PVsyst Soiling Loss Modeling for India and Africa — Regional Calibration

Soiling loss is the single most variable and site-sensitive loss category in a PVsyst simulation. For a coastal US project in a humid climate, soiling might contribute 0.5–1.5% to annual energy loss with moderate rainfall washing modules naturally. For a utility-scale project in the Thar Desert of Rajasthan or the Sahara fringe of Morocco, soiling contributes 4–10% — the difference between a 79% Performance Ratio and a 73% PR, representing millions of rupees or dollars in annual revenue and significant impact on debt service coverage.

Getting soiling right in PVsyst is not a software configuration question — it is an engineering judgment question. The tool provides monthly input fields; the engineer must populate them with values that reflect the actual site’s dustfall rate, cleaning frequency, precipitation pattern, and ambient particulate concentration. Choosing values too optimistic triggers IE review comments and inflates the bankable P50 yield. Choosing values too conservative produces a yield report that undersells the project’s financial viability.

Direct answer. PVsyst soiling loss is configured as a monthly profile (12 independent values) under Project → Losses → Soiling. For bankable IEA documentation, each monthly value should be linked to site-specific dustfall data, published regional soiling benchmarks, or measured soiling curves — and must be consistent with the cleaning frequency specified in the O&M contract. Annual average soiling for Indian utility-scale ground-mount ranges from 1.5% (well-watered coastal Tamil Nadu) to 8% (pre-monsoon peak period, Rajasthan desert). Sub-Saharan Africa ranges from 2–9% depending on proximity to Saharan dust transport corridors.

---

Why Soiling Modeling Is an EPC P&L Decision

The soiling assumption in a PVsyst simulation is not just a technical parameter — it directly determines:

1. Bankable P50 yield — a 2% difference in soiling assumption changes P50 yield by approximately 2%, which for a 50 MW project at ₹3.50/kWh PPA represents ₹1.4–1.8 crore per year in revenue difference
2. O&M contract structure — the cleaning frequency required to achieve the simulated soiling level must be reflected in the O&M contract; a mismatch triggers IE red flags
3. Cleaning cost optimization — the economic optimal cleaning frequency is derived from the marginal revenue per cleaning cycle minus the marginal cleaning cost
4. P90 uncertainty — soiling inter-annual variability is a significant contributor to the P90 uncertainty band; poorly documented soiling expands the P50-to-P90 gap

---

How PVsyst Soiling Input Works

PVsyst accepts soiling loss as a monthly input table: twelve values, each representing the average soiling loss for that month as a percentage of incident irradiance. The monthly profile allows the simulation to capture the seasonal pattern of soiling accumulation and cleaning.

Accessing the Soiling Input:

Project → Losses → Module Array Losses → Soiling → Monthly soiling loss input. For each month, enter the expected average soiling loss percentage. PVsyst interpolates between monthly values when computing hourly simulation results.

Relationship Between Cleaning Frequency and Monthly Soiling:

The monthly soiling value in PVsyst represents the average soiling condition for that month — which depends on how frequently the modules are cleaned. If modules are cleaned every 14 days in a high-dustfall month, the maximum soiling at any point is approximately 14 days × (soiling rate per day), and the average over the month is approximately half the maximum.

Soiling rate per day = soiling per cleaning cycle / cleaning interval (days)

For Rajasthan in April (pre-monsoon, highest dustfall):

• Soiling accumulation rate: approximately 0.4–0.6% per day (based on published dustfall measurements)
• Cleaning interval: 10 days
• Peak soiling before cleaning: 4–6%
• Average soiling over the month: approximately 2–3%
• Monthly input value for PVsyst: 2–3%

For Rajasthan in October (post-monsoon, low dustfall):

• Soiling accumulation rate: approximately 0.05–0.1% per day
• Cleaning interval: 21 days
• Average soiling over the month: approximately 0.5–1.5%
• Monthly input value for PVsyst: 0.5–1.5%

---

India Soiling Reference — Regional Profiles

The PVsyst Monthly Soiling Profile Framework presents recommended monthly soiling input ranges for Indian utility-scale projects by region. Values assume standard manual water washing with the specified cleaning interval.

Region 1 — Rajasthan Desert (Jodhpur, Bikaner, Barmer, Jaisalmer)

The most extreme soiling environment in India’s solar belt. Thar Desert dustfall rates are among the highest globally — comparable to MENA desert sites. Rajasthan accounts for approximately 30% of India’s utility-scale solar capacity.

Month	Daily Soiling Rate (%/day)	Cleaning Interval	Monthly Soiling (avg)
January	0.10–0.15	21 days	1.0–1.5%
February	0.15–0.20	21 days	1.5–2.0%
March	0.25–0.35	14 days	1.5–2.5%
April	0.40–0.55	10 days	2.0–3.0%
May	0.45–0.60	10 days	2.5–3.5%
June	0.30–0.45	14 days	2.0–3.0% (dust storms)
July	0.10–0.15	Rain + 14 days	0.5–1.0% (monsoon washing)
August	0.05–0.10	Rain + 21 days	0.3–0.8%
September	0.05–0.10	Rain + 21 days	0.3–0.8%
October	0.08–0.12	21 days	0.8–1.2%
November	0.10–0.15	21 days	1.0–1.5%
December	0.10–0.15	21 days	1.0–1.5%
Annual average			3.5–5.5%

For the bankable IEA of a Rajasthan project, an annual average soiling of 3.5–5.5% is defensible with reference to published dustfall data. Values below 3% require specific justification (robotic cleaning, high-frequency manual cleaning, or site-specific low dustfall evidence).

Region 2 — Gujarat (Kutch, Saurashtra, Mehsana)

Gujarat’s Kutch district shares many characteristics with Rajasthan desert in terms of dustfall. The Rann of Kutch has extremely high seasonal dustfall. Inland Saurashtra and Mehsana have lower but still significant soiling rates.

Month	Kutch/Rann Soiling	Saurashtra/Mehsana
March–May	3.0–5.0%	2.5–3.5%
July–September	0.5–1.0% (monsoon)	0.5–1.0%
October–February	1.0–2.0%	0.8–1.5%
Annual average	3.0–5.5%	2.0–3.5%

Region 3 — Karnataka, Andhra Pradesh, Telangana

Central Deccan plateau — moderate soiling with distinct wet and dry seasons. Bellary and Kurnool areas are important utility-scale clusters.

Month	Soiling Range	Notes
January–February	1.0–1.5%	Dry season, moderate accumulation
March–April	1.5–2.5%	Peak pre-monsoon; agricultural dust
May–June	2.0–3.0%	Highest pre-monsoon period
July–September	0.3–0.8%	Southwest monsoon; natural washing
October–December	0.8–1.5%	Post-monsoon decline
Annual average	1.5–3.0%

Region 4 — Tamil Nadu (Tuticorin, Tirunelveli, Vellore)

Tamil Nadu has a coastal climate with monsoon rain from two seasons (southwest and northeast monsoon). Soiling is moderate compared to the Deccan interior.

Month	Soiling Range	Notes
January–March	0.8–1.5%	Dry northeast monsoon ending
April–May	1.5–2.5%	Pre-summer peak; lower than Deccan
June–September	0.5–1.0%	Southwest monsoon partial effect
October–December	0.5–1.0%	Northeast monsoon; regular rainfall
Annual average	1.0–2.0%

Region 5 — Maharashtra (Nashik, Solapur, Osmanabad)

Maharashtra covers a range of climates from the rain-shadow interior (high soiling) to the more humid western coast.

Subregion	Annual Average Soiling	Dry Season Peak
Vidarbha (Nagpur, Wardha)	2.5–4.0%	3.5–5.0% May
Marathwada (Aurangabad, Osmanabad)	2.5–4.5%	3.0–5.0% May
Nashik/Solapur	2.0–3.5%	2.5–4.0% May
Konkan Coast (Mumbai area)	0.8–1.5%	1.5–2.5% March

---

Africa Soiling Reference — Regional Profiles

Africa spans a wider range of climates than India’s solar belt, from the hyper-arid Sahara and Namibia to the humid sub-equatorial belt of Central Africa. Soiling behavior varies correspondingly.

North Africa — Morocco, Algeria, Egypt

The MENA transition zone. High irradiance and high soiling — the combination that makes solar economics strong despite cleaning costs.

Country / Region	Annual Average Soiling	Peak Month	Notes
Morocco (Ouarzazate area)	3.0–6.0%	March–April (harmattan sand)	Noor complex projects show 3–5% managed soiling
Algeria (Saharan fringe)	4.0–8.0%	February–April	Extreme dustfall in open desert
Egypt (Eastern Desert)	3.5–7.0%	March–May	Khamsin seasonal dust storms
Tunisia	2.0–4.0%	March–May	Moderate; Mediterranean rainfall in north

West Africa — Senegal, Mali, Burkina Faso, Ghana, Nigeria

The Sahel transition zone is characterized by the harmattan wind — a dry, dust-laden wind that blows from the northeast November through March, carrying Saharan dust southward.

Country	Annual Average Soiling	Harmattan Season	Wet Season
Senegal (northern)	3.5–6.0%	4.0–8.0% (Nov–Mar)	0.5–1.5% (Jul–Sep)
Mali (Bamako area)	4.0–7.0%	5.0–9.0% (Nov–Feb)	0.3–1.0% (Jul–Sep)
Burkina Faso	4.5–8.0%	5.0–10.0% (Nov–Mar)	0.3–1.0% (Jul–Sep)
Ghana (northern)	3.0–5.0%	4.0–6.0% (Nov–Feb)	0.5–1.5% (Jun–Aug)
Nigeria (northern)	3.5–6.0%	4.0–7.0% (Nov–Feb)	0.5–1.5% (Jun–Sep)

East Africa — Kenya, Tanzania, Ethiopia, Uganda

East Africa benefits from relatively moderate soiling compared to West Africa and North Africa, due to lower proximity to major dust transport corridors and more varied precipitation patterns.

Country	Annual Average Soiling	Notes
Kenya (Turkana, Garissa)	2.0–3.5%	Dry north significantly higher than coast
Tanzania (Dodoma, Singida)	1.5–3.0%	Interior higher than coast
Ethiopia (Rift Valley)	2.0–4.0%	Red dust in Rift Valley sites
Uganda	1.0–2.5%	Higher rainfall reduces soiling

Southern Africa — South Africa, Namibia, Zambia, Zimbabwe

South Africa’s solar belt — particularly the Northern Cape and Western Cape — has lower soiling than Sahelian West Africa, but the Namaqualand and Kalahari desert fringe can experience moderate dustfall.

Country / Region	Annual Average Soiling	Notes
South Africa (Northern Cape, Upington)	1.5–3.0%	Arid; moderate dustfall; Kalahari fringe
South Africa (Western Cape, Karoo)	1.0–2.5%	Lower dustfall; Western Cape more humid
Namibia (central plateau)	2.5–5.0%	Arid Namib fringe; high dustfall
Zambia, Zimbabwe	1.5–3.0%	Moderate; seasonal rainfall washing

---

PVsyst Monthly Soiling Profile — Input Workflow

Step 1 — Identify the Regional Soiling Source

For a bankable simulation, the monthly soiling profile should be traceable to at least one of:

1. Site-measured soiling data — actual IV curve measurements or irradiance sensor comparisons from a cleaned vs uncleaned sensor pair at the site. The highest quality source; typically available after 12+ months of monitoring on an operating plant or pilot measurement station.

2. Published regional soiling studies — academic papers, NREL soiling reports, IEA PVPS Task reports, or operator data from comparable existing plants in the region.

3. Dustfall measurement data — particle counter measurements or mass-based dustfall gauges at the site. Can be gathered in 3–6 months of pre-construction monitoring.

4. Comparable operating plant data — production data from an operating plant in the same region with documented cleaning frequency, normalized to monthly soiling loss.

Step 2 — Convert Dustfall to Monthly Soiling Values

If dustfall measurement is available (in g/m²/day or similar units), the conversion to soiling loss percentage requires an empirical factor that depends on particle size distribution and module tilt angle. Published conversion factors for Indian dust:

• Fine PM2.5 fraction: higher soiling impact per gram
• Coarse PM10 fraction: lower soiling impact per gram; sheds more easily
• Module tilt effect: steeper tilt (>20°) sheds coarser particles more effectively; finer dust adheres regardless of tilt

For Rajasthan, published studies reference a soiling rate of approximately 0.3–0.5%/day in peak dustfall months for horizontal modules, with 15–25% lower rate at 20–25° tilt due to particle shedding.

Step 3 — Apply Cleaning Protocol Adjustment

Adjust the daily accumulation rate for the cleaning frequency to derive the monthly average:

Monthly average soiling = (daily rate × interval / 2) × (days in month / interval)

This simplifies to: Monthly average ≈ daily rate × interval / 2

Where interval is the cleaning interval in days.

For Rajasthan in May with 0.50%/day accumulation and 10-day cleaning:

• Monthly average ≈ 0.50% × 10 / 2 = 2.5%

For Rajasthan in October with 0.12%/day accumulation and 21-day cleaning:

• Monthly average ≈ 0.12% × 21 / 2 = 1.3%

Step 4 — Enter Monthly Values in PVsyst

Navigate to Project → Losses → Module Array Losses → Soiling. Enter the twelve monthly values. PVsyst accepts decimal values to two decimal places. Verify that the annual average (sum / 12) is within the expected range for the region.

---

Cleaning Frequency Economics — The O&M Decision Model

The soiling loss assumption in PVsyst should drive the cleaning frequency decision, not follow it. The economic model:

Revenue per cleaning cycle:

ΔE (kWh) = P_installed (kW) × Soiling_delta (fraction) × Full-load-hours_between_cleanings

For a 10 MW plant in Rajasthan in May with 3% average soiling (before cleaning), daily production of ~50,000 kWh, and cleaning every 10 days:

• Average soiling across 10-day period = 1.5% (linear accumulation)
• Additional production from keeping soiling at 0 vs 1.5%: ~750 kWh/day × 10 days = 7,500 kWh/cycle
• Revenue per cleaning cycle at ₹3.50/kWh: ₹26,250

Cost per cleaning cycle:

Cleaning Method	Cost per MW per Cycle	For 10 MW
Manual water wash (tanker)	₹1,200–1,800	₹12,000–18,000
Semi-robotic (brushes + operator)	₹800–1,200	₹8,000–12,000
Fully robotic (autonomous)	₹300–600	₹3,000–6,000 + amortized robot cost

Net revenue per cleaning cycle:

• Manual wash: ₹26,250 revenue − ₹15,000 cost = ₹11,250 net per cycle
• Robotic: ₹26,250 revenue − ₹4,500 cost = ₹21,750 net per cycle

The economics strongly support bi-weekly cleaning in peak soiling months for most Indian utility-scale projects with manual washing, and more frequent cleaning (weekly) for robotic systems where marginal cost per cycle is low.

---

Documenting Soiling for Bankable IEA Submission

The soiling section of a bankable IEA typically includes:

1. Site location and climate context — latitude, annual rainfall, proximity to major dustfall sources (desert, agricultural land, construction, industrial zones)

2. Monthly soiling profile table — twelve monthly values with accompanying column showing: cleaning interval assumed, daily soiling rate source, and resulting average monthly soiling

3. Soiling data source citation — the reference for the daily soiling rate. Examples:

   • “Based on NREL India soiling study (Bergin et al., 2017) for Rajasthan desert sites, adjusted for tilt angle at 22°”
   • “Based on site measurement data from comparable operating 25 MW plant at Jodhpur, Rajasthan, 2021–2023”
   • “Based on regional dustfall measurements conducted at site from [date] to [date] using optical particle counters”

4. O&M contract linkage statement — “The monthly soiling profile assumes cleaning every [N] days during months X–Y (peak soiling season) and every [M] days during months A–B (low soiling season), consistent with the O&M contract cleaning schedule specification in Exhibit [X]”

5. Sensitivity analysis — P90 adjustment for soiling uncertainty. Typical sensitivity: ±1.0% soiling adds approximately ±0.5% to P50-to-P90 uncertainty spread.

The NREL soiling and forecasting methodology report and the IEA PVPS Task 13 performance and O&M reports are the standard external references for soiling methodology in bankable IEA documentation. For India specifically, Bergin et al.’s published soiling research for Indian solar sites is widely cited in IE reviews of IREDA and IFC-funded projects. The MNRE solar project guidelines and SEIA research resources provide additional context for acceptable soiling documentation standards.

---

How Heaven Designs Calibrates Soiling for Regional Projects

Heaven Designs calibrates soiling profiles for Indian and African utility-scale projects based on regional dustfall databases, peer-reviewed soiling studies, and comparable operating plant data from our project portfolio. Every bankable yield report includes a fully documented soiling section with monthly profiles, data sources, cleaning frequency linkage, and sensitivity analysis.

• Solar Ground Mount Design — Utility-scale PVsyst with regionally calibrated soiling profiles for India and Africa. IREDA/IFC-format IEA documentation.
• Solar Rooftop Detailed Engineering Design — C&I PVsyst with soiling profiles calibrated for urban, semi-urban, and industrial sites.
• Site Survey and Land Feasibility — Site soiling measurement and irradiance validation as PVsyst input package.
• MW-Scale PMC — Owner’s engineer oversight of O&M cleaning protocol compliance with PVsyst soiling assumptions.

Related posts: PVsyst Loss Diagram Interpretation | PVsyst vs Helioscope for Utility-Scale | PVsyst Meteo Data: Meteonorm vs SolarGIS vs NSRDB

Glossary: PVsyst, Performance Ratio, Soiling Loss.

---

FAQ

What soiling loss should I enter in PVsyst for a Rajasthan project?

For a utility-scale ground-mount project in Rajasthan with standard manual water washing (bi-weekly cleaning in peak months), the bankable soiling profile typically shows annual average soiling of 3.5–5.0%. Monthly peaks in April–May (pre-monsoon) range from 2.5–4.0%; monsoon months (July–September) are 0.3–1.0% due to natural rainfall washing. Annual average below 3% for a Rajasthan project without documented high-frequency robotic cleaning will typically receive a conservative adjustment from the independent engineer in the IEA review.

How is soiling loss different from degradation in PVsyst?

Soiling loss is the temporary reduction in energy output from particulate accumulation on the module surface — reversed by cleaning. Degradation is the permanent, irreversible reduction in module power output capacity over time due to UV exposure, thermal cycling, moisture ingress, and other aging mechanisms. In PVsyst, soiling is configured as a monthly loss profile under Module Array Losses; degradation is configured as an annual percentage decline under System Properties → Module Degradation. The two are independent and cumulative.

How does rainfall affect soiling modeling for African projects?

Rainfall provides natural module washing that resets soiling accumulation. For West Africa with a distinct wet season (June–September for Sahelian sites), the monsoon months should show significantly lower soiling (0.3–1.0%) compared to the harmattan dry season (4–9%). For East and Southern Africa with bimodal or year-round rainfall, soiling remains more moderate throughout the year. In PVsyst, the rainfall-washing effect is captured implicitly by entering lower monthly soiling values during rainy months — there is no automatic rainfall-washing calculation in PVsyst; it must be reflected in the input values.

What data sources are accepted in a bankable IEA for soiling documentation?

For bankable IEA documentation accepted by IREDA, IFC, AfDB, and US lenders, acceptable soiling data sources in descending order of quality: (1) site-measured IV curve or irradiance sensor data over 12+ months; (2) published peer-reviewed soiling studies for the specific region; (3) comparable operating plant soiling data from the same region with documented cleaning protocol; (4) regional dustfall measurement data from particle counters or mass gauges; (5) IEA PVPS Task 13 regional soiling benchmarks. Generic default values without regional calibration are not acceptable for projects above 5 MW.

How should soiling be documented differently for DFI-financed African projects vs IREDA-financed Indian projects?

DFI-financed African projects (AfDB, IFC, USAID) typically require more explicit soiling methodology documentation than IREDA-financed Indian projects, because published soiling benchmarks for many African subregions are less available than for Indian states. For African DFI projects, the IEA should include: the specific regional soiling study or dataset cited, a sensitivity analysis showing the impact of ±2% soiling assumption error on P90, and a statement on how the O&M contract cleaning protocol aligns with the soiling assumption. For IREDA-financed Indian projects, the soiling section documentation is more standardized because accepted regional benchmarks exist and IEs have established review conventions.

Can I use a single annual soiling percentage instead of monthly values in PVsyst?

PVsyst allows a single annual soiling percentage as a simplified input. For residential and commercial projects where monthly soiling variation is not significant, this is acceptable. For utility-scale projects in India and Africa where seasonality drives large variation between monsoon and dry-season soiling (e.g., 5% in May vs 0.5% in August for a Rajasthan site), using a single annual average significantly distorts the monthly production profile — overestimating monsoon production (where soiling is actually very low) and underestimating dry-season production (where soiling is high). Bankable IEA reports for utility-scale projects in seasonal climates should always use the monthly soiling profile.