose response was found for heart disease at doses of 0 to 3 Gy. The estimated confidence bounds indicated no conclusive answer for an increased risk below 0.75 Gy for cerebrovascular disease and 2.6 Gy for heart disease.
UNSCEAR did not say LNT was wrong at low doses. They said LNT was not confirmed (vs. other models) at low doses. This should not be surprising at all, since at low doses the signal becomes very weak.
There's a difference between "we don't know if LNT is correct" and "we know LNT is incorrect". You are improperly concluding they are saying the latter. They are not.
But yeah, people did return in the end.