Global Satellite Navigation Systems are widely used for geodetic and geodynamics purposes. However the meteorological applications, such as Precipitable Water Vapor (PWV) estimation, are increased with GNSS permanent stations deployments all over the world. The continuity of GNSS observations and the spatial resolution of the permanent GNSS stations are some of the potentials of GNSS remote sensing using permanent arrays. In this study we are demonstrated one of the real-time meteorological applications of GNSS networks.  
The spatial distribution of PWV was investigated during extreme rainfall. The PWV data from the SuomiNet network stations in the Texas state was implemented. Using linear interpolation, the PWV were determined for area within the stations.
It was observed that the estimated water vapor from the GNSS observations progresses gradually towards the precipitation site and then, with accumulation in the area of ​​precipitation, the rainfall begins, and then the PWV decreases. Therefore, using a network of uniformly distributed GNSS stations, GNSS observations can be used to measure the accumulation of atmospheric precipitation in a region and to investigate the probability of rainfall occurring. These predictions will be effective if the network is sufficiently dense and the perceptible water vapor is estimated with a short latency.
The estimation of zenith path delay from GNSS is possible using relative or absolute method. Furthermore the slant delay estimation is one of the possibility in the dense GNSS networks. Tropospheric tomography will aid the scientists in the future applications of GNSS. In this paper the precision of real time PWV estimation via GNSS data is investigated. French RGP GNSS networks data are used for PPP processing. The processing is performed by ultra-rapid IGS orbit and clock products and then it repeated using final IGS products. The precision of Zenith Total Delay (ZTD) of final ephemeris is about 3 mm. The Real time estimation of ZTD using ultra rapid data is compared by final solution and the RMSE for different stations are from 3 to 7 mm approximately that is sufficient for real time estimation of PWV and real time precipitations prediction. Investigation of raining occurrence and the PWV changes is performed in this paper.
In the investigation of PWV it may be possible to follow a pattern or patterns for a region prior to intense rainfall, spatial variations and spatial distribution of PWV, which can predict extreme rainfall. Therefore, it is suggested that by studying the PWV behavior accurately, the probability of such patterns is examined. Also, in order to determine the accuracy of the PWV obtained from GNSS observations by the PPP method with the ultra-rapid orbit and clock products, it is possible to compare the PWV obtained from the above-mentioned method with those obtained from the measurement of radiosondes as a reliable source. Also the results of the ultra-rapid products are compared with the final IGS products the consistency is about 3-7 millimeters for the estimated ZTD values.
It is also possible to predict the rainfall by the permanent GNSS stations in Iran. There are several permanent arrays which may provide the GNSS observation files instantaneously. The national geodynamic network, the Tehran's Instantaneous Network, The national cadastre RTK network and the Isfahan municipality RTK network, could be used for PWV estimation with high spatial and temporal resolution and instantaneous meteorological application of a unified network is possible.