A look in the eyes of hurricanes by Global Positioning System

Nicolas Houlié and Gaetano Festa


The landfall of Hurricanes Frances, Charley, Jeanne, and, more recently, Katrina caused billions of dollars in damage to structures and property, the loss of thousands of lives, and displaced millions of people during the summers 2004 and 2005.

In order to mitigate the risk associated with hurricanes and tropical storms, several studies attempted to predict their occurrence, estimate their number and understand their behavior from birth to death (De Pondeca and Zou, 2001; Powers and Davis, 2002; Wang and Wu, 2004; Larcombe and Carter, 2004; McConochie et al., 2004; Xu et al., 2005; Tolman and Alves, 2005; Lu and Garrido, 2005; Emanuel, 2005). Accuracy in the prediction strongly depends on observations such as the temperature and pressure vertical profiles that constrain the modelling of thermodynamic processes involved inside the convection cell. Due to high-speed wind in the outer part of the turbine, however, actual operational tools are limited to measurements at the Earth surface, above and inside the eye of the hurricane. There is thus a need for an instrument capable of scanning the entire hurricane. In order to scan the outer part of the hurricane, which is the most humid and windy, several instruments are limited because they need light (i.e. LASER system) or because the wind velocity is too high (balloons). GPS is strongly attached to the ground and can work in extreme conditions, night and day. The particular sensitivity of the GPS to water vapour in the atmosphere allows us to constrain the temperature profile in the wet part of a hurricane.

Figure 2.21: The GPS network used in this study. The sites belong to the GPS-MET and IGS networks. Superimposed on the picture, we also plot the paths of the hurricanes that hit the Southeast United States during the summers of 2004 and 2005. The trajectories are derived from MODIS Observation.
\begin{figure}\epsfig{file=F01-network.eps, width=8cm}\end{figure}

Indeed, the radio waves emitted by each GPS satellite are delayed by the water vapor contained in the troposphere and can therefore be used as a powerful scanner of the lower atmosphere (Baby et al., 1988; Bevis et al., 1992; Rocken et al., 1993; Brunner et al., 1993; Bevis et al., 1994; Rocken et al., 1995; Bevis et al., 1996; Rocken et al., 1997; Coster et al., 1998; Ruffini et al., 1999; Pany et al., 2001). These studies encourage the use of GPS in order to determine the atmosphere state by using both GPS and ground based measurements. However, GPS has not yet been used to investigate hurricanes.

The paths of Hurricanes Charley, Frances, Jeanne, and Katrina have crossed the GPS network in the Southeast United States (Figure 2.21), allowing us to explore perturbations in pressure, temperature and relative humidity due to the passage of hurricanes in the troposphere.

In this paper, we specifically quantify the change of the vertical temperature profile in the wet part of Katrina during its passage over the Lousiana coast.

Figure 2.22: Abacus presenting the increase of the tropospheric delay as a function of the temperature $T$ and the relative humidity $H$. Gray lines represent the tropospheric delay contours for the pressure value of $1.013\ bar$, as measured on August 1, while black lines are drawn for the pressure value of $916\ mbar$, as measured on August 29.
\begin{figure}\epsfig{file=F03-test.eps, width=8cm}\end{figure}


Since we fix the ground temperature at $25\, ^oC$, the increase in the wet tropospheric delay can be interpreted as an average increase of the temperature along the whole air column above the station. We quantified the vertical gradient increase to be $1\,^oC/km$. An analogous increase of the temperature vertical gradient has been observed in the central part of the hurricane by dropsonds and Microwave Imager Sounder, with a maximum difference comparable with results presented here.


GPS networks, dedicated to tectonic and geodynamic problems in the US-Caribbean area, could be densified by adding instrumentation on boats or/and islands and could become a forecast tool to study the hurricanes, remotely, in 3D, by night and by day.

Our results suggest that we could evaluate the radial variations of the vertical temperature profile in the wet part of the hurricanes using GPS, improving the real-time state of the hurricanes and allowing for short-term prediction of the trajectory of hurricanes. These profiles would be directly related to the distance between the GPS receivers, the position of which is known, and the eye of the hurricanes, the location of the latter being accurately computed by remote sensing imagery (e. g. MODIS, SPOT). Non-linearity in the hurricane trajectories is owed to sharp changes in the temperature at the sea surface and to the presence of sparse islands. Indeed, the use of GPS measurements could also help describe the air column's state above the starting areas of hurricane, better constraining the prediction of its future path.

The integration of 2D or 3D effects that may not be secondary for the evaluation of the tropospheric delays will be the next step of GPS applications to hurricanes.


H. Baby, P. Golé, and A. Lavergnat, J.
Model for tropospheric excess path length of radio waves from surface meteorological measurements.
Radio Science, 22:-1038, 1988.

M. Bevis, S. Businger, T. A. Herring, C. Rocken, R. A. Anthes, and R. H. Ware.
GPS Meteorology: Remote Sensing of Atmospheric Water Vapor Using the Global Positioning System.
J. Geophys. Res., 97(16):-15801, October 1992.

M. Bevis, S. Businger, S. Chiswell, T.A. Herring, R.A. Anthes, C. Rocken, and R.H. Ware.
Gps meteorology: Mapping zenith wet delays onto precipitable water.
Journal of Applied Meteorology, 33(3):-386, 1994.

M. Bevis, S. Chiswell, S. Businger, T. T. Herring, and Y. Bock.
Estimating Wet Delays Using Numerical Weather Analyses and Predictions.
Radio Sci., 31(3):-487, 1996.

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

M. S. F. V. De Pondeca and X. Zou.
A case study of the variational assimilation of gps zenith delay observations into a mesoscale model.
Am. Meteo. Soc., 40:-1576, 2001.

K.A. Emanuel.
Increasing destructiveness of tropical cyclones over the 30 years.
Nature, 436, 2005.

R. Hanssen.
Atmospheric heterogeneities in ERS tandem SAR interferometry, Chapter 2.
DEOS Report, 98.1:-24, 1998.

P. Larcombe and R.M. Carter.
Cyclone pumping, sediment partitioning and the development of the great barrier reef shelf system: a review.
Quaternary Science Reviews, 23:-135, 2004.

Y. Lu and J. Garrido.
Doubly periodic non-homogeneous Poisson models for hurricane data.
Statistical Methodology, 2:-35, 2005.

J.D. Mc Conochie, T.A. Hardy, and L.B. Mason.
Modelling tropical cyclone over-water wind and pressure fields.
Ocean Engineering, (31):-1782, 2004.

T. Pany, P. Pesec, and G. Stangl.
Atmospheric GPS slant path delays and ray tracing through numerical weather models, a comparison.
Physics and Chemistry of the Earth A, 26:-188, January 2001.

J. Powers and C.A Davis.
A cloud-resolving regional simulation of tropical cyclone formation.
Atmospheric Science Letters, 3(1):-24, 2002.

C. Rocken, T. Van Hove, J. Johnson, F. Solheim, R. Ware, M. Bevis, S. Chiswell, and S. Businger.
GPS/STORM—GPS sensing of atmospheric water vapor for meteorology.
Jour. of Atmos. and Ocean. Tech., 1995, 12(3):-478, 1995.

C. Rocken, T. Van Hove, and R. Ware.
Near real-time GPS sensing of atmospheric water vapor.
Geophys. Res. Lett., 24:-3224, December 1997.

C. and Johnson, J. (1993)]rocken1993 Rocken, C. and Ware, R. and Van Hove, T. and Solheim, F. and Alber, C. and Johnson, J. .
Sensing atmospheric water vapor with the global positioning system.
Geophys. Res. Lett., 20:-2634, 1993.

Flores]ruffini1999 G. Ruffini, L.P. Kruse, A. Rius, B. Burki, L. Cucurull, and A. Flores.
Estimation of Tropospheric Zenith Delay and Gradients over the Madrid Area Using GPS and WVR Data.
Geophys. Res. Lett., 26:-450, 1999.

H.L. Tolman and J-.H. G.M. Alves.
Numerical modeling of wind waves generated by tropical cyclones using moving grids.
Ocean Modelling, 9:-323, 2005.

Y. Wang and C.C. Wu.
Current understanding of tropical cyclone structure and intensity changes – a review.
Meteorol. Atmos. Phys., 87:-278, 2004.

X. Xu, L. Xie, X. Zhang, and W. Yao.
A mathematical model for forecasting tropical cyclone tracks, Nonlinear Analysis.
Real World Applications, In Press, Corrected Proof, Available online 20 April, 2005.


Berkeley Seismological Laboratory
215 McCone Hall, UC Berkeley, Berkeley, CA 94720-4760
Questions or comments? Send e-mail:
© 2007, The Regents of the University of California