Global Horizontal Irradiance

What Is Global Horizontal Irradiance?

Global Horizontal Irradiance (GHI) is the total solar radiation energy incident on a horizontal surface per unit area, measured in watts per square metre (W/m²). It represents the sum of two physical components: Direct Normal Irradiance (DNI): the direct beam from the sun’s disk, and Diffuse Horizontal Irradiance (DHI): sunlight scattered by the atmosphere, clouds, and aerosols across the entire sky dome.

The mathematical relationship is:

GHI = DNI × cos(θz) + DHI

Where θz is the solar zenith angle (the angle between the sun and the vertical). When the sun is directly overhead (zenith angle = 0°), cos(0°) = 1 and GHI equals DNI plus DHI. At sunrise and sunset, the zenith angle approaches 90°, the cosine term approaches zero, and GHI is dominated by diffuse light.

GHI is the universal reference metric for solar resource assessment. Solar atlases, project finance models, and academic research all use GHI as the baseline because horizontal measurement is unambiguous, it does not depend on panel tilt, azimuth, or tracking configuration. For actual solar plant design, GHI is translated to Plane of Array (POA) irradiance using the specific tilt and orientation of the installed modules.

For residential solar and commercial rooftop projects in Gujarat, understanding local GHI is the first step in accurate system sizing and payback calculation. A site with about 5.6 kWh/m²/day GHI (Kutch) receives roughly 20% more annual irradiation than a site with about 4.7 kWh/m²/day (Delhi NCR), and identical equipment produces broadly proportionally more energy, a difference that compounds over 25 years into lakhs of rupees.

Why Global Horizontal Irradiance Matters

GHI is the single most important input variable in solar energy yield prediction. Every other design decision, module selection, inverter sizing, tilt angle, string configuration, builds upon the GHI foundation. An error in GHI assessment propagates through the entire financial model.

Critical impacts include:

• Energy yield prediction: A 5% error in GHI translates directly to a 5% error in predicted annual generation. For a 1 MW plant generating 1.6 million kWh/year, this is 80,000 kWh, worth Rs 5-8 lakh annually at commercial tariffs.
• Project financing: Lenders require P90 GHI assessments (exceeded in 90% of years) to size debt conservatively. Poor GHI data increases financing costs or reduces leverage.
• System sizing: Home solar system sizing starts with GHI. A 5 kW system in high-GHI Rajasthan produces more than a 5 kW system in low-GHI Assam, affecting inverter loading, DC-AC ratio, and cable sizing.
• Revenue forecasting: Power Purchase Agreements (PPAs), open access structures, and net metering returns all depend on accurate GHI-derived generation estimates.
• Technology selection: High-DNI sites favor concentrating solar power (CSP) and tracking systems. High-DHI sites (cloudy, humid) favor fixed-tilt silicon PV with high low-light performance.
• Seasonal planning: Monthly GHI distribution reveals monsoon generation dips, enabling accurate O&M scheduling and battery sizing for hybrid systems.

Important: Heaven Green Energy uses multi-decade GHI data from NIWE Solar Atlas and Solargis for all solar EPC projects in Gujarat, ensuring P90-compliant resource assessments that satisfy lender and investor requirements.

How Global Horizontal Irradiance Works

GHI measurement and application follow a structured technical workflow:

1. Solar radiation at the top of atmosphere:

• The solar constant is approximately 1,361 W/m², the irradiance on a surface perpendicular to the sun’s rays at Earth’s average distance from the sun.
• As sunlight passes through the atmosphere, it is attenuated by absorption (ozone, water vapor, aerosols) and scattering (Rayleigh scattering by air molecules, Mie scattering by particulates).

2. Ground-level measurement:

• A pyranometer mounted horizontally measures the remaining radiation that reaches the surface.
• Thermopile pyranometers (Secondary Standard per ISO 9060) achieve ±2% annual uncertainty.
• Silicon-cell pyranometers are cheaper but spectrally limited, with ±5% uncertainty.

3. Data integration:

• Instantaneous GHI in W/m² is integrated over time to produce:
  • Hourly totals: Wh/m²
  • Daily totals: kWh/m²/day
  • Annual totals: kWh/m²/year
• These integrated values are what solar designers use for energy yield calculations.

4. Satellite-derived estimation:

• Where ground stations are sparse, satellites measure reflected visible and infrared radiation.
• Algorithms (Heliosat, MAGIC, etc.) convert satellite radiances to GHI estimates.
• Modern satellite datasets (Solargis, SolarAnywhere) achieve 5-8% annual uncertainty in India.

5. Translation to POA:

• GHI is converted to Plane of Array irradiance using:
  • Panel tilt angle and azimuth
  • Direct beam and diffuse sky models (Perez, Hay-Davies, Reindl)
  • Ground reflectance (albedo)
• For south-facing panels at 25° tilt in Gujarat: POA ≈ 1.05-1.12 × GHI annually. Heaven Designs’ PVsyst resource center covers these transposition models in more depth for design teams building bankable simulations.

6. Energy yield calculation:

• POA irradiance × module efficiency × system performance ratio = AC energy output.
• This step is typically run in PV yield simulation software such as PVsyst, SAM, PVWatts, or Helioscope.

Visual Explanation

Real-World Example

Scenario: A textile factory in Surat, Gujarat is evaluating a 500 kWp commercial rooftop solar installation.

GHI assessment process:

1. Data source: Heaven Green Energy’s design team pulls 20-year GHI data from NIWE Solar Atlas and Solargis for Surat coordinates (21.17°N, 72.83°E).
2. Annual GHI: 5.35 kWh/m²/day average = 1,953 kWh/m²/year.
3. Monthly distribution (Global Solar Atlas long-term averages for Surat):
   • April: 7.2 kWh/m²/day (clear pre-monsoon sky)
   • July: 3.8 kWh/m²/day (monsoon clouds)
   • Annual range: 3.8 to 7.2 kWh/m²/day
4. P90 analysis: 20-year data shows P90 GHI = 1,835 kWh/m²/year (6% below mean).
5. POA calculation: South-facing panels at 21° tilt receive POA = 2,085 kWh/m²/year (6.8% gain over GHI).
6. Energy yield: 500 kWp × 2,085 kWh/m²/year × 18% performance ratio losses = 855,000 kWh/year.
7. Financial impact: At Rs 9.20/kWh commercial tariff, annual savings = Rs 78.7 lakh. Over 25 years with 0.5% annual degradation: about Rs 18.5 crore cumulative savings.

Key insight: The 6% difference between mean GHI and P90 GHI translates to Rs 4.7 lakh annual revenue variance, sufficient to change debt service coverage ratios and lender comfort. Using P90 GHI for conservative modeling is standard practice at Heaven Green Energy.

Technical Specifications / Benchmarks

City values are long-term averages for each city point from the Global Solar Atlas (World Bank / Solargis, accessed October 2026). Nearby sites can differ.

Benefits / Advantages

• Universal comparability: GHI on a horizontal surface is unambiguous, enabling direct comparison across sites, countries, and time periods.
• Foundation for all solar design: Every energy yield model, financial projection, and system specification starts with GHI data.
• Mature measurement infrastructure: ISO 9060 standardizes pyranometer accuracy classes, ensuring consistent, traceable measurements globally.
• Free data availability: NIWE Solar Atlas, NASA POWER, and PVGIS provide free GHI data for preliminary Indian site assessments.
• Satellite coverage: Where ground stations are absent, satellite-derived GHI fills gaps with 5-8% accuracy, sufficient for early-stage feasibility.
• Long-term stability: Multi-decade GHI records show predictable patterns, supporting 20-25 year project finance with statistical confidence.
• Seasonal insights: Monthly GHI distributions reveal generation seasonality, enabling informed battery sizing, O&M planning, and tariff optimization under TOD structures.
• Technology selection guidance: High-DNI/low-DHI sites favor tracking; high-DHI sites favor fixed-tilt with high low-light performance modules.
• Risk quantification: P50/P90 GHI statistics quantify inter-annual resource risk for lender and investor due diligence.
• Climate change baseline: Long-term GHI records detect gradual shifts in solar resource, informing adaptive asset management.

Limitations / Drawbacks

• Not what panels actually see: GHI is on a horizontal surface. Tilted panels at optimal angle see roughly 3-12% more (POA), depending on latitude. Using GHI directly for energy yield without POA translation causes underprediction.
• Spatial variability: Within a single large project (100+ MW), GHI can vary 2-5% across the site due to terrain, microclimate, and cloud patterns.
• Temporal resolution: Annual average GHI masks critical seasonal variation. Monsoon months may see 40-50% reduction from pre-monsoon (April-May) peaks in places such as Surat and Mumbai.
• Measurement cost: Secondary Standard pyranometers cost Rs 2.5-4 lakh, limiting dense ground station networks. Satellite data has 5-8% uncertainty.
• Urban pollution impact: Aerosols and particulate pollution in Indian cities can measurably reduce GHI compared with cleaner rural sites at the same latitude.
• Data gaps: Northeast India and Himalayan regions have sparse ground station coverage, increasing reliance on less-validated satellite datasets.
• Albedo is not included: GHI excludes ground-reflected light, which varies with surface type (grass, concrete, water) and must be modelled separately for tilted and bifacial arrays.
• Not a complete design input: GHI alone does not account for temperature effects, soiling, shading losses, or module degradation, all critical for final yield.

Comparison Section

For rooftop solar and ground-mount PV, GHI is the starting point that gets translated to POA. For concentrating solar thermal (CSP), DNI is the direct and only relevant metric.

Applications

• Residential solar design: Home solar sizing begins with local GHI to estimate annual generation per kW installed.
• Commercial rooftop feasibility: C&I project developers use GHI data in solar calculators to present savings estimates to CFOs and facility managers.
• Utility-scale resource assessment: Lender-grade projects require 10-20 years of GHI data for P50/P90 yield analysis and debt sizing.
• Agricultural solar: PM-KUSUM pump solarization projects use GHI to size arrays for irrigation load matching.
• Solar park master planning: Ground-mount solar parks use spatial GHI mapping to optimize row spacing, tracker configuration, and MW-block layout.
• Battery storage sizing: Seasonal GHI profiles inform battery bank sizing for hybrid systems needing evening discharge.
• Grid integration studies: DISCOMs use regional GHI data to forecast distributed solar injection patterns and plan grid reinforcement.
• Climate research: Long-term GHI trends inform studies on atmospheric changes, pollution impacts, and climate feedback loops.
• Insurance and risk modeling: Actuarial models use GHI variability statistics to price generation risk and weather derivatives.
• International project comparison: GHI enables investors to compare solar resource quality across India, Middle East, Africa, and other markets on a consistent basis.

Industry Standards & Regulations

• ISO 9060:2018: Specification and classification of pyranometers by accuracy class (Class A, B and C, which replaced the older Secondary Standard, First Class and Second Class names).
• WMO Guide to Meteorological Instruments: Global standard for solar radiation measurement methodology, calibration, and maintenance.
• IEC 61724-1:2021: Photovoltaic system monitoring, guidelines for irradiance sensor selection, installation, and data quality.
• NIWE Solar Atlas: Authoritative India-specific solar resource database maintained by the National Institute of Wind Energy.
• CERC/SERC Tariff Orders: Regulatory frameworks that use normative capacity utilisation factors, which depend on the solar resource, for tariff determination.

Important: Heaven Green Energy’s solar EPC quality protocol requires GHI data from at least two independent sources (NIWE + Solargis or PVGIS) for all projects above 100 kW, cross-validating resource assumptions before financial commitment.

India-Specific Context

India possesses exceptional solar resource quality by global standards:

• Global ranking: Most of India averages roughly 4.5 to 5.6 kWh/m²/day GHI, comparable to southern Spain and well above Germany or the UK (under 3 kWh/m²/day).
• Regional champions: Western Rajasthan (Jaisalmer, Barmer, Jodhpur), Gujarat (Kutch, Banaskantha) and Ladakh receive India’s highest solar resource; Ladakh’s cold, high-altitude air gives tilted panels exceptional yields.
• Monsoon modulation: The Southwest Monsoon (June-September) pulls monthly GHI roughly 25-45% below pre-monsoon peaks across much of India; the dry far northwest sees a smaller dip. This seasonality is critical for system sizing and grid integration.
• Pollution penalty: Delhi NCR and Indo-Gangetic Plain cities see GHI reduced by aerosol loading and fog during winter months (October-January), affecting generation during peak demand periods.
• Gujarat advantage: With about 5.3-5.7 kWh/m²/day across the sites we checked, Gujarat has some of India’s highest annual GHI, though its July-August monsoon dip is deep. Strong resource is part of why Gujarat had about 35 GW of installed solar capacity as of 31 August 2026 (MNRE).
• Data infrastructure: NIWE operates a network of ground solar radiation resource assessment (SRRA) stations across India. Solargis provides satellite-derived coverage at about 250 m spatial resolution.
• Emerging markets: Northeast India (Assam, Meghalaya, Nagaland) has lower GHI (roughly 4.0-4.8 kWh/m²/day) but untapped potential as grid connectivity improves.

Future Trends

• Sub-hourly satellite data: Next-generation geostationary satellites (INSAT-3DS, Himawari-9) enable 10-15 minute GHI resolution, improving cloud-tracking and short-term forecasting for grid operators.
• Machine learning enhancement: AI models combining satellite imagery, weather forecasts, and ground measurements are reducing GHI prediction error.
• IoT pyranometer networks: Low-cost, connected silicon-cell sensors are creating dense urban GHI monitoring networks for distributed solar planning.
• Digital twin integration: Real-time GHI feeds into plant digital twins for predictive O&M, soiling detection, and performance anomaly identification.
• Climate-resilient projections: Climate models are beginning to project GHI changes under warming scenarios, informing long-term asset valuation and insurance pricing.
• Bifacial albedo modeling: Advanced GHI models now incorporate ground-reflected radiation (albedo) more accurately, critical for bifacial module yield prediction.
• Agrivoltaic optimization: GHI data is being combined with crop photosynthesis models to optimize panel spacing and height for dual land use.
• Blockchain-verified data: Decentralized GHI data marketplaces may emerge, allowing project developers to purchase verified, auditable resource datasets for due diligence.

Common Mistakes & Misconceptions

1. Using GHI directly for energy yield without POA translation: This underpredicts generation by roughly 3-12% for optimally tilted panels. Always convert GHI to POA using proper transposition models.
2. Relying on single-year GHI data: One unusually sunny or cloudy year distorts 25-year projections. Use 10-20 year datasets minimum.
3. Confusing GHI with DNI: DNI is only the direct beam; GHI includes diffuse. CSP projects need DNI, not GHI.
4. Ignoring seasonal variation: Annual average GHI hides monsoon dips that affect cash flow timing and battery sizing.
5. Using outdated solar atlases: Older, coarse-resolution datasets can carry larger errors in some Indian regions. Use modern NIWE, Solargis, or PVGIS datasets.
6. Treating satellite GHI as ground-truth: Satellite data requires local validation. Install on-site pyranometers for utility-scale projects.
7. Neglecting spatial variability: Large projects (100+ MW) may have 2-5% GHI variation across the site. Multiple measurement points are needed.
8. Forgetting urban pollution impact: City installations can see lower GHI than rural sites at the same latitude due to aerosol loading.
9. Assuming GHI stability means POA stability: Panel soiling, degradation, and shading change over time, affecting actual yield even if GHI is constant.
10. Using free data for lender-grade projects: Free datasets (NASA, PVGIS) are suitable for preliminary screening. Project finance requires premium paid datasets (Solargis, Meteonorm) with validated uncertainty bounds.

Key Takeaways

• Global Horizontal Irradiance (GHI) is the total solar radiation (direct + diffuse) on a horizontal surface, measured in W/m² or kWh/m²/day.
• GHI is the universal reference for solar resource assessment and the foundational input for all energy yield predictions.
• Most of India averages roughly 4.5-5.6 kWh/m²/day GHI, with annual totals from about 1,450 kWh/m²/year (parts of the Northeast) to about 2,050 kWh/m²/year (western Rajasthan and Kutch).
• GHI must be translated to Plane of Array (POA) irradiance for actual panel yield calculations, adding roughly 3-12% for optimally tilted panels.
• The relationship GHI = DNI × cos(zenith) + DHI connects GHI to its direct and diffuse components.
• Multi-decade GHI data (10-20+ years) is essential for accurate financial modeling; single-year data is insufficient.
• P50, P75, and P90 GHI statistics quantify inter-annual variability for lender and investor risk assessment.
• Ground-measured GHI (Secondary Standard pyranometer) achieves ±2% uncertainty; satellite-derived GHI achieves 5-8%.
• Free GHI sources (NIWE, NASA, PVGIS) suit preliminary analysis; premium sources (Solargis, Meteonorm) are required for project finance.
• Heaven Green Energy uses validated multi-source GHI data for all Gujarat solar projects, ensuring accurate sizing and conservative financial projections.

Frequently Asked Questions

What is Global Horizontal Irradiance? GHI is the total solar irradiance (direct plus diffuse) on a horizontal surface, measured in watts per square metre (W/m²) at any instant. Integrated over time, it becomes kWh/m² per day, month, or year, the standard solar resource metric used globally.

How is GHI measured? By a pyranometer mounted horizontally. The thermopile or silicon-cell sensor absorbs all incoming solar radiation from the full sky hemisphere and produces a calibrated voltage proportional to irradiance.

What is the relationship between GHI, DNI, and DHI? GHI = DNI × cos(zenith angle) + DHI. DNI is Direct Normal Irradiance (beam perpendicular to sun). DHI is Diffuse Horizontal Irradiance (scattered sky light). The cosine term accounts for the angle between the sun and vertical.

What is India’s annual average GHI? Most of India averages roughly 4.5 to 5.6 kWh/m²/day, ranging from about 4.0 in the cloudiest parts of the Northeast to about 5.6 in Jaisalmer and Kutch (Global Solar Atlas long-term averages). Annual integrated values: roughly 1,450 to 2,050 kWh/m²/year.

Why is GHI measured on a horizontal surface? Horizontal measurement is a universal, unambiguous reference. Tilted-surface irradiance (POA) requires defining specific tilt and azimuth angles. GHI enables direct comparison across sites and time periods.

How does GHI vary across India? Strongly. Western Rajasthan and Gujarat Kutch receive about 5.5-5.7 kWh/m²/day. Northeastern states receive roughly 4.0-4.8. Latitude, monsoon intensity, atmospheric pollution, and elevation all drive local variation.

Does GHI change over time? Multi-decade GHI is relatively stable with 3-6% year-to-year variation. Long-term climate cycles (ENSO, PDO) cause minor shifts. Climate change may slightly alter cloud patterns but GHI remains predictable for project finance.

How is GHI used in solar plant design? GHI is the foundational resource input. Site-specific GHI data combined with tilt, azimuth, and shading analysis produces Plane of Array (POA) irradiance, which drives expected energy output and financial returns.

Is GHI the same as insolation? Yes, when both refer to total solar energy on a horizontal surface. GHI in W/m² is instantaneous power density; integrated over time it equals insolation in kWh/m². Different names, identical physical quantity.

How accurate are GHI measurements? Ground-measured GHI: 2% uncertainty (Secondary Standard pyranometer) to 5% (First Class). Satellite-derived GHI: 5-8% uncertainty depending on region, algorithm, and validation data quality.

Where can I find GHI data for India? NIWE Solar Atlas (free, India-specific). NREL SAM and PVWatts (free, global). PVGIS (free, European Commission). Solargis and Meteonorm (paid, premium quality for project finance). NASA POWER (free, satellite-derived).

Does GHI include reflected light from the ground? No. A horizontal, upward-facing sensor sees only the sky hemisphere, so ground-reflected light is not part of GHI. Ground reflection (albedo) enters when GHI is transposed to a tilted plane (POA), and matters most for bifacial modules.

What is the difference between GHI and POA? GHI is on a horizontal surface. POA (Plane of Array) is on the tilted module surface at its actual orientation. For Indian latitudes, optimally tilted panels receive roughly 3-12% more annual irradiance than horizontal, less in the south and more in the north.

Can I use a single year of GHI data for solar projections? No. Multi-decade averages (10-20+ years) are essential for accurate projections. Single-year data may capture an anomalously sunny or cloudy year, distorting financial models.

What are P50, P75, and P90 GHI values? P50 is the median value exceeded 50% of years. P75 is exceeded 75% of years. P90 is exceeded 90% of years (conservative, used by lenders for debt sizing). These statistics quantify inter-annual variability.

Related Glossary Terms

• Solar Irradiance
• Direct Normal Irradiance
• Diffuse Horizontal Irradiance
• Peak Sun Hours
• Insolation
• Pyranometer
• Met Station
• Tilt Angle
• Azimuth
• Performance Ratio
• Capacity Utilization Factor
• Shading Loss
• Soiling Loss
• Degradation

Related Resources

• Solar Panel Efficiency Guide
• How to Choose Solar Modules
• Solar Installation Day by Day
• Home Solar System Size Guide
• Solar Calculator
• Residential Solar

Sources & References

• NIWE Solar Atlas, National Institute of Wind Energy, India
• NREL PVWatts and SAM Solar Resource Databases
• Solargis Satellite-Derived Solar Resource Data
• Meteonorm Global Meteorological Database
• WMO Guide to Meteorological Instruments and Methods of Observation
• ISO 9060:2018, Solar Energy, Specification and Classification of Instruments
• NASA POWER Project, Satellite-Derived Solar Data
• Global Solar Atlas (World Bank / Solargis), city-level long-term GHI averages
• MNRE, State-wise installed capacity of renewable power as on 31.08.2026

Authored by Rajesh Patel, Senior Solar Design Engineer (12+ years in rooftop solar EPC). Reviewed by Vikram Mehta, NABCEP Certified PV Installation Professional.