#3 The most interesting thing to come out of this latest iteration of the Hockey Stick is this, in my opinion. What Anthony Watts calls "Liebeg's Barrel" is something I was sort of aware of - my day job involves customer support for a system which generates fertilizer application prescription files for VRA equipment, based on soil sample analysis & other inputs - but hadn't thought about in the context of using tree ring records for temperature proxies.
"Liebeg's Barrel" is basically a visualization of how botanical yield is derivable from environmental & nutritional resources. The growth yield of any given plant is dependent, roughly speaking, on the most scarce resource required for that plant's growth. If the plant's not getting enough water but enough nutrients, than nothing you can do in terms of dumping more fertilizer on it is going to produce another inch's worth of growth, or grain of corn.
Similarly, if the tree's got enough water, enough nutrients, and enough degree days, then it's going to grow like gangbusters, even though it might be technically cooler than it was twenty years previously, when a drought during a warm period stunted the tree's growth rings.
In short, dendrochronological use in temperature proxies runs up against a massive problem with uncontrolled multiple variables, and I don't quite understand why they think it's at all useful. |