Examining Dartmouth's claims about improving sea ice prediction models