I am not a fan of "believe" or dogma of any kind myself. I tend to go on likelihood. I ask myself is it more likely that the 97% are correct or the 3%?
I think Bayes theorem might be worthwhile here. Besides, going with the majority or authority as a principal is almost the definition of dogma. That's not likelihood in any rigorous sense.
When weighing up the likelihood of something it is valid to consider the opinions of experts. Dogma is believing what an authority says is incontrovertibly true. I am not saying that man made climate change is incontrovertibly true, only incontrovertibly highly likely. It is true that the view of the great majority of the experts in a field has authority, particularly in regards to science, but it is an justifiable authority, not one for example one built solely on tradition.