At the end of the day, it's perfectly fine to have an opinion and feel passionately about something. Just don't do it as a scientist. Do it anonymously. The military has very similar concerns. If you're an expert on an enemy, and you think they're going to attack, shouldn't you warn folks? Well yes, you should warn your commanding officers. And you can feel free to participate in political causes around doing things. But never in uniform. And never as a member of the armed services.
I think the most difficult part of all of this is that government is not about doing the right thing, it's about consent of the governed. So if an asteroid were approaching Earth, everybody might vote to ignore it. People have a right to voluntarily choose to self-destruct if they wish. You don't get to overrule the consent of the governed because you have special knowledge. You can certainly advocate, but not as some representative of a more elite caste of people.
Politics -- and science too, for that matter -- is full of people who absolutely "know" what's right. Unfortunately, 1) they rarely unanimously agree with one another, 2) we don't have a priestly class to make wise decisions, we have elected representatives, and 3) if the only requirement for telling others what to do was an unshakable knowledge that you know better, we'd be out in the street beating each other with clubs. We got over that kind of government. At least I hope so.
You're also misleading about the consent of the governed. The governed don't have to agree with every policy that the government institutes. Elected representatives are often only bound to their conscience and not to the direct will of the electorate. This may be hard to swallow in some instances, but the in your example the people that choose to ignore the asteroid have no right to also initiate the destruction of all those that don't agree.
Btw: The arguments in the reason.com article you posted upstream about lies as part of lobbying campaigns have been refuted in the nytimes article quite well. Knowingly engaging in a dangerous activity and lying about the outcome does not deserve 1st amendment protection.
I will note that the issue the Reason article brings up -- using state AGs to prosecute people who publish research, even if it's fraudulent research, is very serious. It's also the exact flip-side of the complaint that congressman are subpoenaing things. It's not going to end well. What one side can do when they're in power, the other side can do when the tables are turned.
This is what happens when policy is made without principle, and down this path lies authoritarianism and suffering. Might-makes-right doesn't end well.
Now, you could use a different line of lobbying: We're aware that using non-renewable carbon based energy sources are an issue, but they're the best we have and the benefits outweigh the costs. This wouldn't be lying despite better knowledge and certainly be protected. That would also be about policy decisions (what do we want and which risks do we accept) which is a valid target for lobbyism.
If I am a scientist, an I have a choice of paths to explore but cannot explore them all at the same time, then there are many things which may guide my decision making, including expectations of recognition and better employment. That is personal activism, yes? And that's because several hundred years ago scientists decided that "first to publish" was the best way to couple scientific advancement with the desire for personal advancement.
If I submit a proposal to a funding source, or a rare resource like the HST, I need to convince them that the science I'm doing is worthwhile. That's activism, yes?
Even then, you've left out "passivism", which might not be a word but I think it gives the intent.
If Exxon scientists do research in the 1980s which concludes that fossil fuel use leads to global warming, and decide to not publish, is that activism? For another real-world example, if a scientist wants to study "Untreated Syphilis in the Negro Male", and doesn't tell the participants that penicillin could now treat the disease, and those men end up suffering and dying though the willful passivism of the scientist, who is more interested in the science of what happens than the morality, is that "activism"?
On the flip side, how do I conduct research to test if advertisement for flu shots is an effective way to reduce flu deaths without some sort of "activism"?
I also don't like your use of the term "politics". I prefer "ethics", where politics is an applied form of ethics.
That's why I don't like your majoritarian definition of what "government" means. Ethics can lead someone to break the law, and later even be praised for doing so. See Thoreau's "Civil Disobedience" and the many people who followed that principle. Or even further back in time, and to quote de Tocqueville's "Tyranny of the Majority":
> "If it be admitted that a man possessing absolute power may misuse that power by wronging his adversaries, why should not a majority be liable to the same reproach? Men do not change their characters by uniting with one another; nor does their patience in the presence of obstacles increase with their strength. For my own part, I cannot believe it; the power to do everything, which I should refuse to one of my equals, I will never grant to any number of them."
You correctly point out "we don't have a priestly class to make wise decisions" but you uncoupled that from the military. Even in the military, some orders and laws must not be followed, as Nuremburg Priniciple IV:
> "The fact that a person acted pursuant to order of his Government or of a superior does not relieve him from responsibility under international law, provided a moral choice was in fact possible to him".
An Earth-killing asteroid will lead to genocide as a side-effect. The moral choice, if possible, is to prevent genocide. Even if 51.9% of the U.S. votes to die. One's right to commit suicide does not justify genocide.
Or slightly more analogous to climate science: OK, so you've found the bomb. But there are 19 different ways of trying to defuse it and/or limit the impact of it exploding, including just evacuating the area and letting it blow. Bomb-science might tell us something about what either option is likely to result in, but it probably intersects heavily with a lot of other fields in doing so, and there will still be inherently non-scientific trade-offs to be made at the end of the day (do you send in three bomb-technicians with a 70% chance of defusing/30% chance of getting blown up to save a $10m building? Or do you send in a robot with a 60% chance of success? What if it's a $1m building? What if it's a power plant and 100,000 people and a hospital depend on it? etc etc).
Ok, so you've found the bomb. You've found it inside a school, disguised as a soda vending machine. The moment you publish your results, you find yourself under attack from school administrators and soda vendors, and they all use the same argument: "there's thousands of soda vending machines in schools, and not one of them has blown up. Your results must therefore be faulty". The attacks are not based on their own scientific rigour, but only serve to sway public opinion against any action. It seems the soda vendors are happy to let this one bomb explode in order to protect their revenue stream.