ALL data is flawed. It will never be not flawed. It is IMPOSSIBLE to achieve bias free data (Unless your area of study happens to be counting beans). Take for instance, the size of the Earth. For a long time, we simply didn't have the correct tools or methods to create an accurate judgement about the exact size and shape of the Earth. Over time, that has changed, and now that we can step outside the Earth entirely with satellites, we are only now creating a correct Geoidal map of the Earth, and creating new updates to the global datums. Thing is, the first real surveys were pretty close. Even now, the Clarke ellipsoid from 1866, long used by the USGS and other organisations wasn't too far off. It was wrong, it was skewed, but it was the best available, and it did a pretty good job. The NASA temp records are similar. They had data. Flaws were discovered, the data was updated. Similar conclusions can still be drawn. Science rolls on.
Some doubt can't be used to throw out your entire dataset, particularly when it is irreplaceable.
Thats great. Do you even know what data is flawed? Do you know what is based on the flawed data?
Yep, endorsed by just about everyone. It doesn't make it true. But it does make a little thing called Consensus