Special Study Group 1.159
Ground-Based GPS Meteorology
Michael Bevis
Hawaii Institute of Geophysics and Planetology
University of Hawaii
Honolulu, HI 96822


Ground-based GPS meteorology has both advanced and expanded quite dramatically during the last four years. When this SSG was formed in 1995 the notion of using continuous GPS (or CGPS) networks to monitor atmospheric water vapor in support of basic meteorological research, climatology and operational weather analysis and prediction was still fairly new (Bevis et al., 1992, 1994; Rocken et al., 1993; Yuan et al., 1993) and despite the success of the first large proof-of-concept experiment "GPS/STORM" (Rocken et al., 1995; Duan et al., 1996) many meteorologists familiar with the idea remained rather sceptical about the prospect of using emerging regional CGPS networks as meteorological observing systems. Now, at the time of the 1999 IUGG meeting, there are at least one hundred scientists and engineers working on ground-based GPS meteorology in more than one dozen countries, and several large national or multinational groups have been formed to engage collaborative research and development of GPS met, including meteorological impact studies. This influx of researchers has led to very rapid development on the geodetic side of the technique, and far wider interest by the meteorological community. This report provides some background on the scope of these activities, including a listing of some major R&D efforts and a bibliography (1995 – present).

While IAG SSG 1.150 focuses on ground-based GPS meteorology, SSG 2.161 focuses on the space-based or occultation approach. Christopher Rocken has reviewed this complementary class of GPS meteorology in his SSG 2.161 report. While ground- and space-based GPS meteorology have been developed largely independently in the last few years, it is likely that in the next few years considerably more effort will be focused on using them together. In the rest of this report the term ‘GPS meteorology’ will refer solely to the ground-based approach unless explicitly stated otherwise.

National and Regional GPS Meteorology Programs

The first continuous GPS network built specifically for meteorological measurements was the NOAA GPS PW Network led by NOAA’s Forecast System Laboratory (FSL) in the USA. This project is working in a quasi-operational status and is mainly focused on measuring the impact of GPS PW data on numerical weather modeling and on improving quality control for nearly real-time solutions. Perhaps the single largest project to date is the Japanese GPS Met Project which seeks to exploit the 1,000 station CGPS network operated in Japan by GSI. This project has completed its initial proof-of-concept phase, and is now focused on R&D relevant to establishing an operational status. More than any other GPS met project established to date, the Japanese effort emphasises using meteorological products to benefit real-time geodesy as well as exploiting geodesy for meteorology. The first extensive European trials of GPS met were mounted through projects WAVEFRONT, BALTEX and MAGIC funded by the EC. More recently the EC has adopted COST Action 716 "Exploitation of Ground-Based GPS for Climate and Numerical Weather Prediction Applications" which emphasises operational exploitation of GPS met. A new German project associated with this cost action, known as the ‘GPS Atmosphere Sounding Project’, is already performing nearly real-time GPS PW measurements in an operational setting, and hopes to add many more CGPS stations to its network in the near future. This project emphasises integration of both the ground- and space-based approaches.

Several of these projects are described in more detail in the appendices. Here is a list of the larger projects (known to our SSG) with contact information:






USA Forecast Systems Laboratory 


S. Gutman
Japanese GPS  

Meteorology Project

Japan IGS, JMA, CRL, NRAO and  

several universities

I. Naito, Y. Hatanaka

(recently concluded)

Europe See appendix 3 A. Dodson
BALTEX Baltic Sea 

& adjacent areas

  Jan Johansson jmj@oso.chalmers.se
MAGIC Mediterranean 

& adjacent areas

See appendix 5 Jennifer Hase  


COST Action 716 Europe European Community 

See appendix 3

G. Elgered 

A. Dodson

GPS Atmosphere Sounding Project Germany & Europe GFZ and many others, see appendix 4 C. Reigber 





References and Bibliography Bar-Sever Y.E., P.M. Kroger, and J.A. Borjesson, Estimating horizontal gradients of tropospheric path delay with a single GPS receiver, J. Geophys. Res., 103, pp 5019-5035, 1998.

Bevis, M., Businger, S., Herring, T.A., Rocken, C., Anthes, A., and Ware, R., 1992, GPS Meteorology: Remote sensing of atmospheric water vapor using the Global Positioning System, Journal of Geophysical Research, 97, 15,787-15,801.

Bevis, M., Businger, S., Chiswell, S., Herring, T.A., Anthes, R., Rocken, C., and Ware, R, 1994, GPS Meteorology: Mapping zenith wet delays onto precipitable water, Journal of Applied Meteorology, 33, 379-386.

Businger, S., S.R. Chiswell, M. Bevis, J. Duan, R. Anthes, C. Rocken, R. Ware, T. M. Exner, VanHove, F. Solheim, 1996, The promise of GPS in atmospheric monitoring, Bulletin of the American Meteorological Society, 77, 5-18.

Carlsson, T.R., G. Elgered, and J.M. Johansson, A Quality Assessment of the Wet Path Delay Estimated From GPS Data, Proc. of the 11th Working Meeting on European VLBI for Geodesy and Astrometry, Research Report 177, Ed. G. Elgered, Onsala Space Observatory, Chalmers University of Technology, 89-95, 1996.

Coster, A.J., A.E. Niell, F.S. Solheim, V.B. Mendes, P.C. Toor, R.B. Langley, and C.A. Upham (1998). "The Westford Water Vapor Experiment: Accuracy issues involving the use of GPS to measure total precipitable water vapor." Proceedings of the 10th Symposium on Meteorological Observations and Instrumentation, 78th Annual Meeting of the American Meteorological Society, Phoenix, AZ, 11-16 January, pp. J70-J75.

Coster, A.J., A.E. Niell, F.S. Solheim,V.B. Mendes, P.C. Toor, K.P. Buchmann, and C.A. Upham (1996). "Measurements of precipitable water vapor by GPS, radiosondes, and a microwave water vapor radiometer." Proceedings of the ION GPS-96, the 9th International Technical Meeting of the Satellite Division of the Institute of Navigation, Kansas City, MO, 17-20 September, pp. 625-634.

Coster, A.J., A.E. Niell, F.S. Solheim, V.B. Mendes, P.C. Toor, and R.B. Langley (1997). "The Effect of Gradients in the GPS Estimation of Tropospheric Water Vapor." Proceedings of The Institute of Navigation 53rd Annual Meeting, Albuquerque, NM, U.S.A., 30 June - 1 July; pp. 107-114.

Davis, J.L. and G. Elgered, The Spatio-Temporal Structure of GPS Water-Vapor Determinations, Physics and Chemistry of the Earth, 23, 91-96, 1998.

Davis, J.L., M.L. Cosmo, and G. Elgered, Using the Global Positioning System to Study the Atmosphere of the Earth: Overview and Prospects, in GPS Trends in Precise Terrestrial, Airborne, and Spaceborne Applications, Eds. G. Beutler, G.W. Hein, W.G. Melbourne, G. Seeber, IAG Symposia, Vol. 115, Springer Verlag, Berlin, pp. 233-242, 1996.

Dodson A H; Baker H C (1998) ; Accuracy of Orbits for GPS Atmospheric Water Vapour Estimation, Physics and Chemistry of the Earth, Vol 23, No 1, pp 119 - 124, April 1998.

Duan, J. , M. Bevis, P. Fang, Y. Bock, S. Chiswell, S. Businger, C. Rocken, F. Solheim, T. Van Hove, R. Ware, S. Mcclusky, T.Herring, R. W. King, 1996, GPS Meteorology: Direct Estimation of the Absolute Value of Precipitable Water, Journal of Applied Meteorology, 35, 830-838.

Elgered, G., J.M. Johansson, and J.L. Davis, Using Microwave Radiometry and Space Geodetic Systems for Studies of Atmospheric Water-Vapor Variations, in Microwave Radiometry and Remote Sensing of the Environment, ed. D. Solimini, VSP Int. Sci. Publ., Zeist, Netherlands, pp. 69-78, 1995.

Elgered, G., J.M. Johansson, B.O. Ronnang, and J.L. Davis, Measuring regional atmospheric water vapor using the Swedish permanent GPS network, Geophys. Res. Let., 24, 2663-2666, 1997.

Emardson, T.R., G. Elgered, and J.M. Johansson, Three Months of Continuous Monitoring of Atmospheric Water Vapor with a Network of Global Positioning System Receivers, J. Geophys. Res., 103(D2), 1807-1820, 1998.

Emardson, T.R., J.M. Johansson, and G. Elgered, The systematic behavior of water vapor estimates using four years of GPS observations, Trans. IEEE Geoscience and Remote Sensing, accepted, 1999.

Fang, P., M. Bevis, Y. Bock, S. Gutman and D. Wolfe, 1998, GPS meteorology: Reducing systematic errors in geodetic estimates for zenith delay, Geophysical Research Letters, 25, 3583-3586.

Gendt G, G Beutler: Consistency in the Troposphere Estimations Using the IGS Network, Proceedings IGS Workshop "Special Topics and New Directions", Potsdam, May 15-18, 1995, GeoForschungsZentrum Potsdam, Germany, Eds. G Gendt and G Dick, pp.115-127

Gendt, G., Comparison of IGS Troposphere Estimations Proceedings IGS Analysis Centers Workshop, 19-21 Maerz, 1996 Silver Spring, Maryland USA , Eds. R E Neilan, P A Van Scoy, J F Zumberge. pp. 151-164

Gendt, G., IGS Combination of Tropospheric Estimates -- The Pilot Experiment. 1997 Technical Reports, Oct. 1998, Ed. I Mueler, K Gowey, R Neilan, pp. 265-269

Gendt, G., IGS Combination of torposphereic Estimates - Experience from Pilot Experiment. Proceedings IGS AC Workshop, 9-11 Febr. 1998, Darmstadt Eds. J Dow, J Kouba, T Springer, pp.205-216

Gendt, G., M. Bevis: Some remarks on new and existing tropospheric products. Proceedings IGS Network Systems Workshop, Annapolis, MD., November 2-5, 1998Eds. J Dow, J Kouba, T Springer, pp.205-216, , in press.

Gossard, E.G., S. I. Gutman, B.B. Stankov, and D.E. Wolfe,1999. Profiles of radio refractive index and humidity derived from radar wind profilers and the Global Positioning System. Radio Sci.,34, 371-383.

Gutman, S., D. Wolfe, A. Simon, 1996: Status of the GPS Precipitable Water Vapor Observing System. FSL Forum, Dec. 1996, 29-33.

Haines, B.J., and Y.E. Bar-Sever, Monitoring the TOPEX microwave radiometer with GPS: Stability of columnar water vapor measurements, Geophys. Res. Lett., Vol. 25, No. 19, pp 3563-3566, 1998.

Ichikawa, R., M. Kasahara, N. Mannoji, and I. Naito, Estimations of atmospheric excess path delay based on three-dimensional, numerical prediction model Data, J. Geod. Soc. Japan, 41, 379-408, 1995.

Ichikawa, R., M. Kasahara, N. Mannoji, and I. Naito, Positioning Error in GPS Measurements due to Atmospheric Excess Path Delay Estimated From Three-Dimensional, Numerical Prediction Model Data, J. Geod. Soc. Japan, 42, 183-204, 1996.

Johansson, J.M., T.R. Emardson, P.O.J. Jarlemark, L. Gradinarski, and G. Elgered, The Atmospheric Influence on the Results from the Swedish GPS Network, Physics and Chemistry of the Earth, 23, 107-112, 1998.

Kimata, F., N. Kato, and T. Sugiyama, GPS measurements errors for Nagoya-Atsumi baseline at the passing of cold front on November 7, 1995, J. Geod. Soc. Japan, Vol. 42, 119-122, 1996 (in Japanese).

Kruse, L.P., B. Sierk, T. Springer and M. Cocard (1999), GPS-Meteorology: Impact of Predicted Orbits on Precipitable Water Estimates, Geophys. Res. Lett., 26, No. 14, in press.

Mannoji, N., GPS meteorology, Journal of the Visualization Soc. of Japan, Vol.16, 107-111, 1996(in Japanese).

Mousa, A. and T. Tanaka, Tropospheric wet delay of microwaves at Shionomisaki, southwest Japan, and a preliminary evaluation of mapping functions, J. Geod. Soc. Japan, 43, 145-158, 1997.

Naito, I., GPS meteorology, J. Japan Soc. Hydrol. and Water Resourc. Vol. 9, 570-578, 1996(in Japanese).

Naito, I., Y. Hatanaka, N. Mannoji, R. Ichikawa, S. Shimada, T. Yabuki, H. Tsuji, and T. Tanaka, Global Positioning System Project to Improve Japanese Weather, Earthquake Predictions, EOS, Transactions, American Geophysical Union, 79, 301, 308, 311, June, 1998.

Ohtani, R., H.Tsuji, N.Mannoji, J.Segawa and I.Naito: Precipitable water vapor observed by Geographical Survey Institute's GPS network, TENKI, 44, 317-325, 1997(in Japanese).

Rocken, C., R. Ware, T. Van Hove, F. Solheim, C. Alber, J. Johnson, M. Bevis, S. Businger, 1993, Sensing atmospheric water vapor with the Global Positioning System, Geophysical Research Letters, 20, 2631-2634.

Rocken, C., T. Van Hove, J. Johnson, F. Solheim , R. Ware, M. Bevis, S. Chiswell, S. Businger, 1995, GPS/Storm - GPS Sensing of Atmospheric Water Vapor for Meteorology, Journal of Atmospheric and Oceanic Technology, 12, 468 - 478.

Ruffini, G., L.P., A. Rius, B. Burki, L. Cucurull, A. Flores, Estimation of Tropospheric Zenith Delay and Gradients over the Madrid Area Using GPS and WVR Data, Geophysical Research Letters, in press.

Sierk, B., Bürki, B., Becker-Ross, H., Florek, S., Neubert, R., Kruse, L.P., and H.-G. Kahle (1997): Tropospheric water vapor derived from solar spectrometer, radiometer and GPS measurements, Journal of Geophysical Research, 102, No. B10, pp. 22411-22424.

Sierk, B., Bürki, B., Kruse, L.P., Becker-Ross, H., Florek, S. Kahle, H.-G. and R. Neubert (1998): A new instrumental approach for water vapor determination based on Solar Spectrometry. Phys. Chem. Earth, 32, No. 1, pp. 113-117.

Steinhagen,H, S Bakan, J Boesenberg H Dier D Engelbart J Fischer G Gendt et al., Field campaign LINEX 96/1 - Possibilities of water vapor observation in the free atmosphere, Meteorol. Zeitschrift, N.F.7, 377-391.

Tsuda T., K. Heki, S. Miyazaki, K. Aonashi, K. Hirahara, H. Nakamura, M. Tobita, F. Kimura, T. Tabei, T. Matsushima, F., Kimura, M. Satomura, T. Kato, and I. Naito, GPS Meteorology Project of Japan -Exploring Frontiers of Geodesy-, Earth Planet Space, 50, i-v, October, 1998.

Yuan, L., Anthes, R., Ware, R., Rocken, C., Bonner, W., Bevis, M., and Businger, S., 1993, Sensing global climate change using the Global Positioning System, Journal of Geophysical Research, 98, 14,925-14,937.

Yang, X., B. Sass, G. Elgered, T.R. Emardson, and J.M. Johansson, A comparison of the integrated water vapor estimation by a NWP simulation and GPS observation, J. Appl. Meteorol., 38, pp. 941-956, 1999.

Westwater, E.R., Y. Han, S.I. Gutman, and D.E. Wolfe, 1998. Remote sensing of total precipitable water vapor by microwave radiometers and GPS during the 1997 Water Vapor Intensive Operating Period. 1998 IEEE International Geoscience and Remote Sensing Symposium Proceedings: Vol. IV, 2158-2161.

Wolfe, D.E. and S.I. Gutman, Developing an operational surface-based GPS water vapor observing system for NOAA: network design and results. J. Atmos. Oceanic Technol., in press.

Zuffada, C., G. Hajj and E.R. Kursinski, A novel approach to atmospheric profiling with a mountain-based or air-borne GPS receiver, to appear in JGR atmosphere.




by Gerd Gendt, GFZ (gendt@gfz-potsdam.de)

GeoForschungsZentrum Potsdam, Division 1, Telegrafenberg A17, D-14473 Potsdam, Germany; e-mail: gendt@gfz-potsdam.de

The existing global IGS network of permanent GPS receivers installed for geodetic and geophysical applications can be used with marginal additional efforts for the determination of atmospheric water vapor.

The IGS has produced a tropospheric product on a regular basis since the beginning of 1997. After the successful performance of the Pilot Experiment the tropospheric estimates became an official IGS product in 1998. All IGS Analysis Centers submit their individual solutions to the GeoForschungsZentrum Potsdam (GFZ) where the official combined product is generated. It is a weighted least squares estimate for the zenith neutral delay (ZND) at selected IGS stations with a sampling rate of 2 hours, and is available with a delay of 3 to 4 weeks. The product meets the demands for climate studies where for the interesting long-term characteristics a resolution of 2 hours is sufficient, and a delay of a few weeks is acceptable.

Estimates for more than 80 sites are provided by 3 or more Analysis Centers. For these sites reasonable estimates of internal consistency can be obtained. The standard deviation for most of the sites is in the level of 2-5 mm ZND (corresponding to better than 1 mm in the precipitable water vapor (PWV)) and the scattering of the bias from site to site is about 3 mm ZND. For sites in the equatorial region, where partly severe problems with the higher ionospheric activities occur, the scattering is much higher but in most cases below the 2 mm level in PWV. Site solutions delivered by only one Analysis Center are also contained in the product, sampled and transformed into the troposphere format. Most of those sites are located in denser parts of the network, where all sites have nearly the same accuracy and therefore the quality can be deduced from neighboring sites.

In addition to the ZND product RINEX meteorological files are offered for conversion into PWV. Despite the fact that IGS has encouraged its members to add suitable met packages to the IGS tracking stations, very little progress has been made during the last two years. At the moment only 30 sites are equipped with met packages. Unfortunately, for some sites data quality is not good enough. The precision of the sensors is not usually the problem, however. Rather, there are too many meteorological data gaps (often days long), and in such cases no meaningful series of water vapor may be compiled. To support the decision as to where future met packages should be installed, IGS will maintain a list of high priority candidate sites. Criterions for the selection may be the quality of the tropospheric estimates and the location of the sites (e.g. equatorial regions may be especially interesting).

The number of projects and activities involving near real-time monitoring of water vapor using ground-based GPS is steadily increasing. IGS will not be involved in such near real-time activities directly. However, IGS can support regional activities of this kind by making available hourly RINEX data within the global tracking network and by generating predicted orbits. The presently available predictions based on daily data batches have to be predicted over 48 hours and are for a number of satellites often not in the quality needed. Based on the hourly downloads IGS will be able to generate predictions more frequently and the shortened prediction interval will lead to significant improvements. During 1999 within the IGS the development into this new direction will be discussed and technologies will be developed.


Present Status of GPS Meteorology Project of Japan

The GPS meteorology project of Japan is a five-year project launched April, 1997 funded by Science and Technology Agency (STA) of Japan. The project is basically motivated by the nationwide GPS array of GSI (Geographical Survey Institute of Japan), composed of nearly one thousand stations separated by 15-30 km, and has following three major goals. The first goal is to create a system to let the retrieved precipitatble water vapor (PWV) data from GSI’s GPS array flow to the four dimensional data assimilation (4DDA) system for numerical weather prediction in JMA (Japan Meteorological Agency). This improve the forecast systems of meso- to local-scale phenomena that often trigger torrential rains in humid Japan. The second is to improve accuracy in crustal deformation measurements in the nationwide GPS array using a GPS meteorology, in which three dimensional meteorological data of JMA’s 4DDA system enhanced by PWV data from GPS are applied to establish a system diagnosing or removing the errors in estimated site coordinates in vertical due to water vapor. The third is to construct a database of nationwide GPS PWV information for uses in interdisciplinary environmental studies, hydrology, meteorology, and geodesy (see Naito et al., 1998 and Tsuda et al., 1998, for more details). Below are the products achieved through the project so far.

PWVs routinely retrieved from GSI’s GPS array agreed well with those from radiosonde observations by JMA with RMS differences of about 3mm, but showed small biases caused mainly by vertical displacements due to ocean tidal loading. The GPS array also detected dense temporal anomalies of PWVs during heavy rainfalls over the Japan Islands that can be applicable for 4DDA system, though there have been found to exist differences between PWVs obtained from GPS made on real topography and those in 4DDA system based on model topography. Estimated horizontal coordinates in the GSI’s GPS array were found to strongly reflect horizontal gradients in zenith tropospheric delay (ZTD). This was confirmed by improvements in horizontal coordinates made by GPS analysis with horizontal gradient in ZTD. GSI’s GPS array detected a local circulation system associated with land and sea breezes in Kanto area through GPS PWVs. GPS’s PWVs were found to be a useful tool for predicting torrential rain and fog related to the local circulation system. Two dimensional tomography experiments using a dedicated dense GPS array detected vertical profiles of atmospheric refractive index due to water vapor. The tomography coupled with a model simulation was found to be a useful tool for detecting dry and wet regions in local torrential rainfalls.

Moreover, GPS measurements in Thailand under GAME (GEWEX Asian Monsoon Experiment) proved GPS’s ability as water vapor sensor in comparison with radiosonde observation. GPS measurements also played a key role in detecting onsets of the monsoon in Tibetan plateau. A global distribution of potential energy due to atmospheric waves at altitudes of 20-30 km were obtained from the space-based GPS meteorology data of UCAR (University Corporation for Atmospheric Research), which provides valuable information in the tropical middle atmospheric dynamics, for such as the Quasi-Biennial Oscillation (QBO), for example.

References: Naito, I., et al., 1997, Global Positioning System Project to Improve Japanese Weather, Earthquake Prediction, Eos, 79, 301, 308, 311.

Tsuda, T., et al., 1997, GPS meteorology project of Japan --- Exploring frontiers of geodesy ---, Earth Planet Space, 50, 1-5.


Report on Project WAVEFRONT and the new European COST Action

By Helen Baker

WAVEFRONT (GPS WAter Vapour Experiment For Regional Operational Network Trials) is a three year collaborative project funded by the European Commission (EC). The project is coordinated at IESSG (University of Nottingham, UK) in collaboration with Onsala Space Observatory (Chalmers University, Sweden), Eidgenhossische Technische Hochschule (Switzerland) and Institut d'Estudis Espacials de Catalunya (Spain) in association with the UK Meteorological Office, the Astronomical Institute at the University of Berne (Switzerland) and the Danish Meteorological Institute.

WAVEFRONT started in September 1996 under the EC Environment and Climate work programme, aiming to develop the potential for using ground-based GPS to estimate the variable water vapour content of the atmosphere, ultimately for meteorological and climatological applications. More specifically WAVEFRONT aims to fulfil the following objectives:

•· Validate the accuracy and temporal resolution of GPS meteorology, by comparison with ground-based water vapour radiometers (WVR)and radiosonde profiles

•· Assess the feasibility of near real-time (less than three hour latency) estimation of GPS IWV to an accuracy of 1-2 kg/m2

•· Compute a post-processed data set of GPS IWV estimates from a European sub-set of established permanent GPS monitoring stations (part of the International GPS Service) for climate modelling studies

•· Investigate the possibility of obtaining a more detailed tomographic profile of water vapour over a smaller, denser GPS network to develop a data set to study micro-scale synoptic processes

The validation of GPS estimates against WVRs and radiosondes has been undertaken through a number of field campaigns and an extensive examination of the associated error sources within the GPS processing procedure, a-priori statistics, conversion constants applied and network processing strategies. Using findings from these analyses, a comparison of tropospheric estimates from three GPS processing software packages used within the WAVEFRONT project (BERNESE, GAS and GIPSY) has been undertaken using two climatically different three-month data sets. A feasibility study for estimating GPS water vapour on a near real-time basis has been completed in preparation for meteorological forecasting impact assessments at the UK Meteorological Office. The archiving of one-hour averages of atmospheric water vapour is also being performed at approximately forty European IGS stations, ultimately to obtain a continuous water vapour time series for climate studies. A dense network of GPS receivers has also been used over short campaign periods to assess the feasibility of tomographic analysis to examine three dimensional water vapour distribution.

Initial results from WAVEFRONT have indicated agreement between GPS, WVR and radiosonde water vapour estimates at the 1-1.5 kg/m2 level using a recommended processing strategy developed from the analysis of processing procedures and error source examinations. European algorithms have been derived to improve the accuracy of the conversion from GPS atmospheric delay estimates to IWV. In addition to the European GPS tropospheric archive collected for climate study purposes, further data sets have been used in a 'sliding window' processing approach with predicted orbit data, to demonstrate the potential for near real-time water vapour estimation at a corresponding level of accuracy (1-2 kg/m2). Preliminary results from the tomographic examination have produced vertical profiles which compare well with ECMWF profiles.



EC COST ACTION 716: Exploitation of ground-based GPS for climate and numerical weather prediction applications

The memorandum of understanding for COST Action 716, entered into force on 16 September 1998. The Action has as its primary objective:

"Assessment of the operational potential on an international scale of the exploitation of a ground based GPS system to provide near real-time observations for Numerical Weather Prediction and Climate


With secondary objectives of:

•· Development and demonstration of a prototype ground-based GPS System

on an international scale

•· Validation and performance verification of the prototype system

•· Development and demonstration of a data exploitation scheme for NWP and analysis of data exploitation techniques needed for climatic applications

•· Requirements for operational implementation of a ground-based GPS system on an international scale.

The 4-year Action at present involves 14 countries (Austria, Belgium, Denmark, Finland, France, Germany, Hungary, Italy, The Netherlands, Norway, Spain, Sweden, Switzerland and the UK), and the chairman is Gunnar Elgered from Chalmers University in Gothenburg, Sweden.

Initial work has concentrated on the formulation of four 'work packages' addressing the above objectives.


GPS Atmosphere Sounding Project : An Innovative Approach for the Recovery of Atmospheric Parameters

By C. Reigber, GFZ


Reliable information on global climate change processes over future decades, and the perennial need for better weather forecasting on near and medium-term time scales, are only possible on the basis of global and regional data records and modeling that accurately represent atmospheric state parameters with high spatial and temporal resolution. Remote sensing satellite systems, such as multispectral passive radiometers, and interdisciplinary research programs, have already contributed significantly to a better understanding of the climate system. However, coverage for important data such as tropospheric water vapor is sparse and complementary observation systems are needed to reduce this lack of information. The Global Positioning System (GPS) is presently developing into a powerful candidate for such a system. With the rapid development and operation of permanent global and dense regional GPS ground station networks and associated data communications and information systems, and also in view of the rapidly expanding number of Low Earth Orbiting (LEO) satellites carrying GPS or GPS-related instrumentation for limb sounding measurements, there are excellent long-term prospects for very valuable contributions of GPS to operational meteorology and for a permanent, seamless, weather-independent and calibration-free (because of pure time difference measurements) monitoring of important parameters of state of the atmosphere and ionosphere. A basic requirement for this is that the necessary infrastructure components and methodologies are developed, tested, validated and are embedded into an operational environment. With this project the HGF centers GFZ, DLR, GKSS and AWI initiate the development into an expanded infrastructure and the performance of programs to effectively assimilate GPS data products into atmospheric and ionospheric modeling and analyses. This is done to ensure that existing GPS expertise of HGF centers is focused towards an innovative approach, the potential of which is considered high for climate research and applications in the science community, and to ensure that proper use is made of national investments into the forthcoming satellite missions CHAMP and GRACE.

Project goals

The main objective of the proposed project is: The development and/or improvement of largely already existing infrastructure (networks of GPS stations, orbiting GPS receivers, communication links, network and mission operation centers, data information and archival centers, analysis centers and related S/W systems, user interfaces) required for the use of ground- and space-borne GPS tracking data for atmospheric and ionospheric sciences and applications. This should be realized in such a short time and to such an extent, that already atmospheric/ionospheric data products from the forthcoming CHAMP and GRACE satellite missions and from densified regional geodetic GPS networks in Germany can be explored for their potential information content and operational availability in applications such as meteorological weather forecasting, climate variability investigations and space weather monitoring.


To achieve this main objective the project activities are combined into five major project goals:

- Acquisition of GPS data in near-real time from regional networks and from the global network of the International GPS Service IGS, development and implementation of necessary techniques and methods for quasi-operational determination of tropospheric water vapor concentrations and electron densities, as well as the development of optimal strategies with the cooperating partners for meteorological practice and research.

- Development of strategies and methods for the determination of atmospheric and ionospheric parameters in near-real time from limb sounding data of the satellites to be launched within the next three years including the forthcoming missions CHAMP and GRACE and starting with already existing GPS/MET data; systematic validations over a lengthy time period, and assessments of the efficiency of the technique for weather forecasting and for climate research.

- Development of assimilation techniques for ingesting limb sounding results and ground-data derived water vapor fields into existing global and regional models for both climate study and numerical weather prediction and systematic evaluations of the assimilation results.

- Extensive comparisons of GPS-based results for atmospheric parameters with correlative data sets from existing sensor systems (e.g., radiosondes, ground-/satellite-based radiometers) and models such as those from the European Center for Medium Weather Forecast (ECMWF), the Max Planck Institute (MPI) for Meteorology, Hamburg, the GKSS, and the German Weather Service (DWD) to evaluate the GPS method and to assess its effectiveness.

- Design of a data processing, archiving and distribution system and its partial development on component level on the basis of the experiences gained in the project for optimally serving users in future programs.

Project Work Breakdown

The work to be carried out in the proposed project is subdivided into the two subprojects:

Subproject 1 (SP1): Water Vapor Estimation from Ground GPS Networks and Assimilation

into Atmospheric Models

Subproject 2 (SP2): Radio Limb Sounding with Spaceborne GPS

Subproject 1 is led by GFZ with contributions from AWI and GKSS. The total work is broken down into 13 work packages. Subproject 2 is led by DLR with contributions from GFZ and AWI. SP2 is split into 20 work packages. Through these HGF center activities primarily ?Level 2„ data products will be generated, such as orbit ephemerides, profiles of refractivity, temperature and pressure as well as Total Electron Content (TEC) records for the ionosphere. ?Level 3„ products pertain primarily to those products derived from cooperative activities involving HGF centers and external partner institutions engaged in climate research or weather analyses, such as the German Weather Service (DWD) and the MPI for Meteorology. This will include, for example, recovery of tropospheric water vapor fields and wind fields, model development and data assimilation, statistical studies of model performance, validation activities, temporal and spatial averages of refractivity, temperature and geopotential height profiles, etc.

Project Team

The project team will be composed of scientists from the cooperating HGF centers AWI, DLR, GFZ, GKSS. Scientists from the external partner institutions given in the sequel will support the project. Prof. Christoph Reigber, director of GFZ‚s Division 1 and presently project lead of the small satellite mission CHAMP, Co-PI of the GRACE mission and chairman of the Governing Board of the International GPS Service, will lead the team.

The HGF center institutes and the lead scientists from these institutes contributing to the project with level 1 to level 2 products are:

• GFZ, Division 1 Kinematics and Dynamics of the Earth, Potsdam and Oberpfaffenhofen: Dr. G. Gendt (Manager SP1)

• DLR, Remote Sensing Data Center, Neustrelitz: Dr. N. Jakowski (Manager SP2)

• GKSS, Institute for Atmospheric Physics, Geesthacht: Prof. Dr. E. Raschke

• AWI, Section ?Physics of the Atmosphere and Oceans I„, Bremerhaven and Potsdam: Dr. R. Neuber

• External partner institutions, the involved lead scientists and their anticipated contribution to the project are:

• Jet Propulsion Laboratory (JPL), Pasadena (Dr. W.G. Melbourne):

GPS Flight receiver S/W; Radio-occultation measurement resolution; validation/proof-of concept programs

• Institute of Radio Engineering and Electronics (IRE), Moscow (Prof. O. Yakovlev):

Improvements of inversion S/W under low S/N or enhanced multipath conditions

• Max Planck Institut (MPI):

- fuer Meteorologie, Hamburg (Dr. L. Kornblueh):

Assimilation methods, retrieval techniques - fuer Aeronomie, Lindau (Prof. K. Schlegel):

Ionospheric electrodynamics, space weather aspects

• Deutscher Wetterdienst (DWD):

- Offenbach (Dr. W. Wergen):

GPS data products in operational numerical weather forecasting - Observatory Lindenberg (Dr. H. Steinhagen):

Validation of GPS limb sounding measurements

• Freie Universitaet Berlin:

- Institut fuer Weltraumwissenschaften (Prof. J. Fischer):

Validation of ENVISAT MERIS water vapor with CHAMP/GRACE retrievals - Institut fuer Meteorologie (Prof. K. Labitzke): Validation of radio occultation data with model data, inspection of time series

• Universitaet Koeln, Institut für Meteorologie (Prof. P. Speth):

Study of Madden-Julian Oscillation with CHAMP/GRACE water vapor

Project Schedule

Project start is planned for July 1999. In the first phase of the project, which will last until the end of the CHAMP mission commissioning phase in about Spring to Summer 2000, the major activities will take place in Subproject 1 - network enhancement, water vapor recovery and data assimilation from ground GPS data. In Subproject 2 preparatory activities for the usage of radio occultation measurements from CHAMP will take place in this period. Operational activities with radio limb sounding data from CHAMP will start in Summer 2000. They will last until the end of the project, which is in July 2002. In the middle of this second phase the GRACE twin satellite mission will be launched, which will allow to test the performance of the infrastructure for a multi-satellite system being developed in the project.



Summary of the MAGIC Project

by Jennifer Haase

Meteorological Applications of GPS Integrated Column Water Vapor Measurements in the Western Mediterranean

Keywords: water vapor, weather prediction, flooding, and climate Thematic areas: Flood and weather prediction, climate

Background and Rationale:

Humidity is a highly variable parameter in atmospheric processes and it plays a crucial role in atmospheric motions on a wide range of scales in space and time. Limitations in humidity observation accuracy, as well as temporal and spatial coverage, often lead to problems in numerical weather prediction, in particular that of clouds and precipitation. Due to these limitations, the verification of humidity simulations in operational weather forecasts and climate modeling are also difficult. At smaller scales, catastrophic rainfall events due to storm systems unique to the western Mediterranean are difficult to predict with the operational (larger scale) numerical weather prediction models. Such storm events are often associated with flash floods with loss of property, and in some cases, human casualties. While local models have some success in predicting precipitation, key humidity data are required to verify the derived humidity fields. The emerging ground-based Global Positioning System (GPS) networks present appealing opportunities for an improved humidity observation source that can help resolve these difficulties.

In brief:

MAGIC will examine the need for improved water vapor estimates by the Danish Meteorological Institute and the Servei Meteorologia de Catalunya and evaluate the ability of GPS ground based networks to address this need. Automated data processing for retrieval of zenith tropospheric delay data from the GPS data will be implemented. The data will be validated against integrated values from conventional data sources such as radiosondes, and against the DMI HIRLAM model forecasts. Assimilation algorithms will be developed by DMI and the IEEC in Spain to integrate this new type of data into mesoscale numerical weather prediction models to evaluate the impact of the data and evaluate improvements in prediction capabilities, especially in areas at high risk for storms such as Catalonia. The project extends beyond the derivation of the required data from earth observation sources, to the actual development of the technology necessary for the user to fully exploit the data.

Recent results:

Compiling a data set from different processing centers for NWP model validation or assimilation tests requires validation of the consistency of the GPS processing methods and quality control of the output GPS IWV data. One objective of the MAGIC project is to demonstrate that this level of consistency can be achieved in direct response to the operational requirements of the meteorological community. The Danish Meteorological Institute (DMI) is the partner in the project charged with evaluating the impact of the GPS IWV data in the HIRLAM NWP model. DMI has given a target value of 0.5 cm in ZTD as an acceptable error level. A methodology has been adopted for processing the data from all permanent GPS stations in France, Spain, and Italy, in addition to data available from the IGS. In preliminary tests comparing the operational results of two of the MAGIC processing centers (CNRS, IEEC), the difference between the ZTD estimates satisfies these requirements.

MAGIC serves:

•Operational meteorological agencies

•Emergency services concerned with catastrophic flooding The European public which will benefit from better forecast models Climate modeling community

•International organisations studying climate change

Specific end users:

•Danish Meteorological Agency (Consortium partner) Servei Meteorologia de Catalunya

•Generalitat de Catalunya (Catalonian Regional Government) other potential customers:

•Other meteorological agencies of the HIRLAM consortium (including the Instituto Nacional de Meteorologia (Spain), the Italian Meteorological Service, and Meteofrance)

•Other regional weather prediction agencies and emergency services entities (Civil Protection of the Government of Andalucia, Spain). International organisations studying climate change



ACRI - Mécanique Appliquée et Sciences de l'Environnement

CNRS: Centre National de la Recherche Scientifique/ Géoscience Azur ES

ICC: Institut Cartografic de Catalunya ES

IEEC: Institut d'Estudis Espacial de Catalunya

ASI: Agenzia Spaziale Italiana, Centro di Geodesia Spaziale

OGMO: Osservatorio Geofisico dell'Universita di Modena

DMI: Danish Meteorological Institute

ROA: Real Instituto y Observatorio de la Armada en San Fernando

Coordinator: ACRI - Mécanique Appliquée et Sciences de l'Environnement 260, Route du Pin Montard, BP. 234, 06904 Sophia Antipolis Cedex, France

Responsable: Jennifer Haase

e-mail: jh@acri.fr


EUREF Analysis Center for GPS Meteorology

by Jan Dousa, Czech Technical University

GOP analys center (AC) takes advantage of hourly data uploads in the EUREF (EUropean REference Frame) permanent GPS network. In the beginning of 1999 it started to operate (for testing purposes only) as a ground-based GPS meteorology analysis center (approx. 20 sites). Although the final processing strategy is still under the development AC produces the results routinely. Research aims are focused on near-real-time (NRT) GPS data processing in accordance with the strategy for precise zenit total delay (ZTD)/precipitable water vapor (PWV) estimation. The emphasis is currently given for the main error source in NRT ZTD estimation: prediction of precise orbits.

Three main types of processing are set up for test purposes:

1. NRT 1h-proc. 24x/day (delay: 1-6 hours), predicted orbits IGP/COD_P2

2. NRT 4h-proc. 24x/day (delay: 1-6 hours), predicted orbits IGP/COD_P2

3. Post 24h-proc. 1x/day (delay: 20-48 hours), rapid orbits IGR/COD_R

- Post-processing solution should be considered as reference solution and archive solution for climatology studies. Two months of comparision (May+June 99') of routinely resulted PWV with radiosonde observations (4x/day) gives RMS 1.5mm and bias -1.0mm. The solution will be father improved and checked especially for the bias. - NRT-processing solutions are considered in future for meteorology operational forecasting.The comparision of 4h-proc (3rd hour) gives currently RMS lower 3.5mm and bias aprox. -1.0mm. (The ambiguity fixing is rather problematic in this solution !)

To achieve a good accuracy of NRT PWV results, it was already proved the selection of satellites for the final processing is necessary (the prediction of only few satellites is realy bad !). Different satellite exclusion (or de-weightening) procedures were tested while the stability of results has increased by 2-4 times in NRT mode. Currently the satellite exclusion procedure is based on checking the consistency between two following up predicted daily orbit arcs and does not take into account predicted accuracy code for each satellite from the header of SP3 file. Preliminary results of solutions with partial orbit improvements gives us hope to achieve the accuracy of 1-2mm RMS in NRT processing. This variant configuration is currently under the development.


GOP LAC - EUREF local analysis center Geodetic Observatory Pecny

established in collaboration between

- Research Institute of Geodesy, Topography and Cartography

- Dep. of Advanced Geodesy, Czech Technical University in Prague



Report from New Zealnad: Estimating Atmospheric Water Vapour Content using GPS

Mark Falvey

Victoria University of Wellington (VUW), Research School of Earth Sciences, P O Box 600, Wellington, NZ

John Beavan

Institute of Geological and Nuclear Sciences (GNS), Gracefield Research Centre, P O Box 30368, Lower Hutt, NZ

Research in New Zealand has so far focused in two areas: using the country’s nascent (5-station) continuous GPS network to provide an operational water vapour monitoring system, and targeted experiments aimed at studying extreme rainfall events across mountain belts.

At present, water vapour observations are restricted to surface humidity measurements and upper air radiosonde (weather balloon) observations. Surface observations are often not representative of the general state of the atmosphere, and radiosondes are costly and are released only once or twice a day at only five stations in the New Zealand region. Satellite based radiometer images in water vapour channels (NOAA, GMS) may also be received but these can be corrupted by the presence of clouds and are difficult to interpret quantitatively. GPS precipitable water measurements should provide valuable data to supplement these existing observational platforms. The hourly time resolution makes it a useful tool in those analysis situations where water vapour changes rapidly with time. Indeed GPS derived PW time series reveal a great deal of high frequency structure. We have also found the GPS PW to be very accurate. Measurements of PW made at radiosonde stations have compared to within 2 mm of those at nearby GPS sites.

For the PW estimates to be of most use, they must be available on a near real-time basis. In real-time processing predicted GPS satellite orbit data must be used. Such orbits are prone to large errors which can cause spurious features in the resulting PW time series. Much of our work so far has been to test and implement processing techniques capable of identifying and removing poor satellite data. We believe that this is now at a sufficiently refined state that we can soon begin to operate in an automated real-time test mode.

Along with the continuous station work, investigation has begun into applying GPS measurements to the study of terrain induced rainfall, using data collected during two meteorological field experiments. This is an application of GPS precipitable water which we believe is unique to our research group. It is well known that moist airflow over New Zealand's mountainous regions can result in considerable precipitation enhancement (orographic rainfall). However, quantitative models capable of predicting this enhancement and associated river flow levels have yet to be fully developed.

There have so far been two field campaigns during which GPS receivers were deployed to make precipitable water measurements. They were SALPEX'96 (Southern ALPs EXperiment) and TARPEX'99 (TARaruas Precipitation EXperiment). Both were collaborative projects involving several national and international research institutions including the National Institute for Water and Atmospheric Research (NIWA), Victoria and Auckland Universities, and GNS. A variety of instruments were deployed (radar, rain gauge, radiosonde, specially equipped aircraft) along with dedicated networks of GPS receivers. During SALPEX, eight receivers were arranged in a transect stretching from Hokitika to Christchurch while in TARPEX, twelve receivers were deployed between the Kapiti coast and the Wairarapa. Both experiments lasted for roughly three weeks and between them a total of five interesting rainfall events have been observed.

The data from both experiments are being used to investigate the ability of the GPS technique to characterise water vapour variations over small horizontal scales. In particular we investigate the significant gradients in PW across the ranges observed during orographic rainfall events. We attribute much of this variation to condensation of water vapour as air is forced upwards over the mountains, and are attempting to relate this apparent condensation signal to observed precipitation patterns. The GPS measurements are also used to validate regional atmospheric model simulations of the SALPEX and TARPEX events.

As a more complete continuous GPS network is developed in New Zealand, we envisage the network being able to contribute water vapour data to real-time weather forecast models – as well as providing ground deformation data for earthquake and volcanic hazard monitoring.