Radar Wind Profiler Research Articles

Boundary layer evolution over the central Himalayas from radio wind profiler and model simulations

Abstract. We investigate the time evolution of the Local Boundary Layer (LBL) for the first time over a mountain ridge at Nainital (79.5° E, 29.4° N, 1958 m a.m.s.l.) in the central Himalayan region, using a radar wind profiler (RWP) during November 2011 to March 2012, as a part of the Ganges Valley Aerosol Experiment (GVAX). We restrict our analysis to clear–sunny days, resulting in a total of 78 days of observations. The standard criterion of the peak in the signal-to-noise ratio (S ∕ N) profile was found to be inadequate in the characterization of mixed layer (ML) top at this site. Therefore, we implemented a criterion of S ∕ N > 6 dB for the characterization of the ML and the resulting estimations are shown to be in agreement with radiosonde measurements over this site. The daytime average (05:00–10:00 UTC) observed boundary layer height ranges from 440 ± 197 m in November (late autumn) to 766 ± 317 m above ground level (a.g.l.) in March (early spring). The observations revealed a pronounced impact of mountain topography on the LBL dynamics during March, when strong winds (> 5.6 m s−1) lead to LBL heights of 650 m during nighttime. The measurements are further utilized to evaluate simulations from the Weather Research and Forecasting (WRF) model. WRF simulations captured the day-to-day variations up to an extent (r2 = 0.5), as well as the mean diurnal variations (within 1σ variability). The mean biases in the daytime average LBL height vary from −7 % (January) to +30 % (February) between model and observations, except during March (+76 %). Sensitivity simulations using a mixed layer model (MXL/MESSy) indicated that the springtime overestimation of LBL would lead to a minor uncertainty in simulated surface ozone concentrations. However, it would lead to a significant overestimation of the dilution of black carbon aerosols at this site. Our work fills a gap in observations of local boundary layer over this complex terrain in the Himalayas, and highlights the need for year-long simultaneous measurements of boundary layer dynamics and air quality to better understand the role of lower tropospheric dynamics in pollution transport.

Impact of Assimilating Wind Profiling Radar Observations on Convection-Permitting Quantitative Precipitation Forecasts during SCMREX

AbstractTo improve the prediction of heavy rainfall in southern China during the prerainy season, horizontal wind data from wind profiling radars (WPRs) were assimilated in the partial-cycle data assimilation (DA) system based on a three-dimensional variational method. The analyses from the DA system are used as initial conditions for the convection-permitting Global/Regional Assimilation and Prediction System (GRAPES) model. The impact of assimilating WPR data on the quantitative precipitation forecast (QPF) in southern China was evaluated over the period of the Southern China Monsoon Rainfall Experiment (SCMREX) in May 2014, by comparing the results of a control experiment with WPR data assimilated and a denial experiment without WPR data. The positive impact of WPR DA was significant on the forecasts of atmospheric variables in the vertical and diagnostic fields at the surface, especially those of surface wind fields in the 0–6-h range. The inclusion of WPR data also improved the QPF skill of light and heavy rainfall throughout the 12-h forecast period by reducing the predicted spurious precipitation (thereby alleviating overestimations and false alarms), with the largest improvement in 6-h heavy rainfall forecasts. WPR DA considerably alleviated the spinup problem, remarkably improving the QPF of heavy rainfall (especially extreme rainfall). The improved representation of wind and moisture at lower levels in the analyses due to WPR DA was the physical cause of the QPF improvement, as is illustrated using a case study.

Technical Aspects of 205 MHz VHF Mini Wind Profiler Radar for Tropospheric Probing

Wind profiler radars (WPRs) are the state-of-the-art technology used for continuous remote sensing of 3-D wind structures from near surfaces to a height of about 20 km. Depending on the magnitude of atmospheric turbulence, the size, power, and frequency of the radar can vary. In this letter, the technical aspects of the design and the development of a 205-MHz very high frequency WPR installed at Cochin (10.04° N, 76.33° E), India, for troposphere and lower stratospheric studies in the tropical monsoon region are presented. The stratosphere–troposphere (ST) radar at Cochin is unique because it is the first WPR in the world operating at 205 MHz with a bandwidth of 5 MHz. The mini WPR with 49 three-element Yagi-Uda antennas at 205 MHz is working successfully and providing accurate wind measurements from 1 to 3.5–8 km, depending on atmospheric conditions. WPR data are comparable with collocated high-resolution Global Positioning System radiosonde observations. The main ST radar with 619 antenna elements is in the installation stage and is expected to be operational within a period of six months.

The POWER Experiment: Impact of Assimilation of a Network of Coastal Wind Profiling Radars on Simulating Offshore Winds in and above the Wind Turbine Layer

AbstractDuring the summer of 2004 a network of 11 wind profiling radars (WPRs) was deployed in New England as part of the New England Air Quality Study (NEAQS). Observations from this dataset are used to determine their impact on numerical weather prediction (NWP) model skill at simulating coastal and offshore winds through data-denial experiments. This study is a part of the Position of Offshore Wind Energy Resources (POWER) experiment, a Department of Energy (DOE) sponsored project that uses National Oceanic and Atmospheric Administration (NOAA) models for two 1-week periods to measure the impact of the assimilation of observations from 11 inland WPRs. Model simulations with and without assimilation of the WPR data are compared at the locations of the inland WPRs, as well as against observations from an additional WPR and a high-resolution Doppler lidar (HRDL) located on board the Research Vessel Ronald H. Brown (RHB), which cruised the Gulf of Maine during the NEAQS experiment. Model evaluation in the lowest 2 km above the ground shows a positive impact of the WPR data assimilation from the initialization time through the next five to six forecast hours at the WPR locations for 12 of 15 days analyzed, when offshore winds prevailed. A smaller positive impact at the RHB ship track was also confirmed. For the remaining three days, during which time there was a cyclone event with strong onshore wind flow, the assimilation of additional observations had a negative impact on model skill. Explanations for the negative impact are offered.

Observations of vertical wind profiles of tropical cyclones at coastal areas

Vertical wind profiles of tropical cyclones (TCs) below 5000m with marine and rugged exposures are investigated based on synchronized measurements from Doppler radar wind profilers and anemometers at three coastal meteorological stations in Hong Kong. The dependence of TCs׳ mean wind structure on storm-relative position, wind strength and exposure is discussed in the context of both case study and composite analysis. For onshore winds, it is found that the wind speed profiles follow the logarithmic-law below about 400m, and the mean wind veers logarithmically with height throughout the Ekman layer. Moreover, the roughness length scale reaches a maximum value and then decreases with increasing wind speed. As the exposure transforms from a marine type to rugged terrains, the mean wind structures vary markedly, featured by a stronger wind shear in a deepened boundary layer. The low-level jets of mean speed profiles are observed in the range of 330–1600m and tilt outward with height. The gust factor of surface wind and the maximum wind scaled by a layer mean wind speed vary with the storm-relative position as well. The obtained profiles also reveal the occasional existence of a secondary maximum wind near 3000–3500m that is accompanied with more turbulent surface wind.

Validation of 205 MHz wind profiler radar located at Cochin, India, using radiosonde wind measurements

AbstractA novel Mini Wind Profiling radar at 205 MHz has been set up at Cochin (10.04°N, 76.33°E), India, with the main objective of studying the dynamics of the Indian summer monsoon. This paper presents the first time validation of this Mini Wind Profiler using collocated radiosonde wind measurements. The wind profiles within 1–6 km altitude range have been validated. The radar wind profiles match very well with the radiosondes both in terms of magnitude and direction and ideally captures the day‐to‐day wind variations. The correlation between radar and radiosonde's zonal and meridional wind is 0.91 and 0.85, respectively. The standard deviation of the difference between radiosonde and radar for zonal wind is found to be 1.95 m s−1 and 1.56 m s−1 for meridional wind. These observations are seen to agree quite well with other established wind profiler radars operating elsewhere in the world. The results show that the 205 MHz Wind Profiler is capable of providing high‐quality wind data, and this could be pivotal for monsoon studies in years to come.

Raindrop size distribution of easterly and westerly monsoon precipitation observed over Palau islands in the Western Pacific Ocean

This paper explores the characteristics of raindrop spectra in terms of raindrop size distribution (RSD) using 4years of Joss–Waldvogel disdrometer data over Palau islands (7o 20′ N, 134o 28′ E) in Western Tropical Pacific ocean. The RSD characteristics are studied in two seasons (easterly monsoon—EM and westerly monsoon—WM) using three (stratiform, deep convection, and shallow convection) rain types identified from collocated 1290-MHz wind profiler radar (WPR). In addition to the ground-based sensors observations, TRMM and MODIS satellite-derived rain parameters and atmospheric parameters are utilized to study RSD characteristics. RSD characteristics stratified on the basis of rainrate show that the mean values of raindrop concentrations of small (medium and larger) drops are same (more) in WM compared to EM season. Normalized gamma distribution of RSD shows that the mean value of mass-weighted mean diameter, Dm (normalized intercept parameter, log10Nw), is higher (lower) in WM than the EM season. In addition, the mean value of Dm (log10Nw) is higher (lower) in deep convective precipitation as compared to the other two types of precipitation (stratiform and shallow convection) in both monsoon periods. In conjunction with the remote sensing data (MODIS & TRMM), RSD shows that the presences of cold clouds which extend to deeper altitudes are responsible for the higher Dm during WM season. The immediate significance of the present work is that (1) it contributes to our understanding of seasonal variations of RSD and distribution of different rain types, and (2) it provides information which is useful for quantitative estimation of rainfall from weather radar observations.

ウィンドプロファイラとゾンデから求めた乱流エネルギー消散率の比較

ウィンドプロファイラとゾンデから求めた乱流エネルギー消散率の比較

The Microwave Radiative Properties of Falling Snow Derived from Nonspherical Ice Particle Models. Part II: Initial Testing Using Radar, Radiometer and In Situ Observations

AbstractIn this study, two different particle models describing the structure and electromagnetic properties of snow are developed and evaluated for potential use in satellite combined radar–radiometer precipitation estimation algorithms. In the first model, snow particles are assumed to be homogeneous ice–air spheres with single-scattering properties derived from Mie theory. In the second model, snow particles are created by simulating the self-collection of pristine ice crystals into aggregate particles of different sizes, using different numbers and habits of the collected component crystals. Single-scattering properties of the resulting nonspherical snow particles are determined using the discrete dipole approximation. The size-distribution-integrated scattering properties of the spherical and nonspherical snow particles are incorporated into a dual-wavelength radar profiling algorithm that is applied to 14- and 34-GHz observations of stratiform precipitation from the ER-2 aircraftborne High-Altitude Imaging Wind and Rain Airborne Profiler (HIWRAP) radar. The retrieved ice precipitation profiles are then input to a forward radiative transfer calculation in an attempt to simulate coincident radiance observations from the Conical Scanning Millimeter-Wave Imaging Radiometer (CoSMIR). Much greater consistency between the simulated and observed CoSMIR radiances is obtained using estimated profiles that are based upon the nonspherical crystal/aggregate snow particle model. Despite this greater consistency, there remain some discrepancies between the higher moments of the HIWRAP-retrieved precipitation size distributions and in situ distributions derived from microphysics probe observations obtained from Citation aircraft underflights of the ER-2. These discrepancies can only be eliminated if a subset of lower-density crystal/aggregate snow particles is assumed in the radar algorithm and in the interpretation of the in situ data.

Reduced Image Aliasing With Microwave Radiometers and Weather Radar Through Windowed Spatial Averaging

Microwave remote sensing instruments detect and image physical phenomena such as brightness temperature and volume reflectivity. The spatial resolution of these measurements is limited by the physical properties of the instrument such as the antenna size, the spatial scan pattern, and temporal sampling. Analysis shows that common sampling schemes undersample the spatial information present at the antenna. Here, we address methods to better capture the spatial information available by applying the Nyquist–Shannon sampling theory to the spatial averaging and sampling of remote sensing data. The use of overlapping windows for spatial averaging rather than treating pixels independently improves the image fidelity while maintaining the system sensitivity. Additionally, the sensitivity to spatially small targets can be maximized by matching the window shape to the antenna pattern. The spatial imaging of scanning radiometers, radars, and phased-array systems is addressed. These principles are demonstrated with the theory and data from the National Aeronautics and Space Administration Goddard Space Flight Center's High-Altitude Imaging Wind and Rain Airborne Profiler (HIWRAP) radar.

Wind Profilers to Aid with Monitoring and Forecasting of High-Impact Weather in the Southeastern and Western United States

Abstract With funding provided by the 2012 Disaster Relief Act (Sandy Supplemental), NOAA’s Earth System Research Laboratory Physical Sciences Division has installed three Doppler wind-profiling radars and surface meteorology towers along the U.S. Gulf and southeast coasts to help detect and monitor landfalling tropical storms and other high-impact weather events. This same combination of instruments has been used to monitor landfalling atmospheric rivers on the U.S. West Coast. For this reason, we refer to the whole collection of instruments at each site as an Atmospheric River Observatory (ARO). These three new AROs supported by the Sandy Supplemental complement a fourth ARO deployed in coastal North Carolina as part of NOAA’s Hydrometeorology Testbed Southeast Pilot Study. These four AROs were installed in time to capture the 2014 hurricane season and will be operated through the 2015 hurricane season.

The Mesoscale Heavy Rainfall Observing System (MHROS) over the middle region of the Yangtze River in China

AbstractThe Mesoscale Heavy Rainfall Observing System (MHROS), supported by the Institute of Heavy Rain (IHR), Chinese Meteorology Administration, is one of the major systems to observe mesoscale convective systems (MCSs) over the middle region of the Yangtze River in China. The IHR MHROS consists of mobile C‐POL and X‐POL precipitation radars, millimeter wavelength cloud radar, fixed S‐band precipitation radars, GPS network, microwave radiometers, radio soundings, wind profiler radars, and disdrometers. The atmospheric variables observed or retrieved by these instruments include the profiles of atmospheric temperature, moisture, wind speed and direction, vertical structures of MCS clouds and precipitation, atmospheric water vapor, and cloud liquid water. These quality‐controlled observations and retrievals have been used in mesoscale numerical weather prediction to improve the accuracy of weather forecasting and MCS research since 2007. These long‐term observations have provided the most comprehensive data sets for researchers to investigate the formation‐dissipation processes of MCSs and for modelers to improve their simulations of MCSs. As the first paper of a series, we briefly introduce the IHR MHROS and describe the specifications of its major instruments. Then, we provide an integrative analysis of the IHR MHROS observations for a heavy rain case on 3–5 July 2014 as well as the application of IHR MHROS observations in improving the model simulations. In a series of papers, we will tentatively answer several key scientific questions related to the MCS and Meiyu frontal systems over the middle region of the Yangtze River using the IHR MHROS observations.

The Wind Forecast Improvement Project (WFIP): A Public–Private Partnership Addressing Wind Energy Forecast Needs

Abstract The Wind Forecast Improvement Project (WFIP) is a public–private research program, the goal of which is to improve the accuracy of short-term (0–6 h) wind power forecasts for the wind energy industry. WFIP was sponsored by the U.S. Department of Energy (DOE), with partners that included the National Oceanic and Atmospheric Administration (NOAA), private forecasting companies (WindLogics and AWS Truepower), DOE national laboratories, grid operators, and universities. WFIP employed two avenues for improving wind power forecasts: first, through the collection of special observations to be assimilated into forecast models and, second, by upgrading NWP forecast models and ensembles. The new observations were collected during concurrent year-long field campaigns in two high wind energy resource areas of the United States (the upper Great Plains and Texas) and included 12 wind profiling radars, 12 sodars, several lidars and surface flux stations, 184 instrumented tall towers, and over 400 nacelle anemometers. Results demonstrate that a substantial reduction (12%–5% for forecast hours 1–12) in power RMSE was achieved from the combination of improved numerical weather prediction models and assimilation of new observations, equivalent to the previous decade’s worth of improvements found for low-level winds in NOAA/National Weather Service (NWS) operational weather forecast models. Data-denial experiments run over select periods of time demonstrate that up to a 6% improvement came from the new observations. Ensemble forecasts developed by the private sector partners also produced significant improvements in power production and ramp prediction. Based on the success of WFIP, DOE is planning follow-on field programs.

Orographic Precipitation Forcing along the Coast of Northern California during a Landfalling Winter Storm

Abstract This study documents orographic precipitation forcing along the coastal mountains of Northern California during the landfall of a significant winter storm over the period 16–18 February 2004. The primary observing asset is a scanning X-band Doppler radar deployed on the coast at Fort Ross, California, which provides low-level (e.g., below 1 km MSL) horizontal and vertical scans of radial velocity and reflectivity to characterize airflow and precipitation structures. Further context is provided by a wind-profiling radar, a radio acoustic sounding system (RASS), balloon soundings, buoys, a GPS receiver, and surface meteorological sensors. The winter storm is divided into two episodes, each having pre-cold-frontal low-level jet (LLJ) structures and atmospheric river characteristics. Episode 1 has a corridor of terrain-trapped airflow (TTA) that forms an interface with the LLJ. The interface extends ~25 km offshore in a ~0.5-km vertical layer, and the western edge of this interface near the ocean surface advances toward the coast over the course of ~5 h. The TTA acts as a dynamically driven barrier, so that the incoming LLJ slopes upward offshore below 1.5 km MSL and precipitation is enhanced over the ocean and near the coast. The absence of a TTA in episode 2 allows the cross-barrier flow to slope upward and enhance precipitation directly over the coastal mountains. A theoretical analysis favors the hypothesis that a gap flow exiting the Petaluma Gap forces the TTA.

Measuring boundary-layer height under clear and cloudy conditions using three instruments

Boundary-layer height (BLH) under clear, altostratus and low stratus cloud conditions were measured by GPS sounding, wind profiler radar, and micro-pulse lidar during the atmospheric radiation measurement experiment from Sep. to Dec. 2008 in Shouxian, Anhui, China. Results showed that during daytime or nighttime, regardless of cloud conditions, the GPS sounding was the most accurate method for measuring BLH. Unfortunately, because of the long time gap between launchings, sounding data did not capture the diurnal evolution of the BLH. Thus, wind profile radar emerged as a promising instrument for direct and continuous measurement of the mixing height during the daytime, accurately determining BLH using the structure parameter of the electromagnetic refractive index. However, during nighttime, radar was limited by weak signal extraction and did not work well for determining the BLH of the stable boundary layer, often recording the BLH of the residual layer. While micro-pulse lidar recorded the evolution of BLH, it overestimated the BLH of the stable boundary layer. This method also failed to work under cloudy conditions because of the influence of water vapor. Future work needs to develop a method to determine BLH that combines the complimentary features of all three algorithms.

Influence of heterogeneous refractivity on optical wave propagation in coastal environments

Spatial variations of refractivity significantly dictate the characteristics of optical wave propagation through the atmosphere. Consequently, the ability to simulate such propagation is highly dependent upon the accurate characterization of refractivity along the propagation path. Unfortunately, the scarcity of high spatiotemporal resolution observational data has forced many past studies of optical wave propagation to assume horizontally homogeneous (HH) atmospheric conditions. However, the (adverse) impact of such an assumption has not been quantified in the literature. In this paper, we attempt to fill this void by utilizing a mesoscale modeling-based approach to explicitly simulate atmospheric refraction. We then compare the differences of the HH refractivity fields to the mesoscale model-derived refractivity fields by means of a realistic atmospheric event and through ray tracing simulations. In this study, we model a coastal low-level jet, a common coastal atmospheric phenomenon which is associated with heterogeneous thermal and refractivity fields. Observational data from a radiosonde and a radar wind profiler near the northeastern region of the United States are used for model validation. The observed characteristics of low-level jet (e.g., evolution, intensity, location) and associated temperature inversion are found to be reasonably well captured by the mesoscale model. The simulated nighttime refractivity gradient field manifests significant spatial heterogeneity; over land, the refractivity gradient is much stronger and amplified near the ground, whereas it becomes much weaker over the ocean. We quantify the effect of this heterogeneity on optical ray trajectories by simulating a suite of rays and documenting the variability of their altitudes at certain propagation ranges. It is found that the altitude of optical rays may vary tens of meters during a diurnal cycle, and at nighttime the rays may bend downward by more than 150 m at a range of 100 km. We run additional ray tracing simulations using refractivity profiles from a single location and assuming HH refractivity along the propagation path. It is observed that the HH approach yields instantaneous ray bending magnitudes up to 30 % less than the ray bending based on the refractivity simulated by the mesoscale model. At the same time, it is found that the mesoscale model-based refractivity fields may have uncertainty introduced by different factors associated with the model configuration. Of these factors, turbulence parameterization is explored in-depth and found to be responsible for more uncertainty than spatial grid resolution. To be more specific, different turbulence parameterizations are found to produce significantly varying temperature inversion parameters (e.g., height, magnitude), which are critical factors influencing ray trajectories. Collectively, these results highlight the potential advantages and disadvantages of utilizing a mesoscale model to simulate refractivity in coastal areas as opposed to assuming HH refractivity.

Wavelet transform‐based methods for removal of ground clutter and denoising the radar wind profiler data

Various non-atmospheric signals contaminate radar wind profiler (RWP) data, which produce bias in estimation of moments and wind velocity. Especially, in ultra high frequency (UHF) RWPs, ground clutter severely degrades wind velocity estimation. Furthermore, at higher altitudes, noise dominates the clear air signal. Thus, the important tasks of signal processing in a RWP are (i) to eliminate the clutter signal, (ii) to detect the weak atmospheric signals buried inside the noise and (iii) to improve signal-to-noise ratio. Wavelet analysis is a powerful tool to differentiate the characteristics of the ground clutter and noise from the atmospheric turbulence echo at the time series level. The authors have implemented the signal processing for lower atmospheric wind profiler radar at National Atmospheric Research Laboratory, Gadanki, India, using wavelet transforms. In this study, they present the implementation approach and results. The wavelet-based algorithms use different threshold levels to identify and remove ground clutter and to denoise the data. The obtained results using this method are validated with collocated global positioning system radiosonde data.

An assessment of the performance of a 1.5 μm Doppler lidar for operational vertical wind profiling based on a 1-year trial

Abstract. We present the results of a 1-year quasi-operational testing of the 1.5 μm StreamLine Doppler lidar developed by Halo Photonics from 2 October 2012 to 2 October 2013. The system was configured to continuously perform a velocity-azimuth display scan pattern using 24 azimuthal directions with a constant beam elevation angle of 75°. Radial wind estimates were selected using a rather conservative signal-to-noise ratio based threshold of −18.2 dB (0.015). A 30 min average profile of the wind vector was calculated based on the assumption of a horizontally homogeneous wind field through a Moore–Penrose pseudoinverse of the overdetermined linear system. A strategy for the quality control of the retrieved wind vector components is outlined for ensuring consistency between the Doppler lidar wind products and the inherent assumptions employed in the wind vector retrieval. Quality-controlled lidar measurements were compared with independent reference data from a collocated operational 482 MHz radar wind profiler running in a four-beam Doppler beam swinging mode and winds from operational radiosonde measurements. The intercomparison results reveal a particularly good agreement between the Doppler lidar and the radar wind profiler, with root mean square errors ranging between 0.5 and 0.7 m s−1 for wind speed and between 5 and 10° for wind direction. The median of the half-hourly averaged wind speed for the intercomparison data set is 8.2 m s−1, with a lower quartile of 5.4 m s−1 and an upper quartile of 11.6 m s−1.

A Multiscale Overview of the El Reno, Oklahoma, Tornadic Supercell of 31 May 2013

Abstract On 31 May 2013 a broad, intense, cyclonic tornado and a narrower, weaker companion anticyclonic tornado formed in a supercell in central Oklahoma. This paper discusses the synoptic- and mesoscale environment in which the parent storm formed, based on data from the operational network of surface stations, rawinsondes, and WSR-88D radars, and from the Oklahoma Mesonet, a Doppler radar wind profiler, Rapid Refresh (RAP) analyses, and photographs. It also documents the overall behavior of the tornadoes and their relationships to features in their parent supercell based on data from a nearby, rapid-scan, polarimetric, mobile Doppler radar. The supercell formed near the intersection of a cold front and a dryline in an environment of moderately strong vertical shear and high CAPE, at the southern end of a line of multicell convective storms. The tornado damage path was as wide as 4.2 km according to the NWS damage assessment and ground-relative Doppler velocities of at least 135 m s−1 were found at the theoretical beam height of <20 m AGL. The tornado debris signature in the copolar cross-correlation coefficient ρhv was as wide as ~4–5 km. After the strong tornado formed, at least one additional cyclonic tornado formed and rotated cyclonically around the main tornado; it was then absorbed by it and the main tornado broadened. Smaller subvortices, which rotated cyclonically around a common axis of rotation, were subsequently observed. The tornado then weakened but remained broad, while the anticyclonic tornado formed to the southeast along the rear-flank gust front.

High-speed solar wind streams and polar mesosphere winter echoes at Troll, Antarctica

Abstract. A small, 54 MHz wind-profiler radar, MARA, was operated at Troll, Antarctica (72° S, 2.5° E), continuously from November 2011 to January 2014, covering two complete Antarctic winters. Despite very low power, MARA observed echoes from heights of 55–80 km (polar mesosphere winter echoes, PMWE) on 60% of all winter days (from March to October). This contrasts with previous reports from radars at high northern latitudes, where PWME have been reported only by very high power radars or during rare periods of unusually high electron density at PMWE heights, such as during solar proton events. Analysis shows that PWME at Troll were not related to solar proton events but were often closely related to the arrival of high-speed solar wind streams (HSS) at the Earth, with PWME appearing at heights as low as 56 km and persisting for up to 15 days following HSS arrival. This demonstrates that HSS effects penetrate directly to below 60 km height in the polar atmosphere. Using local observations of cosmic-noise absorption (CNA), a theoretical ionization/ion-chemistry model and a statistical model of precipitating energetic electrons associated with HSS, the electron density conditions during the HSS events are estimated. We find that PMWE detectability cannot be explained by these variations in electron density and molecular-ion chemistry alone. PWME become detectable at different thresholds depending on solar illumination and height. In darkness, PWME are detected only when the modelled electron density is above a threshold of about 1000 cm−3, and only above 75 km height, where negative ions are few. In daylight, the electron density threshold falls by at least 2 orders of magnitude and PWME are found primarily below 75 km height, even in conditions when a large proportion of negative ions is expected. There is also a strong dawn–dusk asymmetry with PWME detected very rarely during morning twilight but often during evening twilight. This behaviour cannot be explained if PMWE are caused by small-scale structure in the neutral/molecular-ion gas alone but may be explained by the presence of charged meteoric dust.