Reply to 'Influence of cosmic ray variability on the monsoon rainfall and temperature': a false-positive in the field of solar-terrestrial research

TL;DR

This paper critically examines a previous study claiming a significant link between cosmic rays and Indian monsoon extremes, revealing that the original statistical analysis was flawed and the purported correlation is likely a false-positive.

Contribution

The paper identifies and corrects a statistical error in a prior study, demonstrating that the supposed correlation is not statistically significant when proper distribution assumptions are applied.

Findings

The original study's significance level was overestimated due to incorrect distribution assumptions.

Corrected analysis shows the correlation is likely a false-positive with a high p-value.

The data follow an ergodic chaotic distribution, not Gaussian, affecting significance testing.

Abstract

A litany of research has been published claiming strong solar influences on the Earth's weather and climate. Much of this work includes documented errors and false-positives, yet is still frequently used to substantiate arguments of global warming denial. This manuscript reports on a recent study by Badruddin & Aslam (2014), hereafter BA14, which claimed a highly significant ($p=1.4\times10^{-5}$) relationship between extremes in the intensity of the Indian monsoon and the cosmic ray flux. They further speculated that the relationship they observed may apply across the entire tropical and sub-tropical belt, and be of global importance. However, their statistical analysis---and consequently their conclusions---were wrong. Specifically, their error resulted from an assumption that their data's underlying distribution was Gaussian. But, as demonstrated in this work, their data closely follow an ergodic chaotic distribution biased towards extreme values. From a probability density function, calculated using a Monte Carlo sampling approach, I estimate the true significance of the BA14 samples to be $p=0.91$.