Iranian researchers developed a new model for more accurate atmospheric water-vapor monitoring
Published 30 September, 2026
Water vapor is the invisible moisture in the air. Although it makes up only a small part of the atmosphere, it helps form clouds and rainfall and influences the weather. Measuring it accurately is therefore important for atmospheric research.
Weather balloons measure moisture directly, but they are launched only at certain times and places. Ground-based global navigation satellite system (GNSS) receivers provide a more continuous source of information. As satellite signals pass through moist air, they slow slightly. Researchers can use this delay to estimate precipitable water vapor (PWV), the total amount of water vapor contained in a vertical column above a location.
The conversion of PWV is strongly dependent on atmospheric weighted mean temperature, or Tm. Tm combines the temperatures of different atmospheric layers with the amount of water vapor they contain. Any error in Tm affects the final PWV estimate directly, and a bias of a few degrees can introduce a noticeable error.
Global Tm models, such as the widely used Bevis model, provide generalized values that work reasonably well at a global scale, but they are not optimized for regional climates. In a study published in Geodesy and Geodynamics, Arash Tayfehrostami and Yazdan Amerian developed three regional Tm models calibrated using radiosonde observations and multi-mission GNSS radio occultation profiles over Iran, providing a more representative framework for regional atmospheric conditions. Their main difference is how they use surface temperature (Ts):
(i) Model One does not require Ts. (ii) Model Two assumes a linear relationship: each degree of surface-temperature change has approximately the same effect on Tm. (iii) Model Three uses a curved relationship, meaning that the effect of a one-degree change can vary at different temperatures. This gives Model Three greater flexibility, but also makes it more sensitive to inaccurate input data.
When tested against independent and reliable weather-balloon observations, Models Two and Three performed almost identically. The result changed with satellite radio-occultation profiles, which do not provide a reliable direct measurement of Ts. Models Two and Three therefore became less accurate, with Model Three’s curved relationship magnifying the uncertain input. Model One avoided this source of error and reduced the overall error by 42.6% compared with the Bevis model.
The researchers then tested the models at GNSS stations in Tabriz and Tehran. Model One reduced PWV estimation errors by approximately half at both locations. “No model is best in every situation,” the authors emphasized, “Models Two and Three are useful when accurate surface temperature is available, and Model One turned out to be the most robust one, particularly in contexts where data cannot be reliably obtained.”
However, although the models were calibrated using atmospheric observations over Iran, the proposed methodology provides a transferable regional modeling framework that can be recalibrated and independently validated for application in other climatic regions. The limited number of high-altitude stations and possible seasonal biases in satellite sampling also require further study.
Contact the author:
Arash Tayfehrostami (Faculty of Geodesy and Geomatics Engineering, K. N. Toosi University of Technology, Tehran, Iran)
a.tayfehrostami@email.kntu.ac.ir
Yazdan Amerian (Faculty of Geodesy and Geomatics Engineering, K. N. Toosi University of Technology, Tehran, Iran)
Conflict of interest:
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
See the article:
Tayfehrostami, A., & Amerian, Y. (2026). A new model for estimating atmospheric weighted mean temperature from radiosonde and multi-mission GNSS radio occultation data. Geodesy and Geodynamics, 17(2), 197-210. https://doi.org/10.1016/j.geog.2025.09.009