What is linearly polarized GNSS radio occultation and why does it matter?

Linearly polarized GNSS radio occultation (GNSS RO) enables the detection of precipitation by measuring the phase difference between horizontal (H) and vertical (V) signal components.

This technique adds a new capability to traditional GNSS RO by identifying precipitation-related signatures, including:

• Melting layers
• Rain and snow structure
• Horizontal precipitation bands
• Storm intensity variations

Unlike traditional satellite sensors, this method can detect hydrometeor shape effects, providing new insight into precipitation physics.

What is GNSS radio occultation (GNSS RO)?

GNSS RO measures how signals from GNSS satellites bend and slow as they pass through the atmosphere to a receiver satellite.

It provides global atmospheric profiles of:

• Temperature
• Pressure
• Water vapor
• Winds

Key advantages

• ~100 m vertical resolution
• Works day and night
• Penetrates clouds and storms
• Near-global, unbiased coverage
• Profiles down to near the surface

What does polarization add to GNSS RO?

Traditional GNSS signals are transmitted as right-hand circular polarization (RCP). When received, they can be separated into:

• Horizontal polarization (H)
• Vertical polarization (V)

Precipitation particles (like raindrops and snowflakes) are not spherical.
They tend to be flattened, which causes a measurable H–V differential phase delay

Key signal behavior

• No precipitation: H–V delay ≈ 0 mm
• With precipitation: H–V delay can exceed 20 mm

This makes the differential phase a direct indicator of precipitation presence and structure.

How PlanetiQ measures polarized GNSS RO

PlanetiQ has implemented dual linear polarization measurements on two satellites in orbit.

Instrument configuration

• Forward-facing antenna array
• Each element receives both H and V
• Independent tracking of H and V signals

Data products

• Standard RCP profiles for Numerical Weather Prediction (NWP)
• Additional polarized differential phase product

How precipitation is retrieved from the signal

2D Sliding Window Phase Matching (SWPM)

The retrieval uses a two-dimensional approach based on Jensen et al. (2004):

1. Construct a 2D grid:
   • Bending angle
   • Impact height
2. Model expected Doppler signal
3. Compare modeled vs. observed signal
4. Compute signal-to-noise ratio (SNR)
5. Filter low-quality data (below 4σ noise threshold)

Polarized retrieval method

1. Process H and V channels independently
2. Compute complex conjugate product
3. Extract H–V phase difference
4. Calibrate to zero at 35 km altitude

What precipitation signatures are observed?

1. Melting layer detection (freezing level peak)

• Strong H–V peaks occur near the freezing level
• Indicates transition from snow → melting → rain
• Matches temperature context from atmospheric models

2. Horizontal precipitation bands (2D advantage)

• 2D retrieval reveals sharp horizontal banding
• These features are lost in traditional 1D profiles

 This is a major advancement over earlier GNSS RO methods

3. High-altitude precipitation signals

• Detected up to ~16 km altitude
• Potential extension to ~19 km for anvil cloud detection

4. Strong bending and extreme cases

• Observed bending angles up to ~50 milliradians (~3°)
• Still retains structured precipitation signatures

5. Surface reflection signals

• Some cases show polarized signatures from surface reflections
• Possible sensitivity to:
  • Ocean temperature
  • Salinity

This suggests a secondary application beyond precipitation

Hurricane case study: what does polarized GNSS RO reveal?

A case study of four occultations over ~13 hours (Oct 2025) shows:

Key observations

• Strong precipitation bands
• H–V phase shifts:
  • 20 mm common
  • Peak exceeding 31 mm
• Cloud tops (~14 km) from infrared data
• Precipitation detected up to 12–14 km

Why this matters

• Captures storm structure in 3D
• Detects differences between:
  • Storm core
  • Outer bands
• Identifies features like:
  • Water vapor gradients
  • Sharp atmospheric transitions

Notable anomaly

• Occasional negative H–V phase
• Possible explanation:
  • a change in hydrometeor orientation
• Still under investigation

Why this technique is unique

Compared to traditional sensors

• Limb-viewing geometry detects particle shape effects
• Similar concept to dual-polarization radar, but from space

Key advantage

 Sensitivity to hydrometeor flattening, which nadir-viewing satellites cannot observe

Limitations and challenges

• Precipitation is sparse (~10% of Earth at any time)
• Sampling is inherently limited
• Requires:
  • Advanced retrieval methods
  • Careful calibration

What needs to be developed next?

To make this operationally useful:

1. Forward operator development

• Needed for data assimilation into weather models

2. Higher altitude retrievals

• Extend from ~16 km → ~19 km
• Target: hurricane anvils

3. Improved uncertainty quantification

• Use dual-frequency retrievals

4. Operational integration

• Enable use in forecasting systems like those from NOAA

Key takeaway

PlanetiQ introduces a new way to observe precipitation from space by using linearly polarized GNSS RO measuring H–V phase differences caused by hydrometeor shape.

It provides:

• Vertical structure
• Horizontal banding
• Storm intensity signals

This makes it a powerful new tool for weather analysis and future forecasting systems