Continuous Humidity Monitoring in a Tropical Region with the Equatorial Atmosphere Radar

Abstract

A radar remote sensing technique that estimates humidity profiles using a wind profiler is applied to the equatorial atmosphere radar (EAR) to monitor detailed humidity variations in tropical regions. Turbulence echo power intensity is related to the vertical refractive index gradient squared (M2). Here, M is primarily determined by the vertical gradient of specific humidity in the lower troposphere. These relations are employed to estimate a humidity profile. EAR is a 47-MHz very high frequency (VHF) atmospheric radar installed at KotoTabang in West Sumatra, Indonesia. A humidity-profiling method, which was recently developed for the middle- and upper-atmosphere (MU) radar, was applied to EAR. The aim was to test this new method with a larger dataset observed in the tropical region, where detailed humidity variations have not been fully revealed. EAR observations were carried out in November 2002. Turbulence echoes from one vertical and four oblique beams were obtained with time and height resolutions of 3 min and 150 m, respectively. Spano and Ghebrebrhan's optimized 16-bit code was used for EAR in the entire height range in order to increase the signal-to-noise ratio of the turbulence observation. The effects of pulse compression and coherent and incoherent integrations were removed from the echo power intensity. Because the echo power intensity with the vertical beam was affected by partial reflection, M2 was averaged over four oblique beams. The humidity profiles were estimated using EAR at heights of 1.5?7.5 km. Below 3.0 km the sensitivity of the EAR receiver decreased and the radar-derived |M| was also reduced. Thus, the radar-derived |M| below 3.0 km was adjusted by multiplying the reduction factor calculated from the comparisons with radiosonde results in the campaign period. In this analysis, time-interpolated temperature profiles of radiosonde observations were used to estimate humidity profiles. Detailed variations of humidity corresponded well to rain distribution observed simultaneously with the L-band boundary layer radar (BLR) and the X-band radar, and to the cloud bottom height observed with a ceilometer.