RTNET Real-time ZTD Estimation System

1. Introduction

1.1 GEONET

Figure: GEONET station distribution map.

Figure: Number of GEONET stations for each 0.1 x 0.1 degree grid

Geographical Survey Institute (GSI) in Japan has deployed ground-based GPS stations since 1994, and the GPS network called GPS Earth Observation NETwork (GEONET) consists of about 1,300 ground-based GPS stations now. The mean spacing of the GEONET station is now 17km.

GEONET dataset has been real-time capability since 2003.

1.2 Feasibility

Precise Point Positioning (PPP) processing strategy in GPS analysis is the comutation-time efficient approach for GPS processing, especially for processing giant GPS network fast.

The accuracy and precision of PPP estimates depends on satellite orbit and satellite clock products.

With real-time data stream and real-time satellite clock estimation using some 20 GEONET station dataset we can estimate satellite clock, and the satellite clock can be used for PPP ZTD estimation.

Thus, real-time ZTD estimation is feasibile for GEONET with real-time data streaming.

2. Strategy

2.1 Estimation of satellite clock

Figure: Schematic diagram of double difference strategy.

The selected GEONET stations above were used for satellite clock estimation. RTNET estimate the satellite clocks by using IGU predicted orbit (accuracy is about 10cm). The optimized number of stations for the purpose of real-time ZTD estimation is about 20.

Figure: GEONET stations used for satellite clock estimation.

2.2 PPP estimation of ZTD

Figure: Schematic diagram of precise point positioning strategy.

We can estimate ZTD with RTNET PPP using the estimated satellite clock by RTNET. The sampling interval is from 30 sec to 5 min. Station coordinates are fixed using the averaged coordinate for 7 days. One or two PCs are enough to process whole GEONET stations, and they just need to handle real-time data streaming of GEONET data.

Figure: Weathermap of 00 UTC of AUG 26, 2005 (left), ZTD (middle) estimated by RTNET with real-time strategy, and the approximate ZWD (right) subtracted climatological ZHD from the ZTD.

Figure shows snapshot ZTD and ZWD estimated by RTNET with real-time strategy. The ZTD distribution of the real-time RTNET is spatially coherent, and there are no remarkable gaps of the ZTD.

4. Discussions

4.1 Comparison of renet real-time ZTD with RAOB

Figure: Time series of ZTD estimated with RTNET and GIPSY, and ZTD based on 12 hourly operational RAOB at Akita.

Table: Statistics (mm) of comparison between rtnet ZTD and radiosonde ZTD (radiosonde minus GPS) for two days (Aug 24 and Aug 25) in 2005. 13 pairs of radiosonde-GEONET stations were selected.

	GIPSY	RTNET real-time
Bias	-3.0	0.8
Standard deviation	21.1	16.5
RMS	21.3	16.6

13 pairs of radiosonde-GEONET stations were selected with height difference is less than 50 m. The standard deviations for the comparison are the same order with historical comparison researches in Japan.

---

Meteorology

• Regional and localized forecasts by GPS data assimilations
• Nowcasting of severe rainfall

Geodesy

• Correction of tropospheric delay for space-based geodetic instruments and processing.

Navigation

• Correction of tropospheric delay for GPS navigation

Atmospheric Corrections
Rocken, C. et al.: Real-time Ionospheric and Atmospheric Corrections for Wide Area Single Frequency Carrier Phase Ambiguity Resolution, to be presented at ION GNSS 2006

Nowcasting
Iwabuchi, T. et al.: PPP True Real-time 30 sec Estimation of ZTD in Dense and Giant Regional GPS Network and the Application of ZTD for Nowcasting of Heavy Rainfall, to be presented at ION GNSS 2006

GPS Solutions Inc.

IGS : International GNSS Service

ION : The Institute of Navigation

GEONET

GSI : Geographical Survey Institute

NGS : Nippon GPS Solutions