Soiling losses: from modelling to PV systems simulation

Global estimates suggest that soiling losses reduce annual PV generation by 3-5% on average, and in extreme climates, they can exceed 30%. These losses have enormous economic consequences – it’s estimated that soiling losses cost the solar industry €4-7 billion (US$4.6-8 billion) [1]. At a time when the PV industry is under pressure to deliver truly accurate and bankable outcomes, accurate modelling of soiling losses has become critical.

The nature of soiling losses

Soiling reduces the incoming solar radiation that reaches PV cells. This effect is quantified by soiling ratio (SR) – the ratio between actual irradiance on the soiled solar module surface and the reference case of a clean module. To put it simply, actual versus clean output. The inverse, the soiling loss (SL), directly translates into reduced energy yield.

The magnitude of losses due to soiling depends not only on the concentration of soiling particles, but also on the tilt angle of a module, surface characteristics, rainfall frequency and intensity, and cleaning interventions, whether natural or technical.

What makes soiling particularly challenging is its variability across space and time. Soiling-related losses can differ significantly by region, from just a few percent per year in temperate climate, to becoming the dominant source of performance loss in arid, dusty or polluted environments.

These effects also fluctuate seasonally, influenced by changing rates of particle accumulation and by cleaning events driven by local weather and atmospheric conditions. This variability necessitates accurate modelling that extends beyond empirical loss assumptions.

Modelling soiling losses in PV simulation

Soiling ratio, SR, compares the energy production of a PV module under polluted conditions (Epv-soiled) to its production under clean conditions (Epv-clean). The formula for SR is therefore defined as SR=Epv soiled/Epv-clean. The soiling ratio (SR) value ranges from 0 (complete obstruction by soiling) to 1 (perfectly clean PV module).

The relationship between SR and transmission loss (TL) is often assumed to be linear: TL=1-SR, which is customarily expressed in percentage as the soiling loss: SL=100·TL. This accounts for the reduction in solar energy received by PV cells after passing through the module’s glass. Given that uncertainties in model inputs can exceed the effects of nonlinearity between SR and TL, this assumption is practical for assessing soiling-related losses.

An accurate account of potential soiling losses is one of the prerequisites for accurate, not overly optimistic PV simulation; otherwise, it can lead to financial losses.

Simulating PV performance via Solargis Evaluate platform now allows for automatic simulation of soiling losses using the in-house developed Solargis soiling loss model.

The Solargis soiling loss model (Figure 1) is based on research by Coello and Boyle [2] and consists of two main components:

Accumulation of particles: The model calculates the mass (m) of suspended particles deposited on PV] surfaces, considering both natural and scheduled cleaning events. The formula used is: m=∫(v2.5C2.5+v10-2.5C10-2.5)cosβdt

Soiling ratio calculation: Once the accumulated mass (m) is determined, an empirical relationship estimates the soiling ratio: SR=1-0.3437erf(0.17m0.8473)

The Solargis methodology integrates multiple datasets describing particulate matter (PM2.5 and PM10), meteorological parameters and site-specific PV system characteristics. It integrates data from two leading sources:

• Meteorological parameters, including temperature, atmospheric pressure, wind speed, and rainfall, are sourced from ERA5 (31km resolution) and ERA5-Land (9km resolution)
• Particulate matter parameters (PM2.5 and PM10) are derived through harmonised data from multiple sources, based on CAMS reanalysis (80km resolution, 3-hour intervals). This coarser resolution may reduce local accuracy

The model accounts for deposition rates of airborne particles, the cleaning efficiency of rainfall events, wind speed, humidity and temperature, while also reflecting the influence of PV module tilt, mounting configuration, and scheduled cleaning events.

By modelling the particle deposition mass over time, Solargis estimates how quickly PV modules accumulate soiling and calculates the corresponding soiling ratio. Cleaning events, whether induced by rainfall or by technical intervention, reset the accumulation curve and are explicitly factored into the simulation.

This dynamic representation captures both the accumulation and mitigation processes that determine soiling behaviour in real operating conditions.

Validation and global application

Scientific credibility in soiling modelling relies on validation. Solargis validated its soiling model against high-quality ground measurements from 39 dedicated soiling stations in various geographies worldwide, where soiling accumulation is observed with controlled cleaning intervals and calibrated instruments. These validations confirm that the model reliably reproduces observed soiling dynamics across diverse climates.

In addition to local validation, Solargis has developed global soiling maps, illustrating the spatial distribution and seasonal variability of soiling losses. These maps provide developers with a first-order estimate of soiling risk during site selection and feasibility studies, while detailed simulations can refine the analysis for specific projects (Figure 5).

We’ve also recently launched Solarmaps—an interactive platform where you can explore monthly solar maps and assess the impact of weather and environmental conditions, including soiling, on your PV performance.

From uncertainty to a more predictable PV performance

Soiling is not merely an environmental phenomenon; it’s also a source of significant financial uncertainty. In many PV simulations, soiling contributes 5–10% of the overall uncertainty budget, and in extreme environments, this figure can be higher. When propagated through a project’s energy yield assessment, these uncertainties directly affect performance guarantees, OPEX planning and, ultimately, financial risk.

With the integration of soiling modelling into Solargis Evaluate, these uncertainties can now be systematically reduced. Evaluate simulates long-term time series at sub-hourly resolution and incorporates soiling losses directly into the energy yield chain.

Instead of guessing a fixed annual loss percentage, Evaluate users can model site-specific soiling profiles, compare scenarios with different cleaning schedules and assess their impact on PV power production output (PVOUT). This provides a far more realistic representation of project performance, helping operators optimise maintenance strategies and avoid costly surprises.

Towards a more resilient PV industry

The PV sector can no longer afford to treat soiling as a marginal or secondary issue. Its impact is measurable, material and, if not properly addressed, capable of undermining project economics. By embedding soiling into physics-based models and linking it directly to PV performance simulations, Solargis provides developers and investors with the tools they need to quantify risk, optimise maintenance and improve long-term project resilience.

The message is clear: in order to design PV systems that perform reliably under real-world conditions, PV module soiling must be modelled with the same scientific rigour as solar radiation, temperature or shading.

References

[1] IEA PVPS Task 13, 2022, ‘Soiling losses – impact on the performance of photovoltaic power plants’

[2] Coello, M. & Boyle, L., 2019, “Simple model for predicting time series soiling of photovoltaic panels”, IEEE Journal of Photovoltaics, Volume 9, Issue 5

Authors

Vicente Lara-Fanego holds a degree in physical sciences, an M.Sc. in geophysics and meteorology from the University of Granada, and a PhD in physics from the University of Jaen, Spain. With over 18 years of experience in the solar industry, he joined Solargis in 2017, focusing on meteorological and solar modelling, particularly on albedo and soiling. He has authored over 20 scientific publications.

Tomas Cebecauer holds a PhD in cartography and geoinformatics from the Institute of Geography, Slovak Academy of Sciences. After contributing to the development of the PVGIS portal at the European Commission’s Joint Research Centre in Italy, he became CTO and co-founder of Solargis. Since 2010, he has led the development of Solargis’ global solar resource database and simulation solutions for photovoltaic systems.