Soiling season is here: a field note on when cleaning actually pays

It is June. Across Iberia and the Gulf the spring rain has stopped, and on the plants we watch the soiling ratio has started its slow seasonal decline. This is the time of year the cleaning question gets asked — and answered badly.

The bad answer is the calendar: “we clean in June and September.” It feels disciplined. It is mostly luck. A fixed date over-cleans when a rain front was two days out, and under-cleans through a long dry, dusty spell that is quietly costing more than the clean would. The calendar optimises for the crew’s convenience, not the asset’s yield.

The only question that matters

Cleaning pays when the value of the energy you would recover before the next cleaning rain exceeds the cost of the clean. That is it. The inputs are the site’s soiling rate, the tariff or PPA price on the recoverable kWh, and the rain forecast — and all three are site-specific and time-varying, which is precisely why a fixed date can’t be right two years running.

We wrote the full treatment of that calculation — including how the soiling ratio is estimated when there’s no dust sensor on site — here: Soiling loss: when is cleaning worth it? The mechanics of the loss itself sit on the soiling-loss reference page.

What to do this week

If you have a clean scheduled on the calendar for this month, do one thing before you dispatch the crew: check the soiling trajectory against the ten-day rain forecast. If rain is likely to do the job for free, defer. If you are in a dry run with a high soiling rate and a good tariff, you may already be past break-even — clean sooner, not on the date. Let the numbers, not the month, make the call.