The main conclusions of TOSCA, which are developed in greater detail in the forthcoming handbook, are the following:

Direct heating by solar radiation: an important effect, but not the only one

Solar electromagnetic radiation represents over 99.96\% of the total energy that enters the Earth's atmosphere, and is frequently described as a single quantity, called TSI (Total Solar Irradiance). The variability of the TSI over a Schwabe cycle (with a period of typically 11 years) is of the order of 1 W/m2, resulting in a top-of-the-atmosphere radiative forcing of about 0.2 W/m2. This variation is causing a global mean surface temperature response of likely slightly less than 0.1 ºC. This is of the same order of magnitude as the natural variability of the climate system on decadal time scales, thus making it very difficult to detect a solar signal in most climate observations.

Since the Maunder minimum that extended from approximately 1645 to 1715, solar activity (as expressed by the sunspot number) has progressively reached a peak of activity during the second half of the 20th century, called grand maximum, followed since by much more moderate levels. This pattern is well accepted. However, how this translates into TSI changes remains a matter of intense debate, with differences ranging from 0.6--3 W/m2 since the Maunder minimum. Narrowing this range depends on our ability to understand how the Sun radiates during periods of extreme quietness, such as the Maunder minimum. Comparisons with Sun-like stars provide further constraints. On yearly time-scales and below, however, the agreement between the observed TSI and model reconstructions is within error bars.

The impact of ultraviolet radiation versus that of visible and infrared

Although the energy from UV wavelengths contributes a relatively small amount to the TSI (about 8\%), it is much more variable than the TSI over the Schwabe cycle, and therefore cannot be neglected. In contrast to visible and infrared radiation, most of the UV radiation is absorbed in the middle and upper atmosphere. The main mechanism by which solar UV impacts climate is through ozone production and destruction, which alters the temperature of the middle atmosphere. Major progress has been achieved in the last decade, in incorporating these effects in climate models. While TSI variations impact global mean temperatures, UV variations are more likely to cause regional changes.

How changes in middle atmosphere can impact the much denser lower atmosphere is still a question of debate. Different mechanisms are likely to coexist, but the paucity of direct observations hinders further progress. A major challenge is the proper reconstruction of the solar spectral variability in the UV.

Energetic particles modulated by the solar wind: more important than expected

Variations in the solar wind permanently affect the state of the Earth's environment. One of their consequences is the intermittent precipitation of energetic particles into the atmosphere. There are various mechanisms by which these fluxes of particles can eventually affect the lower atmosphere, and climate, but their level of understanding is not as advanced as for radiative forcing. The primary mechanism involves the precipitation into polar regions of energetic particles (mainly protons and electrons) that produce chemically active species such as HOx and NOx, which in turn affect the ozone balance in the middle atmosphere (at high latitudes, and in winter), similarly to what UV radiation does, but limited to high latitudes.

These mechanisms are still poorly documented due to the lack of observations, and also because the observed quantities are subject to strong internal variability that masks the solar signal. Recent results show that the effects of energetic particle impacts on middle atmosphere ozone are more frequent than previously assumed. In addition, their impact on regional climate may be comparable to that of the UV forcing. These particle precipitations are local and highly intermittent, and one of the challenges is to quantify their long-term climate impact.

Galactic cosmic rays: not a major player in climate change

Galactic cosmic rays (GCRs) refer to the flux of high-energy, extra-solar, particles which constantly impinge upon the Earth's atmosphere from all directions. By colliding with atoms in the atmosphere, they generate cascades of secondary particles that may extend to the Earth's surface, and consequently produce ionization in the middle and lower atmosphere. GCRs are primarily modulated by the variable solar magnetic field and the geomagnetic field, and, as a result, vary inversely with solar activity. A long-debated hypothesis suggested that GCRs may provide a strong solar amplification mechanism, linking solar activity to the Earth's climate system. One suggested pathway by which this may occur, termed ion-mediated nucleation, operates via an ion-enhancement of aerosol formation and growth rates, leading to a modification of cloud properties. While laboratory-based studies have confirmed the existence of ion-mediated nucleation, it is now understood that conditions in the troposphere are usually unfavorable for this mechanism to exert a significant influence. Such conclusions are supported by laboratory experiments, climate model studies, and satellite observations. Although no evidence for a significant and widespread GCR--cloud link has been found, uncertainty still remains as to whether or not signals too small to be reliably detected remain buried within cloud observations. However, it is clear that the remaining uncertainty is not sufficient to account for 20th century climate change.

Increased certainty regarding a GCR--cloud link would be gained from improved aerosol and cloud datasets, in particular more reliable vertical profiles and cloud-type discrimination from satellite observations. Although there is no widespread globally significant GCR--cloud effect of importance to recent climate change, the possibility of locally significant, second-order, impacts on clouds has not yet been excluded: the uncertainty regarding such phenomena are high, to the extent that even the potential sign of cloud changes that may result from variations in the GCR flux is not reliably known. Various processes could account for local-scale relationships, the majority of which relate to the global electric circuit, which has recently received considerable interest. To reduce this uncertainty, a large coordinated effort of ground-based observations and experimentation will be required at multiple sites across the globe, with expertise from a range of disciplines including atmospheric electricity and cloud microphysics.

Internal climate variability blurs the whole picture

Random variability of the climate system is comparable in magnitude to that of the external drivers, which makes it very difficult to detect statistically meaningful correlations without having long-term climate records. In the stratosphere and above, the solar signature becomes the main cause of the observed variability, and so can often be more easily detected.

Long-term monitoring of the climate system is required to discriminate among concurrent mechanisms. Consequently, the key to a better understanding of the climate response resides in our ability to continuously monitor the state of the atmosphere, and the relevant solar forcings. These data are mainly obtained from spacecraft observations, and to a lesser extent from ground-based, or balloon-borne instruments. Unfortunately, the measurement of several key observables, such as the ozone content and the spectral irradiance, will be discontinued soon, thereby depriving us of the possibility to improve our understanding.

Global solar influence with significant regional manifestation

In recent years, considerable evidence has been accumulated in favour of regional effects of solar influences on climate. Recent observations suggest that regional temperature anomalies from solar associated changes may exceed global mean anomalies from global forcings such as TSI changes. For example, changes in the UV flux, and particles precipitation both affect the ozone concentration in a way that eventually leads to an enhanced polar vortex, and a positive phase of the North Atlantic Oscillation, which in turn has a particular impact on European climate.

This role of regional effects is a topic of great interest, and again highlights the limitations of a reductionist position that only emphasises global averages. The possible coupling of natural oscillation modes with the solar cycle has recently called attention to the impact of regional effects. For example, the positive phase of the North Atlantic Oscillation seems to be preferred in the descending phase of the solar cycle.

Models are a precious ally to observations

Although observations are vital for understanding the climate impact of solar variability, models have become a key tool for testing hypotheses, and for investigating regions or regimes that are not properly covered by observations. Significant progress in understanding has been achieved using reanalysis data sets that enhance observations by blending them with model simulations and thereby obtaining realistic and complete high-resolution data. However, each model has its limitations and caveats, and understanding the uncertainties of models has become as important as determining the uncertainties associated with observations.

Future developments will be substantially influenced by the expected increase in computer power, which will eventually allow climate simulations with higher spatial resolution. This may allow to gradually overcome the necessity of parameterising small-scale dynamical disturbances, clouds, etc. Regarding the investigation of solar signatures, an important issue is the proper description of the impact of solar UV radiation and energetic particles on the middle and upper atmosphere. This is a challenging task, as it requires a description of physical and chemical processes involving many species. So far, much more attention has been given to the troposphere only, but climate models are gradually pushing their upper limit to higher altitudes, thereby enabling to study phenomena involving vertical coupling processes such as solar signals.