Ian S. Adams

Evaluation of hurricane ocean vector winds from WindSat

The ability to accurately measure ocean surface wind vectors from space in all weather conditions is important in many scientific and operational usages. One highly desirable application of satellite-based wind vector retrievals is to provide realistic estimates of tropical cyclone intensity for hurricane monitoring. Historically, the extreme environmental conditions in tropical cyclones (TCs) have been a challenge to traditional space-based wind vector sensing provided by microwave scatterometers. With the advent of passive microwave polarimetry, an alternate tool for estimating surface wind conditions in the TC has become available. This paper evaluates the WindSat polarimetric radiometers ability to accurately sense winds within TCs. Three anecdotal cases studies are presented from the 2003 Atlantic Hurricane season. Independent surface wind estimates from aircraft flights and other platforms are used to provide surface wind fields for comparison to WindSat retrievals. Results of a subjective comparison of wind flow patterns are presented as well as quantitative statistics for point location comparisons of wind speed and direction.

Identification of Ocean-Reflected Radio-Frequency Interference Using WindSat Retrieval Chi-Square Probability

Ocean retrievals using passive microwave radiometers are sensitive to small fluctuations in ocean brightness temperatures. As such, the signals emanating from geostationary satellites that reflect off the ocean surface can result in large errors in ocean retrievals. Since geostationary communication satellites maintain fixed positions above the Earth and constantly transmit to predetermined regions while most other error sources, e.g., precipitation, are transient, time-averaged retrieval error statistics can be used to identify regions of measurements contaminated with radio-frequency interference (RFI). This letter describes a new method of identifying regions of ocean where ocean retrievals are affected by geostationary communication (television) satellites by using geophysical retrieval chi-square probability (goodness-of-fit) estimates. A three-month time-averaged collection of retrieval chi-square estimates is used to identify regions of the ocean where RFI may be present. This information is combined with information on geostationary satellite bandwidths, locations, and antenna contours to identify the source of the RFI. A mask derived from the analysis is used, in conjunction with satellite geometry calculations, to flag individual channels for RFI. These channels can then be ignored in the geophysical retrieval processing in order to produce uncontaminated ocean retrievals.

The Impact of Radio-Frequency Interference on WindSat Ocean Surface Observations

To study the effects of radio-frequency interference (RFI) on remote sensing data, we examined five years of WindSat ocean observations from regions affected by reflected X-band emissions from geostationary communication satellites. We compared measured brightness temperatures to modeled brightness temperatures obtained using mitigated retrievals as input to the WindSat parameterized radiative transfer model, demonstrating a considerable contribution to the radiometric measurements from interference. Comparisons of potentially contaminated retrievals of sea surface temperature (SST), wind speed, and wind direction with surface reanalyses confirmed that the presence of RFI can bias retrievals while also increasing retrieval uncertainty. Mitigation removed most of the biases. The quality of mitigated SST and wind speed retrievals was comparable to uncontaminated retrievals; however, the performance of first-rank wind directions was noticeably degraded. Recommendations are given on the use of contaminated and mitigated data, from radiance assimilation to climate studies.

Foreword to the Special Issue on Radio Frequency Interference: Identification, Mitigation, and Impact Assessment

The 14 papers in this special issue cover a combination of software and hardware solutions to the Radio Frequency Interference (RFI) problem, detail the challenges in monitoring RFI, and attempt to quantify the impact that interference has on measurements.

Hurricane wind retrievals using the SeaWinds scatterometer on QuikSCAT

Meteorologists require reliable real-time measurements of tropical cyclone (TC) conditions to issue hurricane forecasts and advisories. The models they utilize assimilate many sources of data including airborne and spaceborne surface wind estimates. One instrument that has the potential for operational use is the SeaWinds Scatterometer on QuikSCAT. Unfortunately, the adverse effects of rain on the scatterometer signal make these measurements unreliable. Intense rain volumes can drown out radar echo with a combination of attenuation and high backscatter, while the standard resolution of wind measurements tends to wash out high wind retrievals. Furthermore, traditional geophysical model functions (GMF), which relate wind speed and direction with radar backscatter (sigma-0), have not been tuned for the high wind conditions of storms because sampling of these events is poor. Thus, scatterometers tend to underestimate TC winds. By utilizing a combined active/passive retrieval algorithm developed specifically for TCs, we are able to simultaneously retrieve wind speed and rain rates using the SeaWinds Scatterometer. Unreliable wind measurements are flagged based on rain estimates. To certify that wind retrievals are ready for operational use, wind speeds are compared with H*Wind wind fields developed by the Hurricane Research Division of NOAA.

Seawinds radiometer (SRad) on ADEOS-II brightness temperature calibration/validation

NASA’s SeaWinds scatterometer, on Japan’s ADEOSII satellite, is a special purpose radar remote sensor used to measure ocean surface wind vector. This paper presents the novel use of SeaWinds as a radiometer (SRad), to measure the ocean’s brightness temperature simultaneously with the radar backscattered power. The calibration/validation of the SRad ocean Tb data processing algorithm uses the Advanced Microwave Scanning Radiometer (AMSR) that also operates on ADEOS-II as a brightness temperature standard. Empirical onorbit comparisons are presented for SRad and independent, simultaneous Tb measurements from AMSR. Keywords-component; SeaWinds scatterometer, AMSR, Brightness temperature, SRad calibration.

Combined active/passive hurricane wind retrieval algorithm for the Seawinds scatterometer

Because of their high wind gradient structures, tropical cyclones (TCs) present a major challenge to space-borne scatterometer measurements of ocean surface wind vectors. Frequently spiral bands of strong rains accompany the high winds, and this precipitation attenuates the ocean backscatter measured by the scatterometer. Furthermore, traditional geophysical model functions (GMF), which relate wind speed and direction with radar backscatter (sigma-0), have not been tuned for the high wind conditions of TCs. The SeaWinds scatterometer has the ability to measure simultaneously the ocean backscatter and brightness temperature. By using this combined active/passive approach, simultaneous wind and rain estimates are made in TCs. Rain rate, determined passively, is used to model both the attenuative and scattering effects of rain. These parameters are used to correct the measured ocean sigma-0 at 12.5 km resolution. Wind speed retrievals are performed using a special TC-GMF developed using airborne scatterometer measurements in hurricanes. SeaWinds wind speed results for several hurricanes occurring between 1999 and 2002 compare well with high-resolution surface wind fields available from NOAAs Hurricane Research Division aircraft flights.

Brightness Temperature Simulation of Observed Precipitation Using a Three-Dimensional Radiative Transfer Model

AbstractThis study demonstrates the capabilities of a three-dimensional radiative transfer model coupled to a polarized microwave surface emissivity model. Simulations are performed at 10, 19, and 37 GHz for TMI and WindSat using three-dimensional fields of rain, snow, and graupel derived from Tropical Rainfall Measuring Mission observations of moderate Tropical Storm Asma in conjunction with atmospheric profiles and surface fields from ECMWF. Simulations are well behaved and compare well with measured brightness temperatures. Comparisons are made between simulations with a wind-roughened surface and simulations assuming a specular surface. This theoretical study, which is supported with WindSat retrievals, shows the frequencies and conditions under which surface emissions may be detected in the presence of rain.

SeaWinds radiometer (SRad) brightness temperature calibration and validation

NASAs SeaWinds scatterometer, on Japans ADEOS-II satellite, is a special purpose radar remote sensor used to measure ocean surface wind vector (speed and direction). The paper presents a novel use of this instrument as a SeaWinds radiometer (SRad) to measure the ocean microwave emissions (brightness temperature). The derivation of the SRad radiometric transfer function is presented, which is used to calculate the apparent brightness temperature collected simultaneously with the radar scattering measurement. Results are presented for the on-orbit calibration and validation of the SRad brightness temperature algorithm performed using simultaneous measurements from the advanced microwave scanning radiometer (AMSR) also on the ADEOS-II.

Hurricane wind vector estimates from WindSat polarimetric radiometer

Abstract : WindSat is the worlds first microwave polarimetric radiometer, designed to measure ocean vector winds. In late 2004, the first preliminary oceanic wind vector results were released, and this paper presents the first evaluation of this product for several Atlantic hurricanes during the 2003 season. Both wind speed and wind direction comparisons will be made with surface wind analysis (H*Wind) developed by the NOAA Hurricane Research Division (HRD) and provided by the NOAA National Hurricane Center (NHC). Examples are presented where HRD aircraft flights were conducted within several hours of the WindSat overpass. These H*Wind surface wind analyses provide the most complete independent surface winds comparison data set available. Both WindSat retrieved wind speeds and wind directions are evaluated (against H*Wind) as a function of storm quadrant. To complement the analysis, rain rates were derived using WindSat brightness temperatures with a modified version of the TMI 2A12 heritage rain algorithm. Effects of rain on the derived wind speeds and directions are discussed.