Study Of Method For Radar Data Quality Control With New Generation Weather Radar

Doppler velocity information can be used for removal of ground clutter, extractionof wind profiles, detection of shear zones, and construction for dual Doppler wind fields.Operational application of Doppler velocity data from weather radars is hampered bythe infamous limitation of the range-velocity ambiguity. The dual pulse repetitionfrequency (DPRF) is a commonly used technique, especially in C Band new generationweather radars that extends the unambiguous Doppler velocity. However, the region ofchaos radial velocity that contained opposite sign velocity values with large differencesoften occurs in velocity data which obtained from C Band new generation weather radarrunning in dual PRF mode. It affected radar data quality severely. The characteristics ofthe chaos radial velocity are analyzed in this paper, and an algorithm for identifies andcorrects the chaos radial velocity has been developed and examined by the volumescanning radar data from Haerbin and Mudanjiang radars in the summer of2010. At thesame time, this study also aims at analyzing the detection results of algorithm aboutradar data quality in Doppler weather radar3-D mosaic system which developed by theChinese Academy of Meteorological Sciences, presents a new automatic algorithm forisolated reflectivity echo and radial interference echo detection, and improves two-stepapproach for ground clutter detection. All of these new algorithms are based on themethod of identify storm segments in SCIT algorithm, and examined by the volumescanning radar data with different type of echoes from Wuhan radar between2004and2011.The main conclusions are gotten as follows.(1) The signal to noise ratio has definite effects on local velocity change. We foundthe signal to noise ratio is smaller, the local velocity change is bigger, the probability ofchaos radial velocity occurred in velocity data is greater, and vise versa.(2) In the region of chaos radial velocity, the right velocity value is related to theunambiguous Doppler velocity value and the error velocity value. Subtracting the unambiguous Doppler velocity value from the error velocity value when the errorvelocity value is greater than zero, or adding the unambiguous Doppler velocity value tothe error velocity value when the error velocity value is smaller than zero, then we couldget the right velocity value.(3) The algorithm uses characteristics of velocity difference that in small region toidentify the region with big local velocity change, and turns out two characteristicparameters V8, absData to distinguish the chaos radial velocity from the margin zone ofthe velocity aliases well. The V8represents the abrupt changes in region, and theabsData shows the average velocity differences in small region within the interval of0to2×(Nyquist velocity).(4) On the basis of image continuity, the average velocity value with maximumratio between the positive velocity points and the negative velocity points which are notthe chaos radial velocity and are all in the region of K neighborhood can be acted as thenew velocity value. In comparison with original velocity data, the region of chaos radialvelocity has been retrieved obviously, and the quality of velocity data has beenimproved.(5) Isolated reflectivity echo with smaller scale than other meteorological echoesusually exists in reflectivity data. Identifying storm segments in the radial reflectivityand then combining them into2D storm components, the site information of echoes inan elevation can be obtained, and the geometric information such as area value can alsobe calculated. Using the area value instead of the traditional method of filtering fordetection of isolated reflectivity echo not only detects exactly but also conserves theedge of the meteorological echo completely.(6) Radial interference echo can fulfill the whole radial from radar to the maximumdistance of radar detection. It is also exist in one or more orientations. The distributionof tangential width value along the radial orientation is unchanging and very small(almost <=5). So the length of the identifying storm segments along the tangentialorientation of radar is almost a constant, which can used to detect the radial interferenceecho, and separate it from precipitation echoes accurately.(7)After processed by the two-step approach, some remainders of ground clutterwith high reflectivity value also exist in the reflectivity data. This is mainly caused bythe differences of frequency between scanning of reflectivity and scanning of Dopplerinformation at lowest two elevations. The remainders always occur at outside of150kmfrom radar. But the reflectivity value on the edge of the remainders is not in accordancewith the characteristics of precipitation echoes. Therefore, we can make full use of the identification on the edge of the echoes to remove those remainders. In comparison withthe original algorithm, the new algorithms appear to work equally well in situationswhere the three type of non-meteorological echoes are either separated from orembedded within precipitation echoes. Moreover, the precipitation echoes aresuccessfully retained.