The null-distribution for the global mean temperature response derived through Monte Carlo simulations, where the annual mean GCR was replaced by random values and repeated 3000 times
Figure 3. The null-distribution for the global mean temperature response derived through Monte Carlo simulations, where the annual mean GCR was replaced by random values and repeated 3000 times. The solid red line shows the results for which only the GCR was replaced by random numbers whereas the dashed red line shows the results for GCR when all co-variates were set to random numbers. The global response was calculated from the global spatial temperature patterns.
Variations in the annual mean of the galactic cosmic ray flux (GCR) are compared with annual variations in the most common meteorological variables: temperature, mean sea-level barometric pressure, and precipitation statistics. A multiple regression analysis was used to explore the potential for a GCR response on timescales longer than a year and to identify 'fingerprint' patterns in time and space associated with GCR as well as greenhouse gas (GHG) concentrations and the El Niño–Southern Oscillation (ENSO). The response pattern associated with GCR consisted of a negative temperature anomaly that was limited to parts of eastern Europe, and a weak anomaly in the sea-level pressure (SLP), but coincided with higher pressure over the Norwegian Sea. It had a similarity to the North Atlantic Oscillation (NAO) in the northern hemisphere and a wave train in the southern hemisphere. A set of Monte Carlo simulations nevertheless indicated that the weak amplitude of the global mean temperature response associated with GCR could easily be due to chance (p-value = 0.6), and there has been no trend in the GCR. Hence, there is little empirical evidence that links GCR to the recent global warming.