Solar Suitability · Kamrup Metro

GIS & Remote Sensing · Drone-derived DSM · Weighted Overlay Analysis

Rooftop solar suitability, mapped roof by roof

A drone survey of an industrial and residential block in Kamrup Metro District, Assam, turned into a rooftop-level solar suitability assessment: solar radiation, slope, aspect and shadow are standardised, weighted and overlaid to rank every roof for photovoltaic installation.

Study area 26.142178° N, 91.725575° E Inputs Orthomosaic + DSM Processing QGIS · ArcGIS Pro Analysis Weighted overlay, 4 criteria

Heaviest criterion

40%

Solar radiation weight in the overlay

Shadow slices modelled

08:00 to 16:00, combined into one sunlight surface

Criteria in overlay

Radiation, shadow, slope, aspect — rooftop area as the unit

Suitability classes

Highly Suitable through Unsuitable

01 — Criteria & weights

What drives the score

Weights were assigned from expert judgement and published literature on GIS-based rooftop PV suitability analysis. Solar radiation carries the most weight because it acts directly on generation; slope, shading and aspect follow.

Solar radiationprimary driver of PV yield

40%

Shadow / sunlighthillshade-derived shading loss

30%

Sloperoof inclination

20%

Aspectroof orientation

10%

02 — Solar geometry

Five moments in the sun's path

Hillshade was computed on the DSM at five times of day to capture how surrounding buildings and roof structures throw shade across the rooftops. Each slice was reclassified so that sunlit roof area scores higher.

Sun altitude across the survey day

Sun azimuth across the survey day

Time of day	Sun azimuth	Sun altitude
08:00	92.7°	40.2°
10:00	115.0°	66.4°
12:00	214.4°	76.3°
14:00	259.5°	52.7°
16:00	274.2°	25.9°

03 — Classification

How a roof is banded

The weighted overlay index is classified from the most to the least suitable roof. Higher scores mean more radiation, less shadow, a flatter surface and a better orientation.

Highly Suitableflat, unshaded, south-facing, high radiation

Suitablegood radiation with minor slope or shading

Moderately Suitableusable, with orientation or shade penalties

Less Suitableconstrained roof geometry or heavy shading

Unsuitablesteep, heavily shadowed or negligible area

Note: the source document describes the scheme both as these five named classes (Step 9) and as a four-class dissolve (1–4) when rooftop areas are summarised per class; the map atlas below follows the five named classes. Suitability values on the underlying rasters are shown on a 1–4 scale in the overlay section and a 1–5 scale in the reclassification section.

04 — Methodology

From drone flight to suitability map

Nine steps chain the drone capture, terrain derivatives and overlay into one decision-ready surface.

Data collection

Drone survey over the study area; orthomosaic and DSM generated in Agisoft Metashape and Spatix.

Data preparation

DSM and orthomosaic imported to QGIS / ArcGIS Pro; building rooftop polygons digitised from the orthomosaic as analysis units.

Terrain analysis

Slope and aspect derived from the DSM — roof inclination and orientation.

Solar radiation

Area Solar Radiation tool (ArcGIS Pro) on the DSM, default atmospheric parameters, annual totals.

Shadow analysis

Hillshade from the DSM at five times of day; reclassified so sunlit roof area scores higher.

Rooftop area

Polygon area computed with the Field Calculator in QGIS.

Zonal statistics

Mean solar radiation sampled per rooftop polygon from the radiation raster.

Reclassification

Each criterion standardised into suitability classes; combined by weighted overlay.

Classification

Suitability index classed into Suitable / Moderately Suitable / Less Suitable / Unsuitable.

05 — Map atlas

The output layers

Every criterion layer and the final overlay, as produced in the analysis. Each map is shown with its legend — what every colour on the map stands for — alongside the legend as exported from QGIS in the source document. The suitability raster was finally polygonised and dissolved by class so each rooftop carries a measurable area.

**Orthomosaic — study area**Drone-derived orthomosaic of the Kamrup Metro sample area (26.142178° N, 91.725575° E). Building rooftop polygons were digitised from this image.

**Fig 1 · Slope map**Rooftop inclination in degrees, derived from the DSM with the QGIS Slope tool. Flatter roofs score higher.

Legend for this layer as it appears in the source document — **Fig 1 · Slope map**Rooftop inclination in degrees, derived from the DSM with the QGIS Slope tool. Flatter roofs score higher.

**Fig 2 · Aspect map**Direction each rooftop surface faces. South-facing roofs receive more radiation in the Northern Hemisphere and score higher.

**Fig 3 · Shadow map, 08:00**Binary shadow surface at sun azimuth 92.7° and altitude 40.2°. One of five such slices across the day.

**Fig 8 · Sunlight availability**The five shadow maps added together into a single sunlight-availability surface — the shading criterion fed into the overlay.

**Fig 9 · Solar radiation**Annual solar radiation from the Area Solar Radiation tool. The highest weighted criterion at 40%.

**Fig 10 · Site suitability map**Weighted overlay of radiation, shadow, slope and aspect. Higher values (dark red) indicate the most suitable rooftop locations.

**Fig 11 · Vectorised rooftops**The suitability raster polygonised and dissolved by class, with area computed per polygon for each suitability class.

**Fig 12 · Sample building**One building classified across suitability bands — the check used to validate the raster result against real roof conditions.

Sample building · rooftop area by class

For the sample building used as the worked check, the polygonised suitability classes were measured with the QGIS Field Calculator. The four class areas sum to the building's classified roof area — the same summary applied across the whole study area to estimate installable rooftop space.

Class	Area
Suitable	147.28 m²
Moderately Suitable	68.81 m²
Less Suitable	28.51 m²
Unsuitable	62.64 m²
Total sample roof	307.24 m²

06 — Findings

What the overlay shows

The overlay separates rooftops into distinct suitability bands rather than a single blanket score, so the large flat industrial roofs and the smaller residential roofs read differently on the final map.

The suitability map was validated by visual comparison against the high-resolution orthomosaic. Randomly selected roofs were checked to confirm the assigned class matched the roof's radiation, slope, aspect and available area. Consistency between the GIS classes and the observed roof conditions supported the reliability of the assessment.

Because the raster was polygonised and dissolved by class, the areas can be summed per class. That turns the map into a quantified estimate of the rooftop area available for PV installation across the study area — the step that makes the output usable for planning.

The framework is repeatable and low-cost: a drone flight plus open GIS tooling is enough to rank roofs and prioritise where panels should go first.

“The results identified rooftops with varying levels of suitability, enabling the prioritization of buildings with high solar potential, favourable roof characteristics and adequate installation space.”

Sources cited in the study

Polat, N. & Memduhoğlu, A. (2024). An Unmanned Aerial Vehicle Based Investigation of Roof Patch Suitability for Solar Panel Installation. International Journal of Engineering and Geosciences.
Ahamed, M. R. (2017). Assessment of Solar Potential Rooftops by Using GIS and Remote Sensing Techniques. Faculty of Geomatics, Sabaragamuwa University of Sri Lanka.