Quote:
Originally Posted by NeilBlanchard
Reconstructed data using proxies is not the same as modeling projections. ...
|
Good points, but not the one I was after - it is about resolution.
We have measurements now which are almost in real time - at the lowest resolution it is day by day. A worked example is the "record" ice reduction in the Arctic last summer - we could watch that figure all the time via satelites.
Reconstructions don't do that. At best they give us year by year values, maybe even season by season but more likely measurements which are decades long. Ice cores are a clear example - some don't record any measurement under 300-500 years in length.
So there could have been a similar ice breakup maybe in 1937 but we don't know, there could have been a really bad melt in 1930 - we don't know. There could have been a really bad freeze in the 1920s - we don't know.
Another worked example - take that map from 1938 - "From Observations" - that means ships navigating around the ice and radioing others at sea, ships arriving at port and reporting what they saw, and maybe even aircraft being sent out to map the ice. This did not happen in one day, it was months long - so the map is an average.
Today we just go to a web page and can get a map of the ice right now, which is what the second map is.
So it is comparing apples and bananas.