Scale-invariant Analysis of Instrumental and Proxy Data Using Bernoulli Variables Demonstrates a Globally Coherent Little Ice Age
Abstract
The validity of using "climate epochs" such as the Little Ice Age (LIA) as a tool to frame global climate forcing over the Anthropocene has been the center of significant discussion in recent years. Many different estimates have been proposed for the magnitude of the global temperature anomaly during the LIA and some have questioned whether it is spatially homogeneous enough to represent a global climate signal at all. Here we use instrumental data from HadCRUT4 to investigate the relationship between spatial homogeneity and global climate signal and, by using proxy data from the PAGES 2k Network, apply this relationship to reconstruct estimates of global temperature anomalies during the LIA. We represent spatial homogeneity as a series of Bernoulli trials, where each spatial grid square is either above the local average or below it. This provides us a simple framework to evaluate probabilistically how global signals affect local records. Applying this framework to gridded instrumental data, we find a linear interannual correlation of 0.98 between the fraction of earth above local average temperature and overall global average temperature anomaly during the years 1850-2019. By randomly sampling the instrumental data to reduce the number of spatial grid squares used and adding noise, we can simulate the same data availability and signal-to-noise ratio that we observe in the proxy dataset. We show that when these degraded proxy-like instrumental records are smoothed, the Bernoulli representation of their spatial homogeneity retains a high interannual correlation (between 0.8 and 0.9) with the global average temperature anomaly. We then generalize these results to proxy data, developing a new scale-invariant temperature reconstruction technique. Through Boolean analysis of the PAGES 2k land and ocean proxy data, we find evidence for a globally consistent Little Ice Age, suggesting a significant global role for preindustrial climate forcing at multidecadal timescales.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2020
- Bibcode:
- 2020AGUFMPP0310013F
- Keywords:
-
- 3305 Climate change and variability;
- ATMOSPHERIC PROCESSES;
- 3337 Global climate models;
- ATMOSPHERIC PROCESSES;
- 3344 Paleoclimatology;
- ATMOSPHERIC PROCESSES