>"I think if you compare historical rates of iceberg calving to current rates of iceberg calving, you should be able to find some significant quantifiable diffe

>"I think if you compare historical rates of iceberg calving to current rates of iceberg calving, you should be able to find some significant quantifiable difference." So? The rates could increase/decrease for many reasons. If all your X mechanism predicts is that the rate should change (or only slightly better, increase/decrease) it isn't going to be distinguishable from any other reason. Edit: It really sounds like you are making the usual NHST error of thinking a significant difference means your theory is true.