18_cruz


84

Cruz and Villarin

Science Diliman (January-June 2003) 15:1, 84-87

An Analysis of the Precipitable Water Vapor
Observed over the PIMO GPS Station

F.T. Cruza* and J.T. Villarinb
aDepartment of Physics, Ateneo de Manila University

Loyola Heights, Quezon City, Philippines
bManila Observatory, Ateneo de Manila University Campus

Loyola Heights, Quezon City, Philippines
E-mail: aastinus@zuma.ps.admu.edu.ph

ABSTRACT

Remote sensing of the atmosphere using Global Positioning Systems has been made possible with the
derivation of the precipitable water vapor (PWV) from the tropospheric wet delay experienced by the
signal propagation. Although limited by the missing observational data from the receiver, it is observed
that the PWV obtained from this data gives reasonable values when considered for the cases of wet and
dry seasons and when analyzed with a measurable meteorological variable, such as the amount of rainfall.
A continual update of the record for PWV is highly recommended for further studies on the behavior of
the atmospheric water vapor and its contribution to the changing climate.

INTRODUCTION

Water vapor plays a significant role in the physical
and chemical processes of the atmosphere. It is also
considered as one of the major contributors to the
greenhouse effect. Analyses of the water vapor would
contribute greatly to studies, such as weather
forecasting and climate change.

Global Positioning Systems (GPS) have been used in
the remote sensing of the atmosphere. Due to their good
spatial and temporal coverage, the GPS’s complement
radiosondes and ground- and space-based water vapor
radiometers which are either limited in cold, dry regions
or in the presence of clouds and rain (Bevis et al., 1992).

GPS signals experience delays as they propagate
through the atmosphere. In the troposphere, the delays

are classified as hydrostatic delay or as wet delay. The
wet delay due to the atmospheric water vapor is more
variable and harder to remove. However, it is from
these delays that the precipitable water vapor can be
estimated (Bevis et al., 1992).

Derivation of the precipitable water vapor from
the zenith wet delay

The precipitable water vapor (PWV) is related to the
zenith wet delay (ZWD) by a factor of Π,

(1)

where both the PWV and ZWD are in units of length
(m). The dimensionless constant of proportionality, Π
is given by:

(2)

PWV ZWD= Π ×

6

'3
2

10

v
m

k
R k

T
ρ

Π =
⎡ ⎤⎛ ⎞

+⎢ ⎥⎜ ⎟
⎝ ⎠⎣ ⎦

* Corresponding author



85

Analysis of the Precipitable Water Vapor

where ρ is the density of water, R
v
 is the specific gas

constant for water vapor, T
m
 is a weighted mean

atmospheric temperature and k
3
 and k’

3
 are the

refractivity constants (Bevis et al., 1994)

The mean atmospheric temperature is related to the
surface temperature, T

s
,

(3)

where T
s
 can be obtained from any meteorological

station (Bevis et al., 1994).

Since the values of ρ, R
v
 are well-defined, the

uncertainty of Π is derived from the uncertainties of
T

m
 and the refractivity constants. If the term k

3
/T

m
 is

large enough, the contribution of k’
2
 can be neglected

and thus, the error in Π can be approximated to 2%
(Bevis et al., 1994).

METHODOLOGY

The PIMO station is one of the National Aeronautics
and Space Administration, Jet Propulsion Laboratory
(NASA JPL) GPS ground-based receivers in the
Philippines. It is located at a latitude of 14°35’ N and a
longitude of 121° E, within the grounds of the Manila
Observatory at the Ateneo de Manila University,
Quezon City.

The GPS receiver records the observation data every
300 seconds in GPS time. These are then sent to the
regional data centers, to the global data centers, and
then to the data analysis centers for processing. The
observation files in the Receiver Independent Exchange
(RINEX) format can be obtained from the Crustal
Dynamics Data Information Service (CDDIS) at
http://cddisa.gsfc.nasa.gov. These files would be
submitted to JPL’s AutoGIPSY for the processing for
tropospheric delay. The ZWD and PWV are derived
using the utility programs written in C language by
Reyes, which was later modified for this study (Reyes
& McNamara, 2000).

Temperature profiles are recorded every hour by the
Advanced Meteorological Station, located within the
grounds of the Manila Observatory.

It is very important to consider the time indicated in
the records. Observation files from the GPS receivers
are in GPS time, which differs from the Coordinated
Universal Time (UTC) by a few leap seconds.
Temperature profiles, on the other hand are recorded
in the local time, which is 8 hours ahead of the UTC.
For consistency in this study, the UTC is chosen as the
standard time.

In the process of obtaining the observation files, it is
noted that there were days with no available data, as
well as days with incomplete data. Therefore, in these
cases, the ZWD and PWV were set to zero, and
consequently, not included in the results.

RESULTS AND DISCUSSION

The Philippine climate is described mainly by its wet
and dry seasons. In this study, two months were
selected for each season based on the relative
completeness of the data for that month.

The dynamic change of the amount of PWV is observed
in Figs. 1 and 2. The amount of PWV for the months
of August and September 2000 (Fig. 1) varies from a
minimum value of 0.134 m to a maximum value of
0.167 m. On the other hand, for the months of
December 2000 and January 2001 (Fig. 2), it varies
from a minimum of 0.122 m to a maximum of
0.164 m. It is observed that the PWV lowered in general
during the dry season.

70.2 0.72m sT T= +

0.17

0.16

0.15

0.14

0.13

0.12
7/31/00

0:00
8/20/00

0:00
9/9/00
0:00

9/29/00
0:00

Time (UTC)

P
W

V
 (

m
)

Fig. 1. Plot of the Precipitable Water Vapor (PWV) for
August 2000 to September 2000 (wet season).



86

Cruz and Villarin

A comparison of the PWV with a measurable
meteorological variable would give some insight into
its behavior or variation.

The PWV is described as the height of the equivalent
column of liquid water directly above the receiver.
Thus, the variation of the amount of PWV can be
analyzed with rainfall data.

To illustrate, the case of December 31, 2000 is selected
since there is a noticeable dip in the PWV at this time
(Fig. 2). The hourly averages of the PWV are derived
(Fig. 3) since the rainfall data are recorded every hour
(Fig. 4).

In this case, after a slight variation during the early
hours, the PWV continually decreased at 1200 UTC. It

0.17

0.16

0.15

0.14

0.13

0.12

11/30/00
0:00

12/20/00
0:00

1/9/01
0:00

1/29/01
0:00

P
W

V
 (

m
)

Time (UTC)

Fig. 2. Plot of the Precipitable Water Vapor (PWV) for
December 2000 to January 2001 (dry season).

0.145

0.14

0.135

0.13

P
W

V
 (

m
)

0:00 4:48

Time (UTC)

9:36 14:24 19:12 0:00

Fig. 3. Plot of the Hourly Average Precipitable Water Vapor
for December 31, 2000.

0:00 7:12 14:24 21:36

Time (UTC)

5

4

3

2

1

0

R
a

in
fa

ll
 a

m
o

u
n

t 
(m

m
)

Fig. 4. Plot of the rainfall amount for December 31, 2000
where the peak occurs at 14:00 UTC.

is at this time when there was a sudden occurrence of
rain, which eventually lessened towards the end of the
day.

Therefore, it is reasonable to infer that when it rains,
there is an associated decrease in the PWV. It should
be noted that one of the considerations for this was the
relatively short distance between the GPS receiver and
the meteorological station since the PWV is defined
directly over the GPS receiver.

The archive for the observation data for the PIMO
station started only in April 1999. Thus, it is
recommended that there should be a continuous update
of the record and the analyses on PWV for studies such
as weather forecasting and climate variability.

ACKNOWLEDGMENTS

The authors are grateful for the help and support offered
by the staff of the Jet Propulsion Laboratory (JPL) and
Crustal Dynamics Data Information Service (CDDIS)
of the National Aeronautics and Space Administration
(NASA); the Environmental Management Bureau of
the Department of Environment and Natural Resources
of the Philippines; and Conexor for the Advanced
Meteorological Station data; Michael Reyes, Gemma
Narisma, Armelle Remedio, Luisito Agustin, and
Joanne Dy.



87

Analysis of the Precipitable Water Vapor

REFERENCES

Bevis, M., S. Businger, T. Herring, C. Rocken, R. Anthes, &
R. Ware, 1992. GPS Meteorology: Remote sensing of
atmospheric water vapor using the global positioning system.
J. Geophys. Res. 97(D14): 15787-15801.

Bevis, M., S. Businger, S. Chiswell, T. Herring, R. Anthes,
C. Rocken, & R. Ware, 1994. GPS Meteorology: Mapping
zenith wet delays onto precipitable water. J. Appl. Met. 33(3):
379-386.

Reyes, M. & D. McNamara, 2000. Observing zenith wet
delays and precipitable water vapor over JPL GPS stations
in the Philippines. Manila Observatory Scientific Report. 4:
1-12.