Weak Radar Echo Research Articles

A PRECISE RESPIRATORY AND HEART RATE DETECTION METHOD FOR MILLIMETER-WAVE RADAR

Millimeter wave radar as a type of noncontact sensor can more conveniently and insensibly obtain breathing and heartbeat signals. However, due to the weak radar echo signal and the influence of noise, there may be phase ambiguity and separation distortion of breathing and heartbeat signals during the processing, which affects the accuracy of detection. In this paper, an enhanced extended differential and cross multiplication (EDACM) algorithm is presented to obtain chest position signals without phase ambiguity. A specific order finite impulse response (FIR) filter with a Blackman window is applied to extract breath signals without waveform distortion from phase signals, and the heartbeat signal without rate distortion is extracted from the phase signal by means of the designed Chebyshev II filter. The measured results show that the proposed method achieves 96.78% accuracy for heart rate detection and 94.44% accuracy for respiratory rate detection, and improves the accuracy of heart rate detection by 2.22% and respiratory rate detection by 4.06% compared to the band-pass filter method under the same parameter conditions.

E2DTF: An End-to-End Detection and Tracking Framework for Multiple Micro-UAVs With FMCW-MIMO Radar

Due to the weak radar echoes and strong background clutters in low-altitude airspace, the detection and tracking for multiple micro-unmanned aerial vehicles (UAVs) have posed formidable challenges in radar surveillance field. Consequently, this paper proposes an end-to-end detection and tracking framework (E <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> DTF) for multiple micro-UAVs by utilizing the frequency modulated continuous wave-multiple input multiple output (FMCW-MIMO) radar. To address the low signal-to-noise ratio (SNR) problem, E <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> DTF presents a frame-range-Doppler-azimuth information fusion filter to integrate the target energy by exploiting the spatio-temporal dependence of positions within a sequence of unthresholded frames. Additionally, considering that a target may enter/leave the radar field-of-view (FOV), E <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> DTF introduces a target model state, updated by an extended Markov state transition matrix sequentially, to realize an unknown, time-varying number of micro-UAVs tracking. Another nice feature of E <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> DTF is that it avoids the complex data association procedure thanks to removing the threshold-decision operation. Finally, both numerical simulations and experiments with five different scenarios, i.e., horizontal line, cross-trajectory, circular loop, rainy condition and 3D trajectory tracking are presented to verify the effectiveness of the proposed method. The results show that E <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> DTF can obtain superior detection and tracking performance for multiple micro-UAVs in contrast to the state-of-the-art methods considering detection and tracking processes independently, especially under low SNR conditions.

A modeling study of cloud physical properties of extreme and non-extreme precipitation in landfalling typhoons over China

Extreme rainfall associated with landfalling tropical cyclones (ERLTC) can bring severe disaster losses. Revealing the typical characteristics and formation mechanisms of ERLTC has important scientific significance and potential application value. In this study, cloud microphysical properties of extreme and non-extreme rainfall in three landfalling TCs over China under the similar climatic backgrounds are comparatively investigated based on the cloud resolving simulations with a novel quantitative separating method of cold (warm) rain from the total precipitation. Results show that the warm (cold) rain may shape (modulate) the area (intensity) of TC precipitation. During the extreme precipitation period of the strongest TC Lekima (2019), the amount of cold rain accounts for >3/4 of the total precipitation, which plays a leading role in the precipitation rapid enhancement. With the increase of precipitation intensity, the proportion of cold (warm) rain to total precipitation tends to increase (decrease). The area near eyewall is commonly dominated by warm rain due to the weak radar echoes there. During the extreme precipitation process induced by Lekima (2019), the hydrometeor contents (mainly rain water and snow) increase rapidly and maintain for a long time. The processes of water vapor condensation, accretion of cloud water by rain water and melting of snow are the most active. In addition, the comparison with Hagupit (2020) and Yagi (2018) reveals that the processes of raindrop collecting graupel then melting, and the weak evaporation of raindrop make considerable contributions to the formation of extreme precipitation caused by Lekima (2019).

Analysis of possibility to improve the civil marine radar efficiency during navigation in the Arctic waters

The paper presents the civil marine radar (CMR) efficiency in the polar waters in the case of advanced signal processing based on the optimal filtering of the voltage coming to the intermediate frequency amplifier. The structure of the radar subsystem was modified by adding two complementary filter units that allow to transform the filter frequency response into the almost sinc-function curve providing much better detection of weak radar echoes. The studies have shown that the updated radar matched filter makes it possible for CMR to observe objects with relatively small scattering cross section at the considerably longer distances than up-to-date CMR. Also it is explained that this radar signal processing doesn’t require the increasing of CMR cost which can also be considered as advantage of the method.

Deep learning‐based unmanned aerial vehicle detection in the low altitude clutter background

Unmanned aerial vehicles (UAVs), widely used due to their low cost and versatility, pose security and privacy threats, which calls for their reliable recognition at low altitudes. However, strong ground clutter and multipath effects severely interfere with the weak radar echoes reflected off the micro-UAVs, resulting in severe degradation of recognition reliability. Based on channel modelling and UAV recognisability analysis, a time-frequency transform-aided contrastive learning model is proposed to suppress the severe ground clutter and reliably recognise low-altitude UAVs. In the proposed framework, a time-frequency transform unit is first applied to suppress the multipath-induced ambiguity effect and ease the semantic feature extraction via Zhao-Atlas-Marks transform and morphological operation. Thereafter, a contrastive-learning-based feature extraction and fusion unit is established to suppress non-target clutter interference and extract recognisable semantic UAV features. Finally, a gated recurrent unit-based classifier is designed for UAV recognition. Sufficient experiments are carried out on both real and simulated data sets, and the comparative results verify that the proposed model outperforms the mainstream algorithms and improves the detection accuracy by more than 5% under severe ground clutter interference.

CNGAT: A Graph Neural Network Model for Radar Quantitative Precipitation Estimation

Radar quantitative precipitation estimation (RQPE) is the most common measurement for area rainfall estimation with high spatial and temporal resolution. The radar reflectivity ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Z$ </tex-math></inline-formula> ) measured by the Doppler weather radar is strongly related to precipitation rate ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R$ </tex-math></inline-formula> ). However, conventional RQPE methods have limited capability of modeling the complex relationship between the radar echoes and the precipitation field. In this article, we propose a graph neural network (GNN)-based RQPE model named categorical node graph attention network (CNGAT) to model complex spatial–temporal features of precipitation field reflected by the radar echo field. CNGAT is derived from graph attention networks (GAT) utilizing attention mechanism to learn the importance of neighboring points to the central point, which is beneficial for learning varying local spatial patterns. Furthermore, CNGAT can handle multiple types of graph nodes by using different transform functions for different types of nodes, which makes it better at capturing diverse features of precipitation field indicated by strong and weak radar echo areas. The proposed model was trained and tested on radar network and rain gauge data distributed in East China during 2017 and 2018. The results of several experiments show that CNGAT greatly improves the estimation precision and detection rate than <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Z$ </tex-math></inline-formula> – <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R$ </tex-math></inline-formula> relation models and conventional data-driven RQPE methods, and alleviates under-estimation of higher precipitation rates, which validates that CNGAT can effectively represent complex spatial–temporal features of precipitation field.

Charge Regions Indicated by LMA Lightning Flashes in Hokuriku's Winter Thunderstorms

AbstractThe charge distribution of some cells in three winter thunderstorms in the Hokuriku region of Japan is investigated based on Lightning Mapping Array (LMA) flash data. The vertical arrangements of charge regions involved in lightning discharges suggest diverse charge patterns, including quad‐polar, tripole, positive dipole, inverted dipole, and inverted tripole. The riming electrification between graupel and ice crystals or their aggregations are thought to be responsible for the electrification of most cells. The charging process between snow/aggregates and ice crystals may be responsible for some inverted charge structure that occurred above 0 °C isotherm and accompanied with weak radar echoes. Convection indicated by the vertical development of radar reflectivity appears crucial to shaping the diverse charge distribution patterns by determining which charging mechanisms occur and where; it also influences changes in height or even the disappearance of the charge regions. The charged cores are distributed from 0.7‐ to 5.3‐km heights and 2‐ to –31‐ °C temperatures, while the distances between adjacent charged cores with opposite polarities change between 0.2 and 3.4 km, with a mean of 1.3 km. The mean flash duration and horizontal distance are 425.0 ms and 19.8 km, respectively. The average height, temperature, and power of flash initiations are 2.8 km, −11.9 °C, and 15.6 dBW, respectively.

Characteristics of Upward Lightning on the Beijing 325 m Meteorology Tower and Corresponding Thunderstorm Conditions

AbstractComprehensive observation on lightning striking the 325 m meteorology tower has been conducted in Beijing, China, during the summer season from 2012 to 2016. For the 20 tower flashes captured by high‐speed video camera, 19 (95%) cases were initiated from the tower (termed as upward lightning (UL)), with duration of 42–844 ms. Most ULs (84.2%, 16/19) were triggered by nearby lightning, of which positive cloud‐to‐ground flashes (+CGs) account for 87.5% (14/16). Analysis on the characteristics of the causal or triggering lightning flashes suggests that the approaching of negative leader process through in‐cloud horizontal channel is the vital condition for the initiation of upward leader. Generally, the UL initiation follows the nearby discharges, while for a special case in our data set, the nearby positive return stroke was found to occur after the establishment of an upward channel from the tower. It is worth noting that this +CG led to a significant enhancement of the tower‐initiating leader, with eight subsequent return strokes and two M component processes occurring to the tower thereafter. The radar echo of the corresponding thunderstorm indicates that the other‐triggered UL (OTUL) tended to occur at the dissipation stage of the thunderstorm with relatively low cloud top height and weak radar echo, while the self‐initiating UL (SIUL) was more likely to occur when the thunderstorm was in the mature stage and the tower was underneath the stratiform clouds behind the convective region. The meteorology condition for SIUL involved lower ambient temperature and higher wind speed, as compared to the OTUL.

On the linkage of daytime 150 km echoes and abnormal intermediate layer traces over Sanya

AbstractThe daytime 150 km echoes, which are associated with the upper E region field‐aligned irregularities, have been observed around the equatorial electrojet region but never before at magnetic latitudes near the northern anomaly crest region. We present first results of daytime 150 km echo over Sanya, China, a station located far away from the dip equator in the Northern Hemisphere. A layer of weak radar echoes with spectral width less than 10 m s−1 was seen in the height range from 145 km at 12:15 LT to 152 km at 13:45 LT on 21 July 2010. The interesting aspect is that the rare observation of daytime 150 km echoes with the small Sanya VHF radar was preceded by the occurrence of an unusual intermediate layer, which is identified as abnormal traces at the upper E region in corresponding ionograms. The abnormal intermediate layer associated with possible gravity wave activity (that implicates the presence of upper E region density gradients) could make a significant contribution to the growth of irregularities responsible for the rarely detected daytime 150 km echoes over Sanya.

Electrical structure of a Qinghai–Tibet Plateau thunderstorm based on three-dimensional lightning mapping

An isolated thunderstorm that occurred on August 6, 2009, in Datong, Qinghai province, China, was studied using a three-dimensional (3D) lightning mapping system distributed at seven stations. This thunderstorm exhibited a dominant positive electric field (corresponding to an overhead positive charge) at the ground. The 3D mapping results show that the thunderstorm took on an inverted dipole charge structure during its developing and mature stages from 09:08 UTC to 09:35 UTC when positive electric field was dominant. During the thunderstorm dissipating stage from 09:35 UTC to 09:58 UTC, the charge structure changed into four layers of positive, negative, positive and negative charges with heights of 5km (height from the ground), 4km, 3km and 1.8km, respectively. A comparison between the location of the lightning radiation sources and the radar echo shows that the radiation sources of the negative charge region corresponded to the region with radar reflectivity about 40dBZ in both the developing and mature stages, while for the positive charge radiation sources, they were located in the region with relative weak radar echo in the developing stage but in the mature stage they corresponded to the region with radar reflectivity greater than the 40dBZ.

On the observation of unusual high concentration of small chain-like aggregate ice crystals and large ice water contents near the top of a deep convective cloud during the CIRCLE-2 experiment

Abstract. During the CIRCLE-2 experiment carried out over Western Europe in May 2007, combined in situ and remote sensing observations allowed to describe microphysical and optical properties near-top of an overshooting convective cloud (11 080 m/−58 °C). The airborne measurements were performed with the DLR Falcon aircraft specially equipped with a unique set of instruments for the extensive in situ cloud measurements of microphysical and optical properties (Polar Nephelometer, FSSP-300, Cloud Particle Imager and PMS 2-D-C) and nadir looking remote sensing observations (DLR WALES Lidar). Quasi-simultaneous space observations from MSG/SEVIRI, CALIPSO/CALIOP-WFC-IIR and CloudSat/CPR combined with airborne RASTA radar reflectivity from the French Falcon aircraft flying above the DLR Falcon depict very well convective cells which overshoot by up to 600 m the tropopause level. Unusual high values of the concentration of small ice particles, extinction, ice water content (up to 70 cm−3, 30 km−1 and 0.5 g m−3, respectively) are experienced. The mean effective diameter and the maximum particle size are 43 μm and about 300 μm, respectively. This very dense cloud causes a strong attenuation of the WALES and CALIOP lidar returns. The SEVIRI retrieved parameters confirm the occurrence of small ice crystals at the top of the convective cell. Smooth and featureless phase functions with asymmetry factors of 0.776 indicate fairly uniform optical properties. Due to small ice crystals the power-law relationship between ice water content (IWC) and radar reflectivity appears to be very different from those usually found in cirrus and anvil clouds. For a given equivalent reflectivity factor, IWCs are significantly larger for the overshooting cell than for the cirrus. Assuming the same prevalent microphysical properties over the depth of the overshooting cell, RASTA reflectivity profiles scaled into ice water content show that retrieved IWC up to 1 g m−3 may be observed near the cloud top. Extrapolating the relationship for stronger convective clouds with similar ice particles, IWC up to 5 g m−3 could be experienced with reflectivity factors no larger than about 20 dBZ. This means that for similar situations, indication of rather weak radar echo does not necessarily warn the occurrence of high ice water content carried by small ice crystals. All along the cloud penetration the shape of the ice crystals is dominated by chain-like aggregates of frozen droplets. Our results confirm previous observations that the chains of ice crystals are found in a continental deep convective systems which are known generally to generate intense electric fields causing efficient ice particle aggregation processes. Vigorous updrafts could lift supercooled droplets which are frozen extremely rapidly by homogeneous nucleation near the −37 °C level, producing therefore high concentrations of very small ice particles at upper altitudes. They are sufficient to deplete the water vapour and suppress further nucleation as confirmed by humidity measurements. These observations address scientific issues related to the microphysical properties and structure of deep convective clouds and confirm that particles smaller than 50 μm may control the radiative properties in convective-related clouds. These unusual observations may also provide some possible insights regarding engineering issues related to the failure of jet engines commonly used on commercial aircraft during flights through areas of high ice water content. However, large uncertainties of the measured and derived parameters limit our observations.

A climatological analysis of heatbursts in Oklahoma (1994–2009)

AbstractHeatbursts are characterized by a sudden and highly localized increase in air temperature, a simultaneous decrease in relative humidity and dewpoint temperature, and strong gusty winds, typically associated with decaying thunderstorms. The small spatial extent and short duration of most heatbursts makes detailed study of these events difficult using the standard federal observation network (e.g. hourly observations) established in most countries. Thus, many discussions of heatbursts note that they are ‘rare’ phenomena. However, observations from the Oklahoma Mesonet indicate that although heatbursts are meso‐alpha scale phenomena, they are not rare. Using multi‐criteria analysis, 207 heatburst events of various magnitudes, areal coverage and duration were identified between 1994 and 2009 across Oklahoma. Significantly more heatbursts were detected in the western two‐thirds of the state as compared to the eastern third. Heatbursts were primarily nocturnal events, with 70% of heatburst events initiated between 0000 and 0800 UTC (i.e. between 6 p.m. and 2 a.m. Central Standard Time). With 62 of the 207 events, June was the most active month for heatbursts. At the other extreme, no heatbursts were detected during February and November, and only one was observed during January and December. Almost all of the heatbursts occurred underneath or adjacent to weak radar echoes at the time of the event. Four basic radar patterns were associated with the events: (1) radar echoes associated with rapidly weakening convection, (2) weak reflectivity not produced by or associated with deep moist convection, (3) weak radar echoes to the rear of a dissipating mesoscale convective system (MCS) and (4) weak reflectivity along the periphery of intense convection that did not appear to be weakening. Copyright © 2010 Royal Meteorological Society

“Warm Rain” in the Tropics: Seasonal and Regional Distributions Based on 9 yr of TRMM Data

Abstract How much precipitation is contributed by warm rain systems over the tropics? What is the typical size, intensity, and echo top of warm rain events observed by the Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar over different regions of the tropics? What proportion of warm raining areas is actually attached to the edges of cold systems? Are there mesoscale warm raining systems, and if so, where and when do they occur? To answer these questions, a 9-yr TRMM precipitation feature database is used in this study. First, warm rain features in 20°S–20°N are selected by specifying precipitation features 1) with minimum infrared brightness temperature &amp;gt; 0°C, 2) with TRMM Precipitation Radar (PR) echo top below freezing level, or 3) without any ice-scattering signature in the microwave observations, respectively. Then, the geographical, seasonal, and diurnal variations of the rain volume inside warm rain features defined in these three ways are presented. The characteristics of warm rain features are summarized. Raining pixels with cloud-top temperature above 0°C contribute 20% of the rainfall over tropical oceans and 7.5% over tropical land. However, about half of the warm pixels over oceans and two-thirds of the warm pixels over land are attached to cold precipitation systems. A large amount of warm rainfall occurs over oceans near windward coasts during winter. Most of the warm rain systems have small size &amp;lt; 100 km2 and weak radar echo with a modal maximum near-surface reflectivity around 23 dBZ. However, mesoscale warm rain systems with strong radar echoes do occur in large regions of the tropical oceans, more during the nighttime than during daytime. Though the mean height of the warm precipitation features over oceans is lower than that over land, there is no significant regional difference in its size and intensity.

The characteristics of cloud-to-ground lightning activity in hailstorms over northern China

The characteristics of cloud-to-ground (CG) lightning for ten hailstorms in Shandong Province of China were analyzed statistically. It is found that the hailstorms in this study present dominant positive CG flashes during periods of falling hail. One specific hailstorm on 16 June 2006 was studied in detail using the data from a CG lightning location network, Doppler radar and cloud images. Comparison between the brightness temperature of cloud-top and CG flash locations indicated that most flashes occurred in the region with temperatures lower than − 40 °C, while dense positive CG flashes occurred in the range between − 40 °C and − 50 °C. Negative CG flashes occurred mostly in the relative weak radar echo region, and positive CG flashes were distributed in the strong echo region especially with a large gradient of echo intensity. CG flashes tended to occur in the cloud region with reflectivity between 25 dBZ and 35 dBZ. Comparison between the wind field retrieved from Doppler radar and the location of CG flashes indicated that the flashes were located in the convergent region at lower to middle levels.

Innovative Signal Utilization and Processing

Shrouded behind jargon and understood only by a small community, “signal processing” in radar meteorology seems to have a certain mystique associated with it. Its roots are embedded in the branches of mathematics and engineering that deal with the processing of time series data, in our case from received radar echoes. These weak radar echoes contain not only returns from weather (e.g., precipitation), but also a variety of artifacts that we wish were not present (e.g., ground clutter) and distract us from our primary goal of measuring precipitation rate, wind speed, turbulence, or any variety of more specialized atmospheric characteristics. Minimally, we wish to measure the intensity or power of the signal (the moment zero of the Doppler spectrum), its mean Doppler frequency (the first moment), and its variability in velocity (the second central moment) within many thousands of range cells per second.

Satellite observations of convective storm tops in the 1.6, 3.7 and 3.9 μm spectral bands

Spatial and temporal characteristics of convective storm tops observed in the 1.6, 3.7 or 3.9 μm and visible satellite spectral bands were examined. National Oceanic and Atmospheric Administration (NOAA)/Advanced Very High Resolution Radiometer (AVHRR) observations over Europe during the 1980s have shown that some storms exhibit a significant increase in the 3.7 μm cloud top reflectivity. Subsequent NOAA/AVHRR observations have shown that similar cloud top phenomena can be found as well for convective storms over the US Great Plains. The launch of the Geostationary Operational Environmental Satellite (GOES)-8, 9 and 10 has enabled the study of the evolution of storm cloud top structures in a similar band (3.9 μm) with high temporal resolution. This imagery shows that the smallest areas of increased 3.9 μm reflectivity (within or above storm tops) appear and fade on the scale of a few minutes, although larger ones can persist for tens of minutes to several hours. Occasionally, cloud top structures resembling plumes have been observed above some of the storms, apparently emanating from cores of these. Selected cases of convective storms exhibiting an increase in the 3.9 μm reflectivity have been studied with respect to internal storm structure as observed by NEXRAD Doppler radars. This revealed that the spots or areas with increased 3.9 μm reflectivity were typically found above relatively weak radar echo regions, though close to storm cores. However, a few of these “spots” have appeared above a mesocyclone near the time of associated tornado touchdown, suggesting that these spots might be the result of relatively small ice crystals present near the top of strong updrafts. One case of high 3.9 μm reflectivity over an entire storm top has been recorded simultaneously by GOES-8 and GOES-9 on 22–23 May 1996. Some aspects of bidirectional scattering are evident from the differences in 3.9 μm reflectivity observed from these two satellites. Finally, the appearance of storm tops is compared from observations in the 3.7 or 3.9 μm bands with those in the AVHRR/3 1.6 μm band, which has been recently implemented on NOAA-KLM polar orbiting satellites.

Mesoanalysis of the Big Thompson Storm

Mesoscale analyses and descriptions of meteorological conditions that produced the devastating dash flood in the Big Thompson Canyon on 31 July 1976 are presented. The storm developed when strong low-level easterly winds to the rear of a polar front pushed a moist, conditionally unstable air mass upslope into the Front Range of the Rocky Mountains. The main thrust of the moisture flux focused initially into the Big Thompson area. Orographic uplift released the convective instability, and light south-southeasterly winds at steering levels allowed the storm complex to remain nearly stationary over the foothills. Heavy rains fell within the storm along a narrow corridor only 5 km wide oriented north-northeast by south-southwest. Minimal entrainment of relatively moist air at middle and upper levels, very low cloud bases and a slightly tilted, updraft structure contributed to a high precipitation efficiency. A deep warm layer of convective cloud fostered precipitation growth through warm cloud processes. The greatest concentrations of precipitation size particles remained at low elevations and as a result of poor vertical beam resolution returned anomalously weak radar echoes to a WSR-57 located 110 n mi away.

Surface slope probabilities from the spectra of weak radar echoes: Application to Mars

Slope probability densities were derived from the power spectra of radar echoes from Mars using integral inversion. The inverse problem is ill-posed; that is, small changes in the data can lead to large changes in the solution. We describe a method of stabilizing the inversion, which was necessary for echoes with signal-to-noise power spectral densities on the order of unity, and for those with broad spectral distributions. The resulting slope probabilities usually consisted of a component due to quasi-specular reflection which decreased rapidly with tilt, plus a broad, slowly decreasing, “diffuse” component due to scattering from (1) surface scales small compared with a radar wavelength, or (2) larger features with high slopes. In the absence of more complete polarization measurements, we are unable to In the absence of more complete polarization measurements, we are unable to distinguish between these possibilities. Root mean square tilts have been determined separately for the two cases. For case (1), values of rms tilt associated with surface features responsible for the quasi-specular echo are normally less than 3°; for case (2), values greater than 8° are common. Knowledge of the depolarization of radar returns would help distinguish between these possibilities.