Abstract

According to the IPCC, ".…Anthropogenic influence on climate occurs primarily through perturbations of the components of the Earth radiation budget." Unfortunately, there are no published peer-reviewed statistical analyses of these perturbations using real hourly data. It is, therefore, not surprising that some researchers have been unable to reject the null hypothesis of natural variability as a driver of rising temperatures, a result that must delight the climate deniers.
This paper represents an attempt to address this gap in the research using hourly CO2, surface radiation, and meteorological data from the polar regions. The following surface radiation variables are employed in the analysis: downward short-wave, downward long-wave, upward short-wave, and upward long-wave. The data set spans from Jan 1, 1998, through Dec 31, 2022. The analysis employs hourly data from two BSRN (https://bsrn.awi.de/project/background/ ) observatories in the Arctic and three from Antarctica.

The analysis employs a spatiotemporal approach that proceeds by employing the vector autoregressive regression method, in which all of the contemporaneous hourly values of each of the variables from across the five observatories are endogenous. The relationships among the variables are then assessed using the concept known as Granger Causality, which is based on whether the lagged values of some variables X, Y, and Z are useful in predicting the current value of some variables A, B, and C. One of the more basic findings is the rejection of the null hypothesis of no radiative transfer between the two regions. The analysis also indicates that hourly CO2 concentrations have Granger Causal consequences for both surface radiation and meteorology. Consistent with this finding, the accuracy of the CO2 contingent out-of-sample hourly predictions of the surface radiation components and the various meteorological outcomes provides more robust support for the scientific consensus on climate change.