چكيده لاتين
The troposphere is considered one of the most significant error sources in spatial geodesy. This layer has posed the most challenges in modeling and estimating errors in spatial methods due to the presence of water vapor. Temporal and spatial variations of this parameter have led researchers worldwide to investigate and model tropospheric parameters in different regions. Among these parameters are the Zenith Total Delay (ZTD), which includes both dry and wet components, horizontal gradients, Weighted Mean Temperature, and two parameters: IWV and PWV. This dissertation focuses on examining the variations and modeling of Weighted Mean Temperature as one of the most crucial tropospheric parameters in our dear country, Iran. It also investigates the impact of various meteorological resources on calculating tropospheric parameters. In the discussion of locally modeling Weighted Mean Temperature using ERA5 meteorological numerical data, two linear models, grouping and harmonic models, were obtained for the region. These models were compared with common global models, including the Bois model and the global model GPT3. Additionally, the impact of these modeling approaches on determining IWV parameters was examined using IGS data for Tehran city. To evaluate the results, data from 12 radiosonde stations in Iran were utilized. As a result, the linear local model (grouping) showed better results in radiosonde stations compared to the Bois model. Furthermore, the harmonic model exhibited higher accuracy, approaching that of the GPT3 model. In the IWV section, the local grouping model outperformed the Bois model, with the harmonic model showing a similar accuracy to the GPT3 model, approximately at 0.13 kg.m-2. For numerical data comparison, four renowned world datasets were utilized alongside the global GPT3 model. Two versions of ERA5 numerical data with spatial resolutions of 0.125 degrees and 2.5 degrees, ERA-Interim, and NCEP data with a spatial resolution of 2.5 degrees were used. According to the study results, ERA5 data with a spatial resolution of 0.125 degrees showed a higher accuracy of approximately 1 to 2 Kelvin compared to other datasets. Additionally, no noticeable dependence on accuracy concerning height was observed in the 12-radiosonde stations. In the IWV section, ERA5 data with a resolution of 0.125 degrees showed an accuracy of around 0.17 kg.m-2, surpassing the other three datasets and the GPT3 model. Furthermore, separating the RMSE results by months revealed that during the warm months (June, July, August), the accuracy was about 50% lower compared to other months due to higher water vapor values in these seasons.
Kaywords: Weighted Mean Temperature, Numerical Data, Integrated Water Vapor, Radiosonde Stations, IGS Station
