Improvement in the Water Vapor Emission Model from GPS/WVR Comparisons

Introduction

An advanced troposphere calibration system, based on a Water Vapor Radiometer (WVR), is being developed for support of Cassini Radio Science experiments. In particular, the noise level (dimensionless strain sensitivity) for the Gravitational Wave Experiment (GWE) will be limited, to a large extent, by the calibration accuracy of tropospheric microwave delay changes on 100-10,000 s timescales.

One significant error component in any WVR-based calibration system involves the water vapor emission model. We measure sky brightness temperatures with a WVR. In order to convert these measurements into microwave path delays, we need to know the emission coefficient of water vapor. This coefficient is currently known to 5%-7% accuracy - we need a factor of two improvement for GWE.

GPS/WVR Comparison

Measurements of transmissions from the Global Positioning System (GPS) satellites can be used to determine the microwave delay from the earth's troposphere. Comparison between GPS and WVR measurements from the same site yields the ratio between these GPS delay measurements and the WVR brightness temperatures, and allows the water vapor emission coefficient to be determined.

Our initial test of this technique was done at Goldstone in the summer and fall of 1996. We operated two WVRs and two GPS receivers nearly continuously at DSS 13. The sampling interval for the GPS data was five minutes. The WVR data was interpolated to the epoch of the GPS data for comparison purposes. Figure 1 shows the GPS and WVR delay time series for the wettest and longest (42 days) segment of data. (The differences are too small to be seen when printed in black and white).

Table 1 gives statistics on the mean wet delay, and GPS/WVR agreement for each of our four data segments.

Future Work

The standard deviations from our Goldstone test are approximately equal to the rms sum of estimated GPS and WVR errors. For the relatively small total wet tropospheric delays at Goldstone, we cannot yet improve our knowledge of the water vapor emission coefficient. However, we expect GPS/WVR comparisons at a warm, humid site (e.g. Florida, the Texas coast, or Hawaii) to yield a more accurate comparison. For our 'humid site' experiment, we plan to add measurements from one more instrument. A microwave temperature profile, using passive observations in the 60 GHz band of molecular oxygen emission, measures the temperature vs. height in the lower troposphere. These temperature profiles reduce the errors in the interpretation of WVR brightness temperatures.

• Top of Page
• Return to Homepage

• TMOD Reorganization: Changes for the Better
• Raster-Scan for Calibration of DSN Antennas
• ARTSN: An Automated Real-Time Spacecraft Navigation System
• Space VLBI Co-Observing at the DSN