When open-source investigators evaluate contested battlefields, ecological disasters, or human rights violations, optical visual photography represents only a narrow slice of available ground truth. Cloud cover, smoke plumes from artillery strikes, nightfall, and deliberate camouflage frequently obscure visible-spectrum satellite passes (such as Google Earth or standard commercial RGB imagery).

To penetrate these observational barriers, investigative newsrooms increasingly rely on multispectral satellite remote sensing. By measuring electromagnetic radiation across infrared, near-infrared, and radar wavelengths, researchers can detect thermal missile strikes hidden under heavy smoke, calculate crop and forest destruction with mathematical precision, and map catastrophic floodwaters through dense overcast skies.

Through the European Space Agency’s Copernicus Sentinel-2 constellation—which photographs the entire Earth’s land surface every five days at 10-meter resolution—this data is available completely free of charge to independent journalists. This guide outlines the technical methodology for interpreting multispectral band combinations, calculating vegetation indexes, and utilizing radar SAR for open-source verification.

1. The Electromagnetic Spectrum: Beyond Visible RGB

Consumer cameras and human eyes perceive light exclusively across the visible spectrum: Red (~665 nm), Green (~560 nm), and Blue (~490 nm). The Sentinel-2 Multi-Spectral Instrument (MSI) captures 13 distinct spectral bands across visible, near-infrared (NIR), and shortwave infrared (SWIR) wavelengths:

        ELECTROMAGNETIC SPECTRUM UTILIZED BY SENTINEL-2
┌───────────────┬───────────────────────────────┬───────────────────────────────┐
│ Visible Light │      Near-Infrared (NIR)      │    Shortwave Infrared (SWIR)  │
│ 400nm - 700nm │        700nm - 1000nm         │       1000nm - 2500nm         │
├───────────────┼───────────────────────────────┼───────────────────────────────┤
│ Band 2 (Blue) │ Band 5, 6, 7 (Vegetation Edge)│ Band 11 (SWIR-1 / 1610nm)     │
│ Band 3 (Green)│ Band 8 (Wide NIR / 842nm)     │ Band 12 (SWIR-2 / 2190nm)     │
│ Band 4 (Red)  │ Band 8A (Narrow NIR / 865nm)  │ High Penetration of Smoke/Haze│
└───────────────┴───────────────────────────────┴───────────────────────────────┘

Band Resolution and Physical Sensor Parameters

• 10-Meter Spatial Resolution: Bands 2, 3, 4 (Visible) and Band 8 (Near-Infrared). Each pixel represents a 10m × 10m grid on the ground—sufficient for identifying large building footprints, bridge spans, runway craters, and major naval vessels.
• 20-Meter Spatial Resolution: Bands 5, 6, 7 (Red Edge) and Bands 11, 12 (SWIR). Critical for calculating moisture content, soil disruption, and active thermal combustion.
• 60-Meter Spatial Resolution: Bands 1, 9, 10. Primarily reserved for atmospheric correction, aerosol optical thickness, and cirrus cloud screening.

2. Band Combinations: Revealing the Invisible

In remote sensing software such as the Sentinel Hub EO Browser or QGIS, researchers map specific spectral bands to the red, green, and blue color channels of your monitor. Switching these band combinations reveals distinct physical realities:

┌────────────────────────────────────────────────────────┐
│ [TRUE COLOR (RGB: 4-3-2)]                             │
│ Looks like a conventional photograph. Obscured by      │
│ smoke, haze, and camouflage netting.                   │
└────────────────────────────────────────────────────────┘
                           │
                           ▼ (Shift to False Color Infrared)
┌────────────────────────────────────────────────────────┐
│ [FALSE COLOR (NIR-Red-Green: 8-4-3)]                   │
│ Healthy chlorophyll reflects intensely in Near-IR.     │
│ Vegetation glows bright scarlet red; burnt ground,     │
│ fresh trenches, and concrete appear dark grey/cyan.    │
└────────────────────────────────────────────────────────┘
                           │
                           ▼ (Shift to Shortwave Infrared)
┌────────────────────────────────────────────────────────┐
│ [SWIR THERMAL COMBUSTION (12-11-8)]                    │
│ Penetrates atmospheric smoke plumes completely.        │
│ Active artillery impacts, wildfires, and industrial    │
│ combustion glow bright neon orange and yellow.         │
└────────────────────────────────────────────────────────┘

1. True Color (Bands 4, 3, 2)

The standard natural visual representation. Use this as your baseline to locate recognizable geographic features, coastlines, urban grids, and major highways.

2. False Color Urban & Vegetation (Bands 8, 4, 3)

Healthy vegetation reflects up to 50% of incoming Near-Infrared radiation due to the internal cellular structure of leaves (spongy mesophyll). * In this band combination, living crops and forests appear in brilliant shades of crimson and red. * Non-vegetated surfaces—asphalt roads, concrete runways, concrete bunkers, and newly excavated trench earth—appear in cold shades of cyan, steel-blue, and dark grey. * Investigative Application: Easily identifies new defensive fortifications, tank tracks through agricultural fields, and military encampments carved out of forest canopies.

3. Shortwave Infrared / SWIR (Bands 12, 11, 8)

Shortwave infrared wavelengths are long enough to pass through small aerosol particles, allowing them to penetrate atmospheric smoke, fog, and light haze that completely block visible light. * Active fires and extreme heat events (such as industrial explosions, oil refinery strikes, and forest wildfires) emit intensely in Band 12 (2190 nm), appearing as vivid neon orange or fiery red hotspots. * Cooled burn scars and ash residues absorb infrared radiation, appearing as distinct deep reddish-brown or charcoal swathes. * Investigative Application: Verifying the exact point of impact during sustained artillery bombardments where optical satellite passes show only opaque white smoke.

3. Quantifying Damage: Normalized Difference Vegetation Index (NDVI)

While false-color imaging provides qualitative visual leads, human rights investigations and environmental accountability reporting require quantitative, peer-reviewed mathematical proof.

The Normalized Difference Vegetation Index (NDVI) calculates the numerical ratio between the red visible absorption band and the near-infrared reflection band:

$$\text{NDVI} = \frac{\text{Band 8 (NIR)} - \text{Band 4 (Red)}}{\text{Band 8 (NIR)} + \text{Band 4 (Red)}}$$

                NDVI MATHEMATICAL SCALE (-1.0 to +1.0)
┌──────────────┬───────────────────────────────┬───────────────────────────────┐
│ -1.0 to 0.0  │ Deep Water, Open Lakes, Ocean │ Intense NIR absorption        │
├──────────────┼───────────────────────────────┼───────────────────────────────┤
│  0.0 to 0.1  │ Bare Rock, Sand, Concrete     │ Equal reflectance in Red/NIR  │
├──────────────┼───────────────────────────────┼───────────────────────────────┤
│  0.2 to 0.4  │ Sparse Shrubland, Disturbed   │ Moderate vegetation density   │
├──────────────┼───────────────────────────────┼───────────────────────────────┤
│  0.6 to 0.9  │ Dense Rainforest, Healthy Crop│ Maximum chlorophyll reflection│
└──────────────┴───────────────────────────────┴───────────────────────────────┘

The NDVI Differential Methodology (dNDVI)

To prove that damage occurred within a specific timeframe: 1. Acquire Pre-Event Scene (T0): Select a cloud-free Sentinel-2 pass 5 days prior to the reported incident. Compute NDVI. 2. Acquire Post-Event Scene (T1): Select the first cloud-free pass immediately following the reported event. Compute NDVI. 3. Calculate the Differential ($\Delta\text{NDVI} = \text{NDVI}{T0} - \text{NDVI}$): * A localized, abrupt drop of $\Delta\text{NDVI} > 0.4$ across specific agricultural parcels confirms severe physical ground disturbance—such as heavy tracked vehicle movements, deliberate crop burning, or munitions cratering—eliminating natural seasonal drying as a variable.

4. Synthetic Aperture Radar (SAR): Penetrating Clouds and Darkness

During prolonged monsoon seasons, winter storms, or night operations, optical and infrared sensors are rendered ineffective by persistent 100% cloud cover.

The solution is Copernicus Sentinel-1 Synthetic Aperture Radar (SAR): * Sentinel-1 does not measure reflected sunlight; it is an active sensor that pulses microwave radar beams (C-band / ~5.4 GHz) down to Earth and measures the backscattered signal returned to the satellite antenna. * Radar waves pass effortlessly through cloud cover, rain, dust, and darkness.

                  RADAR BACKSCATTER DYNAMICS (SENTINEL-1)

       Smooth Water Surface                     Rough Rough Terrain / Urban
     Satellite Radar Pulse                 Satellite Radar Pulse
               ╲                                     ╲
                ╲                                     ╲
                 ╲                                     ▼
        ──────────╲──────────                  ┌──┐  ▲ ┌──┐  ▲
                   ╲ Specular Reflection       │  │  │ │  │  │ Corner Reflectors
                    ▼ (Bounces Away)           └──┴──┴─┴──┴──┴ (Bounces Direct Back)

        Sensor Receives: ZERO BACKSCATTER      Sensor Receives: INTENSE BACKSCATTER
        Visual Render: PURE BLACK PIXELS       Visual Render: BRILLIANT WHITE PIXELS

Radar Surface Interpretations:

• Standing Water & Flooding: Calm water acts as a specular mirror, reflecting the radar pulse forward and away from the satellite. The sensor receives zero backscatter, rendering water bodies and newly flooded floodplains in deep, stark black.
• Urban Concrete & Metal Structures: Buildings, bridges, and military shipping containers act as “corner reflectors,” bouncing microwave energy repeatedly between the wall and ground before sending a massive backscatter return directly to the sensor. Urban centers appear in blinding crystalline white.
• Dam Collapse & Flood Tracking: By overlaying a pre-disaster Sentinel-1 radar pass with a post-disaster pass using the Normalized Difference Water Index (NDWI), investigators can map the exact boundary of floodwaters through unbroken monsoon cloud cover.

5. Satellite Verification Clearance Protocol

Before publishing investigative findings based on satellite imagery, verify these four technical axes:

Conclusion: Remote Sensing as Open Evidentiary Proof

Commercial surveillance tasking is no longer a prerequisite for rigorous geospatial investigations. By mastering Sentinel-2 multispectral band combinations, calculating quantitative NDVI damage differentials, and deploying radar SAR through overcast skies, independent journalists possess the technical capabilities to hold state and military actors accountable from orbit.

Cross-examine satellite coordinates and log case dockets using our Verification Decision Tree & Triage Checklist.