• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.1
• /
• pp.23-28
• /
• 2021

Data analysis research using deep learning has recently been studied in various field. In this paper, we conduct a GNSS (Global Navigation Satellite System)-based meteorological study applying deep learning by estimating the ZWD (Zenith tropospheric Wet Delay) through MLP (Multi-Layer Perceptron) and LSTM (Long Short-Term Memory) models. Deep learning models were trained with meteorological data and ZWD which is estimated using zenith tropospheric total delay and dry delay. We apply meteorological data not used for learning to the learned model to estimate ZWD with centimeter-level RMSE (Root Mean Square Error) in both models. It is necessary to analyze the GNSS data from coastal areas together and increase time resolution in order to estimate ZWD in various situations.

• Journal of Astronomy and Space Sciences
• /
• v.24 no.3
• /
• pp.249-260
• /
• 2007

The crustal deformation due to Ocean Tide Loading (OTL) in the Korean peninsula reaches up to ${\sim}3cm$ in the vertical direction. Considering that the achievable positioning accuracy of current state-of-the-art space geodesy technologies is at the several millimeter level, the centimeter-level OTL effect should be precisely modelled and corrected for. This study begins with comparison of ocean tide models and validation of OTL-prediction softwares. Different ocean tide models caused about ${\sim}6mm$ RMS differences in the vertical deformation in the Kyung-gi Bay area. When we analyzed the OTL displacements in the Seoul, Ulsan, and Seogwipo areas where three VLBI observatories are planned to be installed, the maximum displacement of ${\sim}3.5cm$ was predicted in the Seogwipo area and ${\sim}2cm$ in the Seoul and Ulsan areas. When the OTL corrections were not applied in the GPS data processing, the OTL effect propagates into the Zenith Wet Delay (ZWD) estimates, and the scale factor between ZWD differences and OTL displacements was 3.72.

• Journal of Navigation and Port Research
• /
• v.36 no.10
• /
• pp.825-832
• /
• 2012

GPS signal delay that caused by dry gases and water vapor in troposphere is a main error source of GPS point positioning and it must be eliminated for precise point positioning. In this paper, we implemented to generate tropospheric delay grid map over the Korean Peninsula based on post-processing method by using the GPS permanent station network in order to determine the availability of tropospheric delay generation algorithm. GIPSY 5.0 was used for GPS data process and nationwide AWS observation network was used to calculate the amount of dry delay and wet delay separately. As the result of grid map's accuracy analysis, the RMSE between grid map data and GPS site data was 0.7mm in ZHD, 7.6mm in ZWD and 8.5mm in ZTD. After grid map accuracy analysis, we applied the calculated tropospheric delay grid map to single frequency relative positioning algorithm and analyzed the positioning accuracy enhancement. As the result, positioning accuracy was improved up to 36% in case of relative positioning of Suwon(SUWN) and Mokpo(MKPO), that the baseline distance is about 297km.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.1
• /
• pp.33-39
• /
• 2018

As the retrieval accuracy of PWV estimates from GPS measurements is proportional to the accuracy of water vapor WMT, the WMT model is a significant formulation in the conversion of PWV from the GPS ZWD. The purpose of this study is to develop a MWMT model for the retrieval of highly accurate GPS PWV using the radiosonde measurements from six upper-air observing stations in the region of Korea. The values of 1-hr PWV estimated at four GPS stations during one year are used to evaluate the validity of the MWMT model. It is compared to the PWV obtained from radiosonde data that are located in the vicinity of GPS stations. Intercomparison of radiosonde PWVs and GPS PWVs derived using different WMT models is performed to assess the quality of our MWMT model for Korea. The result in this study indicates that the MWMT model is an effective model to retrieve the enhanced accurate GPS PWV, compared to other GPS PWV derived by Korean annual or global WMT models.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.384-386
• /
• 2003

GPS observation has been performed at Khon Kaen in northeast Thailand to investigate the Precipitable Water Vapor (PWV) change since August 2001 by using a Trimble 4000SSi receiver. The data obtained in the period from March to June in 2002 were processed by using CAMIT software to obtain the Zenith Tropospheric Delay (ZTD) at every one hour referring to some IGS stations around Thailand. We estimated the Zenith Hydrostatic Delay (ZHD) at every three hours with barometer data at Khon Kaen of Thai Meteorological Department, The Zenith Wet Delay (ZWD) was obtained by subtracting ZHD from ZTD and PWV can be calculated from ZTD. The results obtained shows that PWV changes with a large amplitude in March and April before the monsoon onset, and also we can see steep PWV increases before rain and decreases after rain. In May and June after the onset, the PWV is almost constant to be 60 to 70 mm, but there is a semi-diurnal change which has high PWV values at about 8 and 20 o'clock in local time.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.27 no.5
• /
• pp.565-574
• /
• 2009

The water vapor weighted vertically mean temperature(Tm) models, which were developed by the consideration of seasonal characteristics over the Korea, was used in the retrieval of precipitable water vapor (PWV) from GPS data which were observed at four GPS permanent stations. Since the weighted mean temperature relates to the water vapor pressure and temperature profile at a site, the accuracy of water vapor information which were estimated from GPS tropospheric wet delay is proportional to the accuracy of the weighted mean temperature. The adaption of Korean seasonal weighted mean temperature model, as an alternative to other formulae which are suggested from other nation, provides an improvement in the accuracy of the GPS PWV estimation. Therefore, it can be concluded that the seasonally appropriate weighted mean temperature model, which is used to convert actual zenith wet delay (ZWD) to the PWV, can be more reduced the relative biases of PWV estimated from GPS signal delays in the troposphere than other annual model, so that it would be useful for GPS PWV estimation with high accuracy.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.4
• /
• pp.233-242
• /
• 2019

ZHD (Zenith Hydrostatic Delay) model is important parameter in estimating of GNSS (Global Navigation Satellite System) PWV (Precipitable Water Vapor) along with weighted mean temperature. The ZWD (Zenith Wet Delay) is tend to accumulate the ZHD error, so that biases from ZHD will be affected on the precision of GNSS PWV. In this paper, we compared the accuracy of GNSS PWV with radiosonde PWV using three ZHD models, such as Saastamoinen, Hopfield, and Black. Also, we adopted the KWMT (Korean Weighted Mean Temperature) model and the mean temperature which was observed by radiosonde on the retrieval processing of GNSS PWV. To this end, GNSS observation data during one year were processed to produce PWVs from a total of 5 GNSS permanent stations in Korea, and the GNSS PWVs were compared with radiosonde PWVs for the evaluating of biases. The PWV biases using mean temperature estimated by the KWMT model are smaller than radiosonde mean temperature. Also, we could confirm the result that the Saastamoinen ZHD which is most used in the GNSS meteorology is not valid in South Korea, because it cannot be exclude the possibility of biases by latitude or height of GNSS station.