Last week we investigated the performance of German wind farms which, after a massive surge in capacity over the past two decades, are now in a position to contribute substantially to the electricity mix (7.6 % in 2011 according to the national statistical office). However, as we have seen, this comes at a price. The contribution shows large and largely unpredictable variations which have a destabilising effect on the grid.
Another source of renewable energy which has gained a lot of support recently is solar PV. Like wind PV has soared dramatically in the past years. Nevertheless, its overall contribution to the energy mix is very low (3.1 % of total power generation in 2011). Moreover, like electricity produced by wind mills, PV is a factor of instability to the grid. The sun is not shining uniformly throughout the day. Passing clouds may severly impact the output of solar cells causing fluctuations in the supply chain.
Impressive growth rates in both capacity and production, are inclined to mislead the observer. A more thorough consideration of the situation of solar PV, however, will have to look at the output per MW installed. That is the quantity which allows us to assess the investment in that energy source.
Like in the case of wind farms, the expected productivity of solar PV depends heavily on the location. And thus, we should not be surprised to see huge differences between various installations. Here, however, we look at the global picture. This point of view is even more justified since Germany decided to abolish all nuclear plants by 2022 which, in turn, means a higher burden for all other sources of energy. As the country, simultaneously, is highly committed to reducing its carbon footprint renewables are bound to play a much larger role in the future.

Fig. 1 Solar PV in Germany. Average output per MW installed capacity and solar irradiation.
Fig. 1 shows the average PV output (P/C) in MWh/MW (blue curve) while the red curve refers to the average solar irradiation (SI) in kWh/sqm (r.h.s. scale). The performance per MW installed has been adjusted by using a specific model in order to account for extra capacity added over the course of a year. The two curves appear to follow a similar trend. Nevertheless, their correlation is quite weak. An increase in irradiation may even coincide with a downturn in specific production. Thus, on a global scale, more sunshine does not necessarily mean more solar energy produced. Over the period in question we get an average PV output of 868.8 MWh per MW installed with a spread ranging from less than 700 MWh/MW to 1157 MWh/MW.

Fig. 2 Solar PV in Germany. Deviation from mean output per MW installed (P/C) and solar irradiation (SI) in %.
Fig. 2 displays the deviations of both specific output P/C and solar irradiation SI in % over the period in question. This picture confirms the conclusions drawn from Fig. 1. The variations in PV performance are not necessarily reflected in the respective variations of solar irradiation. They may even go in opposite directions. Remarkably, the fluctuations in SI are much larger than the ones in P/C, at least on an annual scale.
As in the case of wind more capacity does not always mean more output. The availability of sunshine comes in as a crucial factor which may be decisive for the performance of a particular facility. And, like the wind, this quantity may vary a lot over the years as can be inferred from the green curve.
There is one more message hidden in Fig. 2. One may argue that due to the relatively small scale of solar PV in terms of the entire German power production, the variations in P/C do not fully reflect the varying nature of solar irradiation. In other words, one might expect them to become even larger as the number of PV installations increases. This in turn implies that the fluctuations in the power grid may be even disrupting than today.
In view of this it is of utmost importance to develop storage facilities for the solar energy produced. This should in fact be a priority.