My understanding is that the notion of null versus
alternate hypotheses came from the statistical testing field where in general a null hypothesis is an assertion that some phenomenon will be constrained to a subset, often a subspace, of the total universe of possible observations.
Physical science is highly constrained by what is known about energy budgets, paleoclimate and radiative transfer in gases, and
alternate hypotheses are hard to
come by as the skeptics well know by now after a decade of trying.