Estimating Signal Amplitudes in Optimal Fingerprinting, Part 1: Theory

摘要

This paper describes a variant of the regression-based technique of climatechange detection and attribution that is generally known as optimal fingerprinting. The fingerprinting approach is to define a pattern of response to external climate forcing using a climate model and then to estimate the amplitude of that pattern, or signal, in the observed climate record. If the hypothesis of zero pattern-amplitude can be rejected with confidence, then that signal is said to be detected.