What Is DHI (Diffuse Horizontal Irradiance)? Definition & Guide

Key Takeaways

DHI is the portion of solar radiation scattered by the atmosphere before reaching a horizontal surface — it arrives from all directions across the sky dome, not just the sun’s position
On a clear day, DHI typically represents 15–25% of GHI; under overcast skies, it can reach 100% of total irradiance
The fundamental irradiance equation is GHI = DNI x cos(theta_z) + DHI, making DHI one of the two core inputs for any solar energy simulation
DHI is the dominant energy source for flat-mounted and low-tilt systems, especially in cloudy climates like Northern Europe or the Pacific Northwest
Accurate DHI data from sources like NREL’s NSRDB or Meteonorm is critical for reliable production estimates in solar design software
Four distinct types of diffuse radiation exist: clear-sky (Rayleigh), cloud-diffused, aerosol-scattered, and circumsolar — each with different spectral and angular properties

What Is Diffuse Horizontal Irradiance (DHI)?

Diffuse Horizontal Irradiance (DHI) is the component of solar radiation received on a horizontal surface from all parts of the sky except the direct solar disk. It results from sunlight being scattered by atmospheric molecules, water vapor, aerosols, and clouds before it reaches the ground.

Unlike Direct Normal Irradiance (DNI), which travels in a straight line from the sun, DHI arrives from every direction in the sky hemisphere. On a perfectly clear day, the sky still scatters a significant amount of blue light (Rayleigh scattering), producing a baseline of diffuse radiation. On overcast days, nearly all radiation reaching the surface is diffuse.

DHI is the “hidden” solar resource. On a cloudy day in Berlin or Seattle, direct beam radiation drops to near zero — but DHI keeps panels producing at 20–40% of rated capacity. Ignoring or underestimating DHI leads to systematically pessimistic yield forecasts in high-latitude and maritime climates.

Types of Diffuse Radiation

Diffuse radiation is not a single phenomenon. It comes from four distinct scattering mechanisms, each with different characteristics:

Clear-Sky DHI (Rayleigh Scattering)

Even under cloudless conditions, atmospheric molecules scatter short-wavelength (blue) light in all directions. This produces a baseline DHI of 50–120 W/m² depending on sun elevation and atmospheric clarity. Rayleigh scattering is strongest at shorter wavelengths, which is why the sky appears blue. It sets the minimum DHI value for any given solar geometry.

Cloud-Diffused DHI

Clouds scatter sunlight across a wide range of wavelengths and directions. Thick overcast layers convert nearly all incoming radiation to diffuse. Thin cirrus clouds scatter selectively, adding 20–80 W/m² of diffuse radiation while still allowing partial direct beam transmission. This is the largest source of DHI variability between climate zones.

Aerosol-Scattered DHI

Dust, pollution, smoke, and sea salt particles scatter light via Mie scattering. Unlike Rayleigh scattering, aerosol scattering is less wavelength-dependent and tends to forward-scatter light into a halo around the sun. In polluted or dusty regions, aerosol-scattered DHI can add 30–60 W/m² above the clean-atmosphere baseline.

Circumsolar DHI

The bright ring of scattered light immediately surrounding the solar disk (within about 5 degrees). Circumsolar radiation behaves almost like direct beam for tracking systems but is technically diffuse. It can represent 5–15% of total DHI on hazy days. Many irradiance models separate circumsolar from isotropic diffuse for better accuracy on tracked arrays.

DHI by Climate Type

DHI varies dramatically by location and climate. The table below shows typical daily values and the fraction of total GHI that comes from diffuse radiation:

Climate Type	Typical DHI (kWh/m²/day)	DHI as % of GHI	Location Examples
Arid desert	1.0–1.5	15–25%	Phoenix, Riyadh, Atacama
Mediterranean	1.5–2.2	25–35%	Southern Spain, Rome, Los Angeles
Subtropical humid	1.8–2.5	30–40%	Houston, Mumbai, Brisbane
Temperate oceanic	1.5–2.3	40–55%	London, Amsterdam, Seattle
Continental	1.3–2.0	30–45%	Berlin, Chicago, Warsaw
Tropical monsoon	2.0–2.8	35–50%	Bangkok, Jakarta, Manila
Subarctic	0.5–1.5	45–65%	Reykjavik, Anchorage, Tromsø

Why DHI Matters More for Flat-Mounted Panels

A horizontal or low-tilt panel receives irradiance primarily from the sky dome rather than from a focused beam. In cloudy climates like Northern Europe, DHI dominates the solar resource. A rooftop system in Amsterdam with a 15-degree tilt gets roughly 50% of its annual energy from diffuse radiation. This means accurate DHI data is more important than DNI data for flat residential and commercial rooftop designs. Use shadow analysis software with location-specific irradiance data to capture this correctly.

The Fundamental Irradiance Decomposition

DHI is one half of the equation that defines the total solar resource at any location:

Irradiance Decomposition Equation

GHI = DNI × cos(θ_z) + DHI

Where:

GHI (Global Horizontal Irradiance) = total solar radiation on a horizontal surface
DNI (Direct Normal Irradiance) = beam radiation measured perpendicular to the sun’s rays
θ_z (solar zenith angle) = angle between the sun and the vertical
DHI = diffuse horizontal irradiance (scattered component)

This equation is the foundation of every solar energy simulation. When you input a location into solar design software, the tool retrieves GHI, DNI, and DHI from a weather database and uses transposition models (like Perez or Hay-Davies) to convert these horizontal values into the actual irradiance reaching the tilted panel surface.

DNI × cos(θ_z) gives the direct component on a horizontal plane. DHI adds everything else — the scattered light from the entire sky dome. Together, they account for all solar radiation reaching a flat surface.

How DHI Is Measured and Modeled

Ground Measurements

DHI is measured using a pyranometer fitted with a shadow band or tracking disk that blocks direct beam radiation. The sensor then records only the diffuse component. Shadow-band measurements require a correction factor (typically 5–15%) to account for the band also blocking some diffuse sky radiation.

High-quality DHI measurement stations are part of the Baseline Surface Radiation Network (BSRN) and similar national networks. However, ground stations are sparse — there are fewer than 1,000 research-grade stations worldwide.

Satellite-Derived Data

Because ground stations are limited, most solar projects rely on satellite-derived DHI estimates. Key databases include:

NREL NSRDB — Covers the Americas with 4 km × 4 km spatial resolution and 30-minute temporal resolution. Uses the Physical Solar Model (PSM) to decompose satellite cloud observations into GHI, DNI, and DHI.
Meteonorm — Global coverage with interpolated data from over 8,000 weather stations and satellite imagery. Widely used in European project development.
PVGIS (EU JRC) — Free European and African coverage using SARAH satellite data. Provides monthly and hourly typical meteorological year (TMY) datasets.
SolarAnywhere — Commercial high-resolution data for the Americas.

Decomposition Models

When only GHI data is available (the most commonly measured quantity), DHI must be estimated using decomposition models. Common approaches include:

Erbs model — Correlates the diffuse fraction (DHI/GHI) with the clearness index (GHI / extraterrestrial irradiance)
Orgill and Hollands — Similar correlation-based approach with different coefficients
DISC/DIRINT models — Used by NREL in the NSRDB; incorporate solar zenith angle and atmospheric parameters
Engerer2 — A newer model that accounts for cloud enhancement effects

Each model introduces uncertainty, typically 5–15% on an hourly basis and 3–8% on an annual basis. This is why using measured DHI (or high-quality satellite-derived DHI) rather than decomposed values produces more reliable energy yield estimates.

DHI in Solar Design Workflows

Site Assessment

During site assessment, DHI data helps determine whether a location’s solar resource is beam-dominated or diffuse-dominated. This affects technology selection:

High DHI fraction (more than 40% of GHI): Flat or low-tilt fixed systems perform relatively better than single-axis trackers, since trackers primarily boost direct beam capture
Low DHI fraction (under 25% of GHI): Tracking systems gain maximum advantage since most radiation arrives as direct beam

Shading Impact

DHI also determines how much energy a shaded panel still produces. A panel in full shade loses all direct beam radiation but continues to receive diffuse radiation from the unshaded sky. In locations with high DHI fractions, shading losses are proportionally smaller. Shadow analysis software accounts for this by separating direct and diffuse shading losses in the simulation.

Energy Yield Simulation

Modern solar design software uses DHI as a direct input to transposition models that calculate Plane of Array (POA) irradiance. The Perez model, the most widely used transposition model, divides DHI into three sub-components:

Isotropic diffuse — uniform radiation from the entire sky dome
Circumsolar brightening — concentrated ring around the sun
Horizon brightening — increased diffuse near the horizon

This three-component approach captures the non-uniform distribution of diffuse radiation across the sky, improving accuracy over simpler isotropic models by 2–5% in annual yield estimates.

Model Site-Specific Solar Resource with Accurate Irradiance Data

SurgePV integrates high-resolution irradiance databases including DHI, DNI, and GHI to deliver accurate energy yield simulations for any location.

Book a Demo

No commitment required · 20 minutes · Live project walkthrough

DHI vs. DNI: When Each Component Dominates

Understanding when DHI or DNI dominates helps designers make better decisions:

Condition	DHI Dominant?	DNI Dominant?	Practical Implication
Overcast sky	Yes (80–100% of GHI)	No	Fixed-tilt systems lose less relative to trackers
Clear desert sky	No (15–20% of GHI)	Yes	Tracking systems gain 25–35% over fixed-tilt
Partly cloudy	Moderate (30–50%)	Moderate	Cloud enhancement can briefly push GHI above clear-sky values
High aerosol (haze)	Yes (35–50%)	Reduced	Forward-scattered circumsolar partially offsets DNI loss
Low sun angle (morning/evening)	Yes (increasing fraction)	Reduced	DHI fraction rises as zenith angle increases

Improving DHI Accuracy in Your Projects

Use location-specific TMY data rather than regional averages. A 10 km shift in location can change annual DHI by 5–10% near coastlines or mountain ranges.
Check the data source year range. TMY files built from 10+ years of satellite data are more reliable than single-year datasets for capturing DHI variability.
Validate with ground measurements when available. For utility-scale projects, compare satellite-derived DHI against the nearest BSRN or national network station.
Account for local aerosols. Urban, industrial, and agricultural regions have higher aerosol loads that increase DHI and decrease DNI relative to clean-atmosphere models.
Update for climate trends. Cloud cover patterns are shifting due to climate change. Check whether your TMY data reflects recent (post-2010) conditions rather than historical averages from the 1990s.

Sources

NREL National Solar Radiation Database (NSRDB) — High-resolution solar irradiance data including DHI for the Americas
World Meteorological Organization (WMO) — Standards for solar radiation measurement and instrumentation
Meteonorm — Global irradiance database with interpolated DHI values from station and satellite data

Frequently Asked Questions

What is DHI in solar energy?

DHI (Diffuse Horizontal Irradiance) is the solar radiation that reaches a horizontal surface after being scattered by the atmosphere. It includes light scattered by air molecules, clouds, aerosols, and dust. Unlike direct beam radiation (DNI), DHI arrives from all directions across the sky dome. It typically accounts for 20–50% of total solar radiation depending on location and weather conditions.

Why is DHI important for solar panel design?

DHI determines how much energy solar panels produce under cloudy and overcast conditions, and how much output remains when panels are partially shaded. In cloudy climates, DHI can represent the majority of the available solar resource. Accurate DHI data is a required input for energy yield simulations in solar design software, and errors in DHI estimation directly translate to errors in production forecasts and financial projections.

How is DHI different from GHI and DNI?

GHI (Global Horizontal Irradiance) is the total solar radiation on a horizontal surface. DNI (Direct Normal Irradiance) is the beam component measured perpendicular to the sun. DHI is the scattered component on a horizontal surface. They are related by the equation GHI = DNI x cos(zenith angle) + DHI. GHI is the easiest to measure, DNI matters most for tracking systems, and DHI matters most for fixed flat-mounted panels in cloudy climates. All three are used together in solar simulation to calculate energy yield at any tilt and orientation.