Prediction and Monitoring of Precipitable Water Vapor (PWV) Profiles Using GNSS-Radio Occultation (GNSS-RO) Satellite Technique in Turkey

Water vapor is a natural greenhouse gas that shapes weather patterns and can condense in the atmosphere. Almost all of the water vapor, which plays an active role in the hydrological cycle, is found in the troposphere. The concentration of water vapor is highly variable according to location and temperature. In recent years, the increase in temperatures on the earth's surface is causing global warming. Therefore, water vapor analysis as an indicator of global warming is becoming more important. The increase in global temperatures is caused by natural processes and human activities. In particular, human activities are causing more carbon dioxide and methane to be released into the atmosphere. As a result, the absorption of greenhouse gases is increasing and the earth and climate are getting warmer. Water evaporating from the oceans, seas and land surfaces with more heat causes more water vapor to enter the atmosphere. The increase in the amount of water vapor held in the air leads to more cloud formation, heavy rainfall and extreme weather events such as floods and droughts. The ability to predict and monitor the occurrence of extreme weather events depends on precisely analyzing the amount of precipitable water vapor (PWV) in the atmosphere. In this context, the GNSS-RO technique offers unique advantages. The GNSS-RO technique does not require calibration, can operate in all weather conditions, and provides rich data at high vertical resolution with global coverage.
In this study, PWV forecasts for the period 2020-2023 were made by applying the GNSS-RO technique in the regions where radiosonde stations (RS) are located in Turkey. WetPf2 profiles of the FORMOSAT-7/COSMIC-2 Low Earth Orbit (LEO) satellite were used for GNSS-RO PWV forecasts. PWV time series obtained from wetPf2 profiles were compared with RS PWV calculated from IGRA (Integrated Global Radiosonde Archive) atmospheric profiles and PWV time series calculated from European Medium Range Weather Forecast (ECMWF) ERA5 hourly specific humidity profiles. For the comparisons, the distance between the COSMIC-2 RO and RS locations was limited to a maximum of 300 km and the ERA5 locations were limited to be at the closest latitude and longitude to the RS locations. COSMIC-2 RO events can occur at different locations at different times of the day. Therefore, each RO file of a day is analyzed separately with the RS and ERA5 profiles of the same day. Then, average values were calculated for the obtained PWVs. Linear regression was performed to examine the relationship between COSMIC-2 RO PWVs and RS and ERA5 PWVs. Squared mean errors and correlation coefficients were also calculated for statistical evaluation. The results obtained were found to be consistent with each other.