Interpretation with many variables

This is an excellent example of the difficulty of interpreting historical data. Tree rings: Chainsaws at dawn. Tree ring widths are determined by several factors, including temperature and rainfall. We need other information if we are to deduce temperature from tree ring width. In practice this seems to be choosing only trees in areas, “You can extract climate information from trees – mostly pines – from more extreme environments, like northern Scandinavia and Siberia; trees from high altitudes and high latitudes where the summer warmth was very important.” There is no escape from complex modelling.