Internal Variability of All‐Sky and Clear‐Sky Surface Solar Radiation on Decadal Timescales

Abstract

Internal variability has been identified together with model response and emission scenario uncertainty as one of the sources of uncertainty in climate projections ( e.g., Deser et al., 2010;Hawkins & Sutton, 2009 ). In fact, on time scales of typically a few decades into the future, internal variability is a major contributor to the overall uncertainty in climate projections. The advances in numerical modeling in recent decades ( Eyring et al., 2016;IPCC, 2013IPCC, , 2021 work toward a better estimation of models' responses to radiative forcing, but the influence of internal variability remains important, especially on regional scales. Due to its dominant relevance on shorter time scales, internal variability has the potential to diminish or even reverse the long-term effect of anthropogenic forcing over a limited period ( Hawkins & Sutton, 2009 ). Being stochastic in nature, internal variability restricts us from making a direct comparison between historical model simulations and observations, and concerning future projections, it interferes with the time of emergence of the forced signal. Since the effect of these natural fluctuations in the climate system cannot be neglected for decadal time scales, statistical approaches of quantifying and understanding it need to be considered.The relevance of internal variability for a climate variable may be gathered from the ratio between the variable's response to a forced signal and the fluctuations that it exhibits due to the chaotic nature of the climate system, also known as the signal-to-noise ratio. A climate variable that is suspected to have a low signal-to-noise ratio, but still maintains a pronounced response to anthropogenic forcing on the decadal timescale, is the downward surface solar radiation ( SSR ) (

Internal Variability of the Climate System Mirrored in Decadal‐Scale Trends of Surface Solar Radiation

“…The magnitudes of clear‐sky SSR trends are always less than of all‐sky, which is in line with Chtirkova et al. (2022) who show that clear‐sky variability in CMIP6 models is generally less than all‐sky SSR variability (approximately 6 times less on a global mean basis). Further, there are many regions with opposite sign trends in all‐sky and clear‐sky SSR.…”

“…Looking at clear-sky patterns alone, one might associate warmer SSTs with more water vapor, thus less clear-sky SSR mean and variability and vice versa; the all-sky patterns during the same periods however have a more complex structure, showing areas with both positive and negative trends. The magnitudes of clear-sky SSR trends are always less than of all-sky, which is in line with Chtirkova et al (2022) who show that clear-sky variability in CMIP6 models is generally less than all-sky SSR variability (approximately 6 times less on a global mean basis). Further, there are many regions with opposite sign trends in all-sky and clear-sky SSR.…”

“…The data suggest that the estimates for unforced clear-sky variability as made in Chtirkova et al (2022) based on a pre-industrial climate may not be directly applicable to the present-day state, but rather underestimate the latter. This is in line with the discussion section and Table 1 in Chtirkova et al (2022), where clear-sky σ SSR in observations is found to be almost twice as large as σ SSR in CMIP6 piControl. From a physical perspective, it is plausible that clear-sky SSR variability differs between pre-industrial and present-day conditions.…”

“…Results for clear-sky SSR in Section 3.2 show that the effect of the absolute value of aerosol concentrations on unforced clear-sky trends can be substantial. The data suggest that the estimates for unforced clear-sky variability as made in Chtirkova et al (2022) based on a pre-industrial climate may not be directly applicable to the present-day state, but rather underestimate the latter. This is in line with the discussion section and Table 1 in Chtirkova et al (2022), where clear-sky σ SSR in observations is found to be almost twice as large as σ SSR in CMIP6 piControl.…”

Causes for decadal trends in Surface Solar Radiation in the Alpine region

“…Even though Krüger and Graßl (2002) have identified a pronounced decrease in cloud albedo in Europe during the period of decreasing aerosols in the 80s and 90s, Ruckstuhl et al (2010) did not find evidence of a significant indirect aerosol effect on SSR changes at 15 lowland stations (altitude lower than 1000 masl) in Switzerland during the same period. Folini et al (2017) and Chtirkova et al (2022) highlighted that the effect of internal variability at individual locations should not be neglected. All of these studies provide evidence for the existence of different players controlling the decadal SSR trends, from both natural and anthropogenic origins.…”