It appears as if the authors put in 10 times as much freshwater (0.4 Sv) as observed for the freshwaterflux from Greenland (0.04 Sv) (see below). If 10 times more freshwater is needed to bring the model in agreement with the observation-based AMOC slowdown estimate, maybe this estimate is wrong?
Comments
"Since 2002, Greenland lost 5,900 billion tonnes (gigatonnes) of ice. To put that into perspective, imagine if the whole state of New South Wales was covered in ice 8 metres thick."