Mitigation
Manual cleaning
- ID
mit:manual-cleaning- IRI
https://solarlattice.org/kb/mit/manual-cleaning- Status
- active
Manual cleaning means crews wash module glass by hand to remove soiling, usually with water. It is the targeted fix for Widespread Fungal Soiling Losses. Black fungal biofilms stick to the glass and stay there through rain until someone cleans them off. On sites with persistent biological deposits, treat cleaning as required maintenance and put bio-soiling in the cleaning schedule. A simple, economical method is a bucket of water with a strip cleaner and squeegee, using overlapping vertical strokes. Do not use high-pressure water, brushes, solvents, abrasives or harsh detergents. Make sure the method fits the module maker's warranty. Use high-purity water, and if detergents are used they must be biodegradable. Check that cleaning worked by comparing cleaning records with normalized production. A repeatable step up in output after cleaning, where rain had not restored it, confirms the effect. Limits: hand cleaning is costly and slow, so whether it pays depends on local soiling rate, energy value and labour cost. Long gaps between cleans let lichen take hold, and lichen is very hard to remove.
<!-- review notes
- sentence 5: support check: partial - Warranty fit and biodegradable detergents are stated, and ultra-pure water is required, but 'if detergents are used' is a minor paraphrase; also detergents must comply with local regulations, which is omitted, so only minor detail is missing.
-->
Claims
typical_power_loss
c1 verified · fact
5–11 PERCENT
Conditions: Persistent production loss from fungal bio-soiling in humid subtropical regions, which may not recover after rainfall
This is the loss that cleaning addresses, not a figure for how well cleaning works.
verified by: joshuahamsaverified on: 2026-10-20review note: support check: partial - Passages give a 5-11% persistent loss from fungal bio-soiling in humid subtropical regions that may not recover after rain, but not that it is a 'typical' loss; also it is described as an impact factor, not a manual-cleaning figure.; reviewer: the compiler-flagged conflict is a different scenario or technology, not disagreement
c2 verified · fact
nominal 1.91 PERCENT
Conditions: Annual soiling loss without cleaning, simulated for a system in Mesa, Arizona (Naeem 2014)
In the same simulation, one cleaning a year cut the loss to 1.52 %/yr, two to 1.32 %/yr and three to 1.20 %/yr.
verified by: joshuahamsaverified on: 2026-10-20review note: support check: partial - The 1.91% annual soiling loss in Mesa and the 1.52/1.32/1.20 %/yr reductions are stated, but the passage does not say 1.91% is the loss without cleaning explicitly as 'manual cleaning typical power loss'; it is the baseline with rain only and cleaning is not stated to be manual.; reviewer: the compiler-flagged conflict is a different scenario or technology, not disagreement
Referenced by
Data