Melting Layer Attenuation at Ka- and W-bands as Derived from Multi-frequency Radar Doppler Spectra Observations

Moisseev, D. N., University of Helsinki

Cloud Processes

High Latitude Processes

Li H and D Moisseev. 2019. "Melting layer attenuation at Ka‐ and W‐bands as derived from multi‐frequency radar Doppler spectra observations." Journal of Geophysical Research: Atmospheres, , 10.1029/2019JD030316. ONLINE.


Differential attenuation and X-band observed reflectivity factor (a). Spectra observations in snow and rain areas indicated by the grey and black arrows are shown in panels (b, c) and (d, e) respectively. The grey shading identifies regions of spectra where hydrometeors are Rayleigh scatterers.



Differential attenuation in the melting layer for May 11, 2014 precipitation event. Radar reflectivity profiles observed by X-SACR are given in panel (a), DWR profiles for X-/Ka-SACR in (b), and KAZR/WMACR in (d). The retrieved melting-layer differential attenuation from X-/Ka-SACR observations is shown in panel (c) and KAZR/WMACR in (e). Panels (c) and (e) compare the DWR values calculated at cloud top and DSR values derived from the Rayleigh parts of spectra observations at the melting-layer top, respectively. The gaps during 03:10 ~ 03:15 in (b) and (c) are due to missing X- and Ka-SACR spectra data.



Differential attenuation and X-band observed reflectivity factor (a). Spectra observations in snow and rain areas indicated by the grey and black arrows are shown in panels (b, c) and (d, e) respectively. The grey shading identifies regions of spectra where hydrometeors are Rayleigh scatterers.


Differential attenuation in the melting layer for May 11, 2014 precipitation event. Radar reflectivity profiles observed by X-SACR are given in panel (a), DWR profiles for X-/Ka-SACR in (b), and KAZR/WMACR in (d). The retrieved melting-layer differential attenuation from X-/Ka-SACR observations is shown in panel (c) and KAZR/WMACR in (e). Panels (c) and (e) compare the DWR values calculated at cloud top and DSR values derived from the Rayleigh parts of spectra observations at the melting-layer top, respectively. The gaps during 03:10 ~ 03:15 in (b) and (c) are due to missing X- and Ka-SACR spectra data.

Science

While the melting layer is a relatively narrow layer in precipitation systems, it has a significant impact on telecommunication and remote-sensing applications. For spaceborne and ground-based radar measurements, unaccounted attenuation in the melting layer may cause significant errors in retrievals of rainfall rate and ice-cloud properties, respectively. There are two approaches to quantify attenuation in the melting layer, namely through modeling or observations. Modeling studies have been carried out and are reported in the scientific literature; however, a shortage of observational studies that could help us to constrain the models is acknowledged. In this study, we use radar Doppler spectra recorded at X-, Ka-, and W-bands to retrieve melting-layer attenuation. We show that the estimated attenuation at Ka- and W-bands agrees reasonably well with previous investigations, but there are indications of differences at higher rain rates.

Impact

This is the first study deriving the melting-layer attenuation at Ka- and W- bands from cloud radar observations. This study confirms that the integrated attenuation above the melting layer can be as large as 2 dB and 7 dB at Ka- and W-bands, respectively. If this attenuation is not taken into account, retrievals of ice cloud physical properties in stratiform precipitation systems would be heavily biased. We propose new parameterizations of melting-layer attenuation to address this problem.

We show that the estimated attenuation at Ka- and W-bands agrees reasonably well with previously reported studies, but there are indications of differences at higher rain rates. This agreement indicates the possibility of adopting simplified melting snowflake models for forward-scattering computations. It is, however, found that the modeling-based attenuation parametrization seems to overestimate the attenuation for moderate to heavy rainfall. This difference could possibly be explained by the assumed snow microphysical properties that are used as input into previously reported melting-layer models. This stresses the need for more comprehensive modeling- and observation-based studies of melting-layer properties.

We report how multi-frequency radar Doppler spectra can be used to retrieve the melting-layer attenuation at Ka- and W-bands. Multi-frequency radar Doppler spectra observations recorded during the Biogenic Aerosols Effects on Clouds and Climate (BAECC) field campaign are used to retrieve the melting-layer attenuation at Ka- and W-bands. The technical rationale is the differential attenuation between the strong- and weak-attenuation frequencies, which avoids other factors such as combined wet radome, rain, and melting layer, as well as possible calibration offsets.

Summary

We present a technique for deriving the melting-layer attenuation at Ka- and W-bands using X-, Ka-, and W-band vertically pointing radar Doppler spectra. The presented analysis is based on identifying Rayleigh scattering regions in radar Doppler spectra measurements where dual-wavelength spectral ratios can be related to differential attenuation. At vertical incidence, radar Doppler spectra separate radar signals from hydrometeors according to their fall velocities. Since smaller particles usually fall slower, there is a part of the Doppler spectrum where the scattering from observed hydrometeors can be described using the Rayleigh approximation even at W-band.

Since the proposed retrieval is based on the measurements of differential attenuation, it avoids potential radar calibration and wet radome issues. This research uses multi-frequency Doppler spectra observations collected by the US DOE Atmospheric Radiation Measurement (ARM) second mobile facility (AMF2) during the Biogenic Aerosols Effects on Clouds and Climate (BAECC) field campaign in Finland (Petäjä et al. 2016).