Retrieval of total columnar precipitable water vapour using radio occultation technique over the Indian region
Introduction
Currently, various constellations of navigation satellites are operational, mainly used for precise positioning. Apart from navigation, the signals obtained from them are also used for other atmospheric and ground-based applications. Global Navigation Satellite System (GNSS) meteorology is the latest concept for the use of GNSS signals in atmospheric sounding. One of the important implications of GNSS observations for atmospheric studies is the estimation of precipitable water vapour (PWV). PWV is an important parameter that significantly influences the accuracy of weather nowcasting and forecasting. Measurement of PWV is not easy in all weather conditions using the available optical remote sensing and in situ monitoring techniques. The space-based GNSS Radio Occultation (RO) technique provides valuable data for global atmospheric monitoring of PWV. As a radio signal travels through the limb of the atmosphere from a transmitter on a Global Positioning System (GPS) satellite, it is delayed and its path bent by gradients in the refractivity field. If this signal is received by a GPS receiver on another satellite behind the Earth's limb, then the time of reception and the direction from which the signal is received differ from those obtained from an unrefracted path. RO calculates the time delay in the occult signal that can be converted into an atmospheric bending angle and refractivity which is further utilised for the derivation of PWV. The GNSS-RO derived PWV are used mainly to enhance the accuracy of weather nowcasts, forecasts and work related to climate change.
Various studies have been performed using ground and space-based observations for PWV estimation. Bevis et al. (1992) first proposed a new approach based on the global positioning system (GPS) for remote sensing of water vapour. After that several efforts were made to derive PWV using GNSS signal received by ground-based observations throughout the globe (Jin et al., 2009; Bender et al., 2011; Karabatić et al., 2011). The accuracy of GPS derived PWV was extensively evaluated by various investigators using radiosonde observations (Musa et al., 2011, Perdiguer-López et al., 2019), satellite observations from Atmospheric Infrared Sounder (AIRS) and reanalysis products obtained from National Centers for Environmental Prediction (NCEP) and ERA Interim (Isioyeet al., 2017; Zhang et al., 2017). Karabatić et al. (2011) and Yuan et al. (2014) utilised GNSS observations for the derivation of near real-time water vapour and investigated its implication in the weather nowcasting. Ground-based GNSS PWV analyses have also been carried out over India using GPS observations (Jade et al., 2005; Suresh Raju et al., 2007). Jade and Vijayan (2008) estimated GPS PWV using NCEP reanalysis meteorological data and compared it with GPS PWV derived using meteorological sensor data over the Indian region. Prasad and Singh (2009) validated the integrated water vapour derived from Moderate Resolution Imaging Spectroradiometer (MODIS) Terra, AIRS, NCEP/DOE AMIP-II Reanalysis-2, and AERONET Sun photometer using ground-based GPS receivers over India and found good agreement (R2 = 95%, RMSE 3.87 mm) between GPS and AERONET water vapour. Singh et al. (2014) developed a weighted mean temperature model for India's extratropical region that is much better than the global model and very useful for estimating PWV.
Extensive studies were also performed using the space-based GNSS signal for PWV estimation. COSMIC (Constellation Observing System for Meteorology, Ionosphere, and Climate) derived water vapour are extensively validated over different global regions using special sensor microwave imager (Wick et al., 2008; Teng et al., 2013), radiosondes (Kishore et al., 2011; Zhang et al., 2018), advanced microwave scanning radiometer (Tenget al., 2013), ground-based GNSS observations (Fonseca et al., 2018), microwave radiometer (Ho et al., 2018), ECMWF, ERA-Interim, ERA5 reanalysis data (Zhang et al., 2018) etc. These studies established the excellent accuracy of COSMIC derived water vapour products. However, such studies highlighted the significant change in the accuracy of these products over different global regions. Zhang et al. (2018) investigated the variability and trends of PWV using COSMIC RO and radiosonde observations and found that the IGRA PWV is at least 5 mm larger than the COSMIC derived PWV in the tropical area. Xue et al. (2019) analysed sampling errors from COSMIC, SSMIS and HIRS and shown COSMIC has small sampling errors in higher latitudes and large sampling error in tropical regions which leads to a 3 mm error in the estimation of global total precipitable water. In view of this, it is extremely important to investigate the ability of space-based GNSS observations for PWV retrieval over the tropical Indian region.
The utilisation of space-based GNSS observations is not significant over the Indian region for PWV estimation. The present study is a step forward in this direction. In this work, space-based radio occultation technique is used for total column PWV estimation over the Indian region. The data products retrieved from the COSMIC RO mission are used to derive total column PWV over northern, eastern, western, and southern Indian regions. Total column PWV estimated for these regions are validated with PWV observations obtained from AIRS instrument aboard Aqua satellite and MERRA2 reanalysis data.
Section snippets
Data description
COSMIC is a constellation of six microsatellites. Assembled on each COSMIC satellite, the GPS antenna and receiver determine the refraction and time delay that comes when a GPS signal passes through Earth's atmosphere (Anthes, 2011). Raw measurements of the phase of the two signals are converted to bending angles and further to neutral bending angles. Refractivity is estimated from the bending angle, and from refractivity, temperature and water vapour pressure are derived.
COSMIC provides
Methodology
Using the COSMIC RO tangent points (also termed as impact points) information, PWV values can be obtained at different atmospheric layers. Total 147567 impact points are available at different pressure levels over the Indian subcontinent (5° to 40° N and 66° to 96° E). Fig. 1 shows 1564 impact points that are very close to the surface. Out of these points, uniformly distributed 293 points are chosen for which pressure difference between initial point pressure and surface pressure is less than
Results and discussions
Data from COSMIC RO over the four regions have been used to estimate PWV at different pressure levels as shown in Fig. 5. For R1, more variation is seen in PWV values from the surface to 400 hPa. After 400 hPa it appears to be constant with height as not much amount of water vapour is present above 400 hPa. The PWV value for the layer which is very near to the surface is estimated around 1 mm and it decreases with increasing altitude. For this region, vapour pressure is starting from 12 hPa and
Conclusions
Using the COSMIC satellite data total column PWV estimation is performed in four different regions of India representing different geographical conditions. PWV is also derived using measurements from AIRS and MERRA2 over these regions and a comparison is made among all PWV values at different pressure levels. The results show that, the COSMIC derived PWV matches very well with the PWV values obtained from other sources at various pressure levels. The mean differences are observed close to zero
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We thank all those data provider teams (COSMIC, AIRS and MERRA2) that are responsible for making it freely available. We are also grateful to anonymous reviewers for their comments and suggestions which have significantly improved the content of this manuscript. We would also like to thank Dr. Prakash Chauhan, Director, IIRS, Dr. S. K. Srivastav, Dean, IIRS, Mrs. Shefali Agrawal, Group Head, GTOPG and Dr. Sameer Saran, Head, Geoinformatics Department for their encouragement and support.
References (27)
- et al.
Development of a GNSS water vapour tomography system using algebraic reconstruction techniques
Adv. Space Res.
(2011) - et al.
Estimates of precipitable water vapour from GPS data over the Indian subcontinent
J. Atmos. Sol. Terr. Phys.
(2005) - et al.
Systematic errors between VLBI and GPS precipitable water vapour estimations from 5-year co-located measurements
J. Atmos. Sol. Terr. Phys.
(2009) - et al.
Near real-time estimation of tropospheric water vapour content from ground based GNSS data and its potential contribution to weather now-casting in Austria
Adv. Space Res.
(2011) - et al.
Global (50 S–50 N) distribution of water vapour observed by COSMIC GPS RO: comparison with GPS radiosonde, NCEP, ERA-Interim, and JRA-25 reanalysis data sets
J. Atmos. Sol. Terr. Phys.
(2011) - et al.
GPS meteorology in a low-latitude region: remote sensing of atmospheric water vapour over the Malaysian Peninsula
J. Atmos. Sol. Terr. Phys.
(2011) - et al.
Weighted mean temperature model for extra tropical region of India
J. Atmos. Sol. Terr. Phys.
(2014) AIRS/Aqua L3 Daily Support Product (AIRS-Only) 1 Degree X 1 Degree V006
(2013)Exploring Earth's atmosphere with radio occultation: contributions to weather, climate and space weather
Atmos. Meas. Tech
(2011)- et al.
GPS meteorology: remote sensing of atmospheric water vapour using the Global Positioning System
J. Geophys. Res.: Atmosphere
(1992)