Document
15Ice-0051
Initial Results from Radiometer and Polarimetric Radar-based
Icing Algorithms Compared to in situ Data
David Serke
National Center for Atmospheric Research
Andrew L. Reehorst and Michael C. King
NASA Glenn Research Center
ABSTRACT
In early 2015, a field campaign was conducted at the NASA INSTRUMENTATION AND ALGORITHMS
Glenn Research Center in Cleveland, Ohio, USA. The purpose of the campaign is to test several prototype algorithms meant to detect the NIRSS location and severity of in-flight icing (or icing aloft, as opposed to ground icing) within the terminal airspace. Terminal airspace for this project is NASA and the National Center for Atmospheric Research currently defined as within 25 kilometers horizontal distance of the have been developing NIRSS since 2003 (Fig. 1)[4]. The system terminal, which in this instance is Hopkins International Airport in employs a elevation and azimuth scanning multi-channel radiometer, Cleveland.
built by Radiometrics Corporation, which passively collects incoming Two new and improved algorithms that utilize ground-based microwave radiation at a number of channels in the K and V-bands of remote sensing instrumentation have been developed and were operated the electromagnetic spectrum [5]. The K-band lies within an during the field campaign. The first is the 'NASA Icing Remote Sensing atmospheric water vapor resonance feature and thus variations at System', or NIRSS. The second algorithm is the 'Radar Icing Algorithm', specific frequencies within the band are primarily caused by or RadIA. In addition to these algorithms, which were derived from variations in the amount of liquid and gaseous water. The V-band is ground-based remote sensors, in-situ icing measurements of the profiles of on the shoulder of an atmospheric oxygen resonance, so that supercooled liquid water (SLW) collected with vibrating wire sondes progressively varying frequencies from the peak of the absorption attached to weather balloons produced a comprehensive database for feature yields information on the atmospheric temperature further comparison. Key fields from the SLW-sondes include air temperature, from the radiometer (in range).
humidity and liquid water content, cataloged by time and 3-D location.
This work gives an overview of the NIRSS and RadIA products and results are compared to in-situ SLW-sonde data from one icing case study. The location and quantity of supercooled liquid as measured by the insitu probes provide a measure of the utility of these prototype hazard- sensing algorithms.
INTRODUCTION
When subfreezing air becomes supersaturated with respect to water and ice, ice crystals begin to grow on freezing nuclei such as mineral dust, aerosols, other pollution particles or existing ice crystals by the process of diffusion [1]. In the absence of significant populations of such ice nuclei, liquid water drops begin to condense out of the air from the water vapor. In these situations, supercooled liquid water (SLW) drops can form, which can result in aircraft icing. In-flight icing occurs when an aircraft impacts with these SLW drops, which instantly freeze to the airframe. The resulting ice accretion builds up on the leading edges of the airframe’s surfaces. A supercooled large drops (SLD) environment is defined as having a spectrum with a maximum drop diameter larger than 100 μm and can accrete well beyond the leading edges of the airframe Figure 1 – Components of NIRSS, shown when they where ice protection systems typically exist. Ice buildup while in flight were in Platteville, CO.
acts to increase drag and reduce the aerodynamic lift generated by a vehicle’s control surfaces. Significant ice buildup can degrade an Algorithms used within the radiometer’s software use aircraft’s flight characteristics enough to be a significant safety hazard and neural networks trained on large historical archives of integrated has been recognized as a contributing factor to many crashes with liquid water (ILW), integrated water vapor (IWV) and temperature resulting loss of life.
profiles combined with the measured radiometric brightness No single instrument has yet been developed which can temperatures to generate real-time profiles of ILW, IWV and unambiguously detect in-flight icing conditions remotely. For this reason, temperature. NIRSS ingests the radiometer profiles, a Ka-band combinations of sensors have been under development for some time to cloud radar profile built by Metek Corporation and a cloudbase detect in-flight icing [2,3]. This study will focus on methods to utilize two height measurement from a laser ceilometer into a ‘fusion’ machine, ground-based remote sensing platforms with the purpose of providing then combines the instrument fields into an in-flight icing product.
accurate and timely warnings on in-flight icing hazard. Each instrument o The height range of the 0 and -20 Celsius isotherms are targeted as platform or source of comparison data are described in detail in the the area where in-flight icing is most likely to exist from previous following paragraphs .
research flight campaigns. The vertical extent of cloud boundaries are provided by the ceilometer and K-band cloud radar. If cloud exists NEXRADS were designed to detect [12]. This means that if a radar within the height range where icing temperatures exist, any liquid sensed was viewing a sample volume that contained one large snowflake and by the radiometer is then distributed vertically with fuzzy logic based on a million much smaller SLW drops, the returned power to the radar previous experience with years of research flights in icing conditions [6]. receiver would be dominated by the single large snowflake. This A recent study found that NIRSS detects the absence, presence and fact, along with knowledge that the dual-polarimetric characteristics severity of in-flight icing at least as well as the FAA’s current operational somewhat overlap for liquid and various generalized ice crystal system for icing detection [7] in the profile viewed by NIRSS. habits, limits the radar’s capacity to directly detect SLW drops.
This technology has recently been extended to provide In the past several years, research flights into small drop, volumetric coverage of an airport terminal environment [8 ] . Building on large drop and mixed-phase icing have shown a possible way around the existing vertical pointing system, the new method for providing these seemingly insurmountable shortcomings in point-by-point radar volumetric coverage utilizes a vertical pointing cloud radar, a multi- moment analysis for icing detection. RadIA [13] uses polarimetric frequency microwave radiometer with azimuth and elevation scanning moment fields from operational S-band weather radars plus a capabilities, and a NEXRAD radar. In addition to the vertical profiles temperature profile from a numerical weather prediction model as provided by NIRSS, the new volumetric product adds slant-angle inputs. The height of the freezing level is first determined from the measurements from the radiometer to derive icing parameters along the polarimetric moment data and then this height is used to make airport's runway headings within 12.5 km of the airport center. Also, adjustments to the model temperature profile. Non-meteorological features in the NEXRAD reflectivity field are tracked with time to provide targets and radar return at non-freezing heights in the volume are an advection vector with which to laterally move the NIRSS icing hazard masked out. Fuzzy logic membership functions are then defined, profiles. Advected hazard values are then compared to predefined based on the work in the following references, to determine the volumes of airspace in the approach paths for each runway heading as a presence of freezing drizzle ('FRZDRZ' algorithm in RadIA)[14 ] , final hazard product for ranges of 12.5 to 25 km from the airport center. homogeneous supercooled liquid drops that are smaller than drizzle- This new terminal area icing remote sensing system can return sized ('SLW' algorithm in RadIA)[15 ] , mixed phase conditions temperature, estimates of liquid water content [6] and mean cloud drop ('MIXPHA' algorithm in RadIA)[16 ] and the existence of plate- size [9] as well as a qualitative icing hazard determination for each point shaped crystal species ('PLATES' algorithm in RadIA)[ 16] . The in the terminal airspace. PLATES algorithm identifies regions of crystals whose major axis dimension is much larger than their minor axis, similar to a dinner plate. These crystals, either dendrites or solid plates, tend to fall with RadIA their major axis preferentially oriented horizontally, resulting in The national network of Doppler radars, known as NEXRADs positive Zdr values. These RadIA-specific categorical labels for ('NEXt-generation RADars', or WSR-88Ds), were originally designed to particle species are referenced for the rest of this work and are not to detect the presence of precipitation-sized particles, such as snow, rain and be confused with traditional definitions for large (freezing rain, hail (Figure 2). These operational radars were recently upgraded to have freezing drizzle) and small (cloud drop size) in-flight icing. Merging both vertical and horizontal transmit and receive capabilities, known as RadIA drop detection parameters to traditional definitions of 'dual-polarimetric'. Dual-polarimetric radars collect additional microphysical particle species would require a research aircraft flight information that can give researchers insight as to the mean particle shape, campaign, which so far has not materialized. The component size, phase (liquid or solid), bulk density and preferred particle orientation weighting functions for SLW, FRZDRZ, MIXPHA and PLATES are in a given sampled volume. Polarimetric radars collect differential shown in Figure 3 and range from a value of 0 for 'no icing interest' reflectivity (Zdr), correlation coefficient ( HV ) and specific differential ρ phase ( DP ) . Differential reflectivity is the ratio of horizontal co-polar Φ received power to the vertical co-polar return. One way to interpret this field is a power-weighted mean axis ratio of the particles in the volume.
Cloud drops are nearly spherical, so Zdr values are near zero and tend to have small reflectivities. Larger drops become somewhat oblate as they fall and thus have Zdr values between +0.3 and +2.0 dB. Oriented ice crystals also have Zdrs that are positive and large. HV is the correlation ρ between horizontal and vertical co-polar received returns. Wet or tumbling particles and mixed phase conditions result in decorrelation in polarizations so that HV values are typically below 0.92. Homogeneous ρ rain or ice result in HV values above 0.95.
ρ Detection of SLW with particle identification algorithms [10] was previously attempted with relatively poor results [11]. This was true because the cross-sectional dimension of SLW drops are often several factors of ten smaller than the precipitation particle categories that the Figure 3– Polarimetric moment weighting functions for the component algorithms of RadIA Figure 2 – NEXRAD antenna with radome to 1 for 'maximum icing interest'. The y-axis indicates the field in capital letters (example: DBZ is reflectivity) and the prefix letter of 'm' refers to 'mean' or 's' refers to 'standard deviation'. The resulting interest values of each of these four calculations are combined through rule-based thresholding to identify areas of the radar volume that are deemed to have a high, medium or low interest in the presence of these three previously mentioned forms of in-flight icing and to define no icing interest when plate crystals are present.
SLW-sondes A new vibrating wire sonde has been under development by Anasphere, Inc. for the past several years (Figure 4). The sondes have recently been flown in known icing conditions and compared to output from ground-based radiometers [17] as well as an in-flight icing research Figure 5 – Location of field campaign instrumentation.
flight [18]. A separate presentation at this c onference (no available paper) The NASA hangar is 20 x 76 meters.
detailing recent developments with the SLW-sondes is given by Michael King from NASA Glen Research Center.
o 238 and 281 azimuth, which is repeated every 5 minutes. In the absence of expensive icing flights flown with specially outfitted research aircraft, verification and validation of the presence of in- flight icing were provided by pilot reports and by the balloon-borne SLW-sondes. PIREP times and icing severities were used only as an indicator of the presence of in-flight icing conditions within the study area.
nd Case Study: January 22 , 2015 nd On January 22 , 2015, a short-wave trough moved through the Cleveland area. The short wave was associated with a persistent low pressure system located over the Hudson Bay region (not shown). The local surface temperatures were -4.4 °C with dew point -1 temperatures around -6.6 °C. Winds were around 2 ms from the northwest, shifting to the southwest at the time of the sonde release.
The local automated weather reported station was reporting light snow, fog and mist changing to light snow at the surface at the time of the sonde release. Several positive icing PIREPs were reported in the Cleveland terminal airspace from 13:14 to 15:15 UTC on this date, including two 'moderate' icing severities. The PIREPs around the time of the sonde indicated icing between 670 and 2100 m AGL.
A SLW-sonde was released from the NASA Glen Reseach Center Figure 4 – SLW-sonde. Wire vibration is orthogonal to the grounds at 14:17 UTC. The thermodynamic profile (Figure 6, part A) relative airflow.
shows a vertically complex cloud that is formed from multiple layers merging to create one continuous cloud from 0.2 to 2.3 km above Pilot Reports ground level (AGL), with temperatures in the prime icing range of -6 C near cloud base to -12 C at cloud top. The vibrational frequency of Pilot Reports (PIREPs) are voluntary reports made by the wire (part B) is seen to start decreasing from a starting value near commercial airline pilots of the time and location of meteorological 44.5 Hz as it gets into the cloud layer and levels out near cloud top at conditions that their aircraft has encountered. In-flight icing is one of a value near 43.5 Hz. The total decrease in frequency through the many possible conditions that can be reported. The existence of icing can cloud layer was approximately 1 Hz. LWC results are estimated from be reported as 'none' or 'icing exists' as ‘trace’, ‘light’, ‘moderate’, ‘severe’ the change in frequency per unit height per unit time [17] and the or ‘heavy’ severity. The authors converted these qualitative conventions to derived profile is shown in Figure 6, part C. Temperature inversions a 0-8 scale of severity for quantitative comparisons. Any icing PIREPs in a saturated environment at 0.8 and 1.4 km AGL (black horizontal within 50km of Cleveland-Hopkins Airport were used as qualitative lines) are colocated with relative or absolute maximums in SLW- comparisons of the existence of icing for the two ground-based sonde LWC. The temperature inversion at 2.3 km corresponds with algorithms. the height of cloud top and SLW-sonde LWC extinction. The presence of these temperature inversions within the cloud layer, even very small ones, indicate the presence of atmospheric lifting and
2015 WINTER FIELD CAMPAIGN
subsequent saturation of discrete layers within the cloud depth.
-3 Maximum LWC of 0.35 gm occurs at 0.8 km AGL and then rapidly
OVERVIEW
reduces to near zero at 1 km AGL. The top half of the cloud has a -3 relative maximum of 0.15 gm just above 1.2 km AGL before the LWC gradually is reduced back to zero by the cloud top height of 2.3 To address the goals of this study, the authors operated NIRSS km.
and RadIA at NASA Glenn Research Facility in Cleveland, Ohio Reflectivity (Figure 7) and differential reflectivity (Figure throughout the winter of 2015. Positions of the respective instrumentation 8) plots from the polarimetric Cleveland NEXRAD at the 3.5 degree for this field campaign are shown in Figure 5. The field instrumentation elevation 'tilt' are shown, as they are key ingredients to RadIA. Only are within 200 meters of each other. The radiometer scan volume includes o weak reflectivities in the range of -20 to 0 dBZ are evident, with zenith and 10 elevation scans along the four runway headings at 58, 101, Figure 6 – Profiles of temperature (red, A) and dewpoint temperature (blue, A), sonde wire frequency (blue dots, B) and LWC (C ) for the balloon launch at 14:17 UTC on January 22nd, 2015.
o Figure 8 – 3.5 elevation NEXRAD tilt Zdr [dB] nd at 14:01 UTC on January 22 , 2015.
o Figure 7 – 3.5 elevation NEXRAD tilt reflectivity[dBZ] nd of plate-shaped crystals in RadIA's PLATES algorithm is likely due at 14:01 UTC on January 22 , 2015.
to the rapid growth of needles, dendritic or hexagonal crystals, as discussed in reference 16. Dendrites grow in water saturated detectable returns terminating at around 37 km range at this tilt angle.
environments at the expense of the supercooled liquid detected near This range corresponds to the highest precipitation sized particles existing the top of the cloud by RadIA's FRZDRZ algorithm. The overall up to about 2.1 km AGL. The corresponding Zdr plot has very high values hazard output of RadIA is somewhat speckled in nature due to the in the top half of the cloud, on the order of +4 to +7 dB, which indicates removal of ground clutter pixels by the particle identification the presence of highly oriented, axially asymmetric crystals. Near zero to algorithm in an earlier processing step. No smoothing of any kind slightly negative Zdr values in the lower third of the cloud indicate the has been done on the RadIA output.
dominance of round particles.
The heights of the dominant algorithm-detected icing RadIA output is shown in Figure 9. The colorbar shows where conditions, as defined by RadIA and delineated by the rings in Figure positive icing is identified by specific algorithms within RadIA. The 9, are plotted in a vertical format on the right side of Figure 10. Icing salmon color is icing due to drops smaller than drizzle (RadIA's 'SLW'), detected by RadIA's cloud-sized SLW drop algorithm dominates from brown is freezing drizzle and larger (RadIA's 'FRZDRZ', abbreviated the surface up to roughly the height of the maximum SLW-sonde 'FRZ'), yellow is both SLW and FRZDRZ, white are plate-shaped crystals LWC value at 0.8 km AGL. Above that, freezing drizzle dominates (RadIA's 'PLATES'), dark grey is when all algorithms show no icing and up to around 1.1 km, where sonde LWC values drop significantly light grey is when RadIA has no classification. Horizontal range rings back toward zero. Above 1.1 km AGL, sonde LWC reaches a relative were added to the figure, within which certain RadIA icing hazard maximum before eroding away significantly with height as the cloud classifications are seen to dominate. SLW is prevalent within about 10 km top is approached. In this portion of the cloud, freezing drizzle range of the radar, FRZDRZ dominates from 10 to 18 km range and pockets still exist in RadIA, but large, continuous areas dominated by patches of FRZDRZ separated by areas of PLATES from 18 to 37 km plate-shaped crystals are interspersed. Perhaps the crystals are range. The development through time of the spatially continuous regions o Figure 9 – 3.5 elevation tilt RadIA classification of the NEXRAD Figure 11– NIRSS volumetric product at 14:15 UTC nd nd on January 22 , 2015.
moments at 14:01 UTC on January 22 , 2015.
also shown on Figure 10. It is important to note that the collection efficiency of the SLW-sonde's vibrating wire, and thus the derived LWC profile, depends on drop size [17]. The sonde values are not really from a vertical profile, rather they are a slant path defined by the ascent rate and horizontal winds experienced by the balloon. The NIRSS LWC profile is computed by distributing the radiometer- sensed integrated liquid through the depth of the Ka-band sensed cloud layer through a fuzzy-logic algorithm that relies on idealized relationships between temperature, reflectivity and the assumption of wedge-shaped profiles. Significant differences exist in these two estimated LWC profiles, such as the height of the absolute maximum in LWC. This is due to NIRSS having a more wedge-shaped output due to the dominant weighting of an idealized wedge-shaped profile in the NIRSS logic. The overall magnitude and height bounds of the NIRSS-derived LWC is comparable to that derived from the sonde.
Differences in these two LWC products can be partially attributed to o the radiometer channels sampling 4-6 angle viewing cones [5] whereas the sondes are essentially time-varying point measurements.
The terminal area NIRSS product, shown in Figure 11, is the result of a recent development effort to extend the ground-based remote sensing aviation hazard detection to the runway approach vectors. In this plot, the maximum NIRSS hazard values within the range-segmented, FAA-defined glide paths for commercial aircraft operating in the Cleveland airspace are color-coded for 'none', 'trace', 'light', 'light/moderate', 'moderate', 'moderate/heavy' and 'heavy' icing severity categories. The box which includes the immediate terminal area is assigned the maximum hazard level given in the profile as determined by the vertically pointing NIRSS (In this case 'light' severity, colored blue). The hazard value for the four boxes along each runway heading within 12.5 km (6.7 nautical miles) of the -3 Figure 10 – SLW sonde (blue) and NIRSS (black) LWC [gm ] with terminal are assigned the maximum NIRSS hazard within each box RadIA classification profile.
volume defined by the 10 degree slant angle radiometer ILW distributed over the given vertical cloud profile. Finally, the hazard responsible for the erosion of the LWC values as one approaches the upper value for boxes along each runway heading between 12.5 and 25 km levels of the cloud. Ice crystal growth scavenging existing SLW is a (6.7 and 13.5 nautical miles) are defined by maximum NIRSS hazard commonly observed and well understood path to the erosion of in-flight within each box volume as defined by the closest slant angle NIRSS icing conditions.
hazard profile as advected by radar detected features. This is the first The derived LWC 'profile' from the sonde (blue line) and the case that the terminal area NIRSS product has been applied to, and derived LWC profile from the colocated NIRSS platform (black line) are the magnitude of the resulting icing in each approach volume seems at 6. Reehorst, A., Politovich, M., Zednik, S., Isaac, G. and least reasonable given the SLW-sonde results that have been discussed in Cober, S., “Progress in the development of practical remote previous figures. detection of icing conditions”, NASA/TM 2006-214242 , NASA, 2006.
SUMMARY
7. Johnston, C., Serke, D., Adriaansen, D., Reehorst, A., Politovich, M.,Wolff, C. and McDonough, F., “Comparison During this past winter, an in-flight icing hazard detection field of in-situ, model and ground based in-flight icing severity”, campaign was conducted in Cleveland, Ohio, USA where output from two AMS Conference Preprint , Seattle, WA, Jan 24-27, 2011.
ground-based remote sensing platforms/algorithms were compared to SLW profiles collected from vibrating wire sondes attached to weather 8. Reehorst, A. and Serke, D., “A Terminal Area Icing Remote balloons. This work explores the results from one case study conducted nd Sensing System”, Report NASA/TM 2014-218417, 2014.
on January 22 , 2015 which had several in-flight icing reports by commercial pilots. The polarimetric radar-based RadIA detected SLW 9. Serke, D., Reehorst, A., and Politovich, M. K., from cloudbase up to the height at which the SLW-sonde detected a -3 “ Supercooled large drop detection with NASA's Icing maximum LWC of 0.35 gm . Above the maximum liquid height, RadIA Remote Sensing System" , SPIE Remote Sensing Preprint, detected freezing drizzle/SLD. In the top half of the cloud layer, LWC Sept. 19-22, Toulouse, FR, 2010.
appeared to erode away toward zero with height. RadIA suggests a [DOI:10.1117/12.863176].
possible cause of that is an increase with time of preferentially-oriented plate-shaped crystals, either dendrites, hexagonal crystals or needles, 10. Vivekanandan, J., Zrnic, D., Ellis, S., Oye, R., Ryzhkov, A.
which are known to grow at the expense of supercooled liquid. This high Zdr 'brightband' , which in this case is not associated with a freezing and Straka, J., “Cloud microphysics retrieval sing S-band level, From this and other cases collected during the January to April of dual-polarization radar measurements”, Bulletin of the American Meteor. Soc ., 80 , 381-388, 1999.
2015 period (not shown here, see https://wiki.ucar.edu/display/TAIWIN/), RadIA has shown skill in identifying the location of icing within 11. Kimberly L. Elmore, 2011: The NSSL Hydrometeor NEXRAD radar echo through texture and absolute values of the areally Classification Algorithm in Winter Surface Precipitation: averaged polarimetric moments when compared to SLW-sonde flights.
The NIRSS-derived LWC profile was wedge-shaped with a Evaluation and Future Development. Wea. Forecasting , 26 , 756–765. [doi: 10.1175/WAF-D-10-05011.1] maximum in the upper half of the cloud, but the overall magnitude of the LWC maximum and ILW was quite comparable to the SLW-sonde. When the significant differences with which LWC was measured are considered, 12. Bringi, V. and Chandrasekar, V., “Polarimetric Doppler Weather Radar – Principles and applications.”, Cambridge the differences in shape and magnitude of LWC from NIRSS seem quite University Press , 656 pp., 2001.
reasonable. A first look at an airport terminal area NIRSS product was provided and discussed. The new product provides a two-dimensional 13. Serke, D., Scott Ellis, John Hubbert, David Albo, output of a three-dimensional product for range-segmented airspace Christopher Johnston, Charlie Coy, Dan Adriaanson and volumes centered along FAA mandated runway approach vectors. The product gave reasonable qualitative 'light' icing severity values for this Marcia Politovich, In-flight icing hazard detection with dual and single-polarimetric moments from operational case, considering that the segmented approach airspace volumes (shown 2- NEXRADs , AMS Radar, September 16-20, Breckinridge, dimensionally in Figure 11) were all within the lowest 1 km AGL of the NIRSS LWC profile (Figure 10). CO, 2013.
Future work includes the archiving, quality control and analysis for all field campaign datasets that were collected during the January to 14. Ikeda, K., Rasmussen, R., Brandes, E. and McDonough, F., March of 2015 case study period. An extensive analysis catalog should be “Freezing Drizzle Detection with WSR-88D Radars”, J.
Appl. Meteor. Climatol. , 48 , 41–60, 2009. [doi: available by early 2016. Comparison to numerical weather prediction model output will also be conducted as part of the overall final analysis. 10.1175/2008JAMC1939.1] 15. Plummer, D., Göke, S., Rauber, R. and Di Girolamo,
REFERENCES
L.,“Discrimination of mixed- versus ice-phase clouds using dual-polarization radar with application to detection of 1. Rogers, R., and Yau, M., “A Short Course in Cloud Physics”, aircraft icing regions”, J. of Appl. Meteor. and Clim. , 49 , Elsevier Science, Oxford, UK, 1988.
Issue 5, pp. 920-936, 2010. [doi: 10.1175/2009JAMC2267.1] 2. Politovich, M.K., B.B. Stankov and B.E. Martner, “Determination of liquid water altitudes using combined remote 16. Williams, E., D. Smalley, M. Donovan, R. Hallowell, K.
sensors”, J. Appl. Meteor ., 34 , pp. 2060—2075, 1995.
Hood, B. Bennett, R.Evaristo, A. Stepanek, T. Bals-Elsholz, J. Cobb, J. Ritzman, A. Korolev, and M.Wolde, 2014: 3. Bernstein, B., McDonough, F., Politovich, M., Brown, B., Measurements of Differential Reflectivity in Snowstorms Ratvasky, T., Miller, D., Wolff, C., and Cunning, G., “Current and Warm Season Stratiform Systems. J. Appl. Meteor.
Icing Potential: algorithm description and comparison to aircraft Climatol. [doi:10.1175/JAMC-D-14-0020.1], in press.
observations”, J. Appl. Meteor ., 44 , pp. 969-986, 2005.
17. Serke, D., Hall, E., Bognar, J., Jordan, A., Abdo, S., Seitel, 4. Reehorst, A.L., Brinker, D.J., Ratvasky, T.P., “NASA Icing K., Nelson, M., Ware, R., McDonough, F. and Marcia Remote Sensing System: comparisons from AIRS–II,” Politovich, “ Supercooled liquid water content profiling NASA/TM—2005-213592, 2005. case studies with a new vibrating wire sonde compared to a ground-based microwave radiometer”, Atmospheric th Research, June 6 , 2014. [DOI: 5. Solheim, F., Godwin, J., Westwater, E., Han, Y., Keihm, S., 10.1016/j.atmosres.2014.05.026] Marsh, K., and Ware, R., “Radiometric profiling of temperature, water vapor and cloud liquid water using various inversion 18. Serke, D., Doyle, R., King, M., Geerts, B., Steiger, S. and methods”, Radio Sci. , 33 , pp. 393-404, 1998.
Politovich, M., A new SLW-sonde compared to research aircraft th flights, AMS 17 Symposium on Meteorological Observation and Instrumentation, June 9-13, Westminster, CO, 2014.
CONTACT INFORMATION
David Serke, serke@ucar.edu , (303)-497-8311, 3450 Mitchell Lane, Boulder, CO, 80301
ACKNOWLEDGMENTS
This work was done under the NASA Transformative Aeronautics Concepts Program and the NASA Aviation Safety Research Program . The National Center for Atmospheric Research is sponsored by the National Science Foundation. Any opinions, findings and conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect views of the National Science Foundation.
DEFINITIONS/ABBREVIATIONS
FAA - Federal Aviation Administration FRZDRZ - RadIA's algorithm for detecting at least as large as freezing drizzle ILW - integrated liquid water MIXPHA - RadIA's algorithm for detection of mixed-phase NASA - National Aeronautical and Space Administration NCAR - National Center for Atmospheric Research NEXRAD - Next-Generation Radar, or WSR-88d radar NIRSS - NASA Icing Remote Sensing System NSF - National Science Foundation PLATES - RadIA's plate-shaped crystal detection algorithm PIREP - Pilot report RadIA - Radar Icing Algorithm REFL - Reflectivity ρ - cross-polar correlation coefficient HV SLD - supercooled large drops SLW - generally Supercooled Liquid Water , also RadIA's algorithm for detection of drops smaller than freezing drizzle Zdr - differential reflectivity