Simple Explanation (ELI5)

Imagine you want to know if you'll need an umbrella tomorrow: instead of just peeking out the window, you ask three friends with different superpowers — one feels the heat, one sees humidity, one uses GPS to read the air. Together, they give you a much better forecast!

Key Advantages & Current Limits

🟢 Pros: - Greater accuracy for temperature and humidity near the ground, with automatic hourly corrections. - Combines different sensors by leveraging each one's strengths, making the system more reliable. 🔴 Cons: - Tested at only one station and for a few months, so results need validation in other climates. - Requires expensive infrastructure (lidar, radiometer, GNSS receivers) not always available at scale.

Future Outlook & Takeaways

In 3–5 years, these multi‑sensor networks could become standard in cities, improving local forecasts for extreme heat, pollution, and sudden storms, helping weather services issue faster and more accurate alerts to the public.

How Does the New System Combining Radar, Lidar, and GPS Read the...

Article

100%

A Chinese research team has developed an innovative method to measure temperature and humidity in the atmosphere using three ground‑based instruments: a microwave radiometer (GMWR), a Mie‑Raman lidar (MRL) and a GNSS/MET sensor. The new framework, called EnKF1D‑Var, merges data from these devices with GFS model forecasts, updating hourly with a vertical resolution of 15 m.

During three months, from May to July 2024, the system was tested at Anqing station in subtropical China. Results show a temperature bias reduction of over 3 % and humidity corrections up to 0.075 g/kg (120 PPMV), especially during daytime. The trick lies in the algorithm dynamically adjusting observation error matrices, weighting each instrument according to its contribution. GNSS/MET leads temperature corrections, while MRL drives humidity improvements. GMWR provides additional support.

This approach proves that integrating multiple ground‑based data sources yields more accurate thermodynamic profiles in the lower atmosphere, crucial for reliable weather forecasts and studying atmosphere‑urban interactions.

Bonus

100%

The EnKF1D‑Var framework merges two data assimilation techniques: the Ensemble Kalman Filter (EnKF) and one‑dimensional variational analysis (1D‑Var). EnKF generates an ensemble of possible atmospheric states from a forecast model (GFS). Each ensemble member is updated with observations via a linear correction based on the state‑observation covariance matrix. 1D‑Var, on the other hand, solves a local optimization problem along the vertical column, minimizing a cost function that includes background and observation departure terms weighted by their error matrices.

In this study, observations come from three sensors with distinct characteristics: GMWR measures microwave emission linked to water‑vapor content; MRL provides backscatter profiles at multiple wavelengths, sensitive to aerosol particles and water molecules; GNSS/MET detects GPS signal delays caused by atmospheric refractivity, related to temperature and humidity. Error covariance matrices are updated in real time using statistical diagnostics, allowing the filter to reduce uncertainty during rapid changes, such as daytime heating.

Assimilation proceeds in two steps: first EnKF updates the ensemble with observations, then 1D‑Var refines the solution along each vertical column, ensuring physical consistency (e.g., monotonic temperature). This hybrid approach leverages EnKF robustness against non‑linear errors and 1D‑Var precision for vertical scales, yielding a significant bias reduction for temperature and humidity in the atmospheric boundary layer.