IX. Propagation Delays Induced in GPS Signals by Hydrometeors and Other Particulates

Dry air, water vapor, hydrometeors and other particulates (sand, dust and volcanic ash) in the atmosphere introduce microwave propagation delays. These delays must be properly characterized to achieve the highest accuracy in surveying and other measurements using Global Positioning System (GPS) signals. The 24 GPS satellites broadcast 1.2 (L1) and 1.6 (L2) GHz carriers based on atomic clocks. Water vapor is typically the largest source of variable atmospheric delay.

Typical zenith GPS propagation delays are listed in Table 1.

Table 1: Zenith Path Delay at Sea Level

Source Magnitude,
< or = cm
Scale height,
km
"dry" air 245 8
water vapor 40 2

Radio Propagation Delays Induced by Hydrometeor Scattering

Propagation delay is produced by the forward scattering effects of raindrops in a rain medium. J. Vivekanandan used a rigorous T-matrix approach for computing the scattering characteristics of raindrops, hail, and snow particles. Using a single scattering approximation, the scattering matrices were averaged over a Marshall-Palmer particle size distribution. Rain droplets are approximately spherical up to 0.5 mm in diameter, and tend to be oblate spheroids in the larger diameters.

The propagation phase delay through the rain medium is shown in Figure 1.

Figure 1. GPS delays for L1 and L2 signals due to forward scattering in rain.

The propagation delay for hail is shown in Figure 2, assuming a particle bulk density of 0.9 g/cm3, oblate spheroid shape and an axis ratio of 0.8. The maximum hail size is truncated at 10 mm. The ice water content of the medium varies between 0.01 and 7 g/m3. The propagation delay in an ice cloud is 50% of the delay in a comparable rain medium. The delay is lower for ice medium because the dielectric constant of ice is smaller than the dielectric constant of raindrops.

Figure 2. GPS delays for L1 and L2 signals due to forward scattering in hail.

The propagation delay for a snow shower is shown in Figure 3. A snow particle is modeled as a low density oblate spheroid. We assumed a bulk density of snowflakes of 0.1 g/cm3 and an axis ratio of 0.8. The snow mass content varies between 0.01 and 0.7 g/m3. For a snow shower with 0.7 g/m3, the propagation delay is 0.6 mm. The lower bulk density results in a reduced delay.

Figure 3. GPS delays for L1 and L2 signals due to forward scattering in snow.

Delays induced by hydrometeors are widely variable and can be comparable to dry air and water vapor delays. A long path such as, 5 to 10 km through a squall line may produce a 50 to 100 mm delay and may introduce significant error in water vapor estimation using the GPS. Also, for a given liquid or ice water content, delay through an ice and snow cloud is smaller than delay through a rain medium.

NCAR | UCAR | NSF | NCAR FY96 ASR | NCAR FY97 ASR