ALBERT

Description: Cloud radar reflectivity profiles can be an important measurement for the investigation of cloud vertical structure (CVS). However, extracting intended meteorological cloud content from the measurement often demands an effective technique or algorithm that can reduce error and observational uncertainties in the recorded data. In this work, a technique is proposed to identify and separate cloud and non-hydrometeor echoes using the radar Doppler spectral moments profile measurements. The point and volume target-based theoretical radar sensitivity curves are used for removing the receiver noise floor and identified radar echoes are scrutinized according to the signal decorrelation period. Here, it is hypothesized that cloud echoes are observed to be temporally more coherent and homogenous and have a longer correlation period than biota. That can be checked statistically using ∼ 4 s sliding mean and standard deviation value of reflectivity profiles. The above step helps in screen out clouds critically by filtering out the biota. The final important step strives for the retrieval of cloud height. The proposed algorithm potentially identifies cloud height solely through the systematic characterization of Z variability using the local atmospheric vertical structure knowledge besides to the theoretical, statistical and echo tracing tools. Thus, characterization of high-resolution cloud radar reflectivity profile measurements has been done with the theoretical echo sensitivity curves and observed echo statistics for the true cloud height tracking (TEST). TEST showed superior performance in screening out clouds and filtering out isolated insects. TEST constrained with polarimetric measurements was found to be more promising under high-density biota whereas TEST combined with linear depolarization ratio and spectral width perform potentially to filter out biota within the highly turbulent shallow cumulus clouds in the convective boundary layer (CBL). This TEST technique is promisingly simple in realization but powerful in performance due to the flexibility in constraining, identifying and filtering out the biota and screening out the true cloud content, especially the CBL clouds. Therefore, the TEST algorithm is superior for screening out the low-level clouds that are strongly linked to the rainmaking mechanism associated with the Indian Summer Monsoon region's CVS.

Description: One of the key parameters that must be included in the analysis of atmospheric constituents (gases and particles) and clouds is the vertical structure of the atmosphere. Therefore high-resolution vertical profile observations of the atmospheric targets are required for both theoretical and practical evaluation and as inputs to increase accuracy of atmospheric models. Cloud radar reflectivity profiles can be an important measurement for the investigation of cloud vertical structure in a resourceful way. However, extracting intended meteorological cloud content from the overall measurement often demands an effective technique or algorithm that can reduce error and observational uncertainties in the recorded data. In this work a technique is proposed to identify and separate cloud and non-hydrometeor returns from a cloud radar measurements. Firstly the observed cloud reflectivity profile must be evaluated against the theoretical radar sensitivity curves. This step helps to determine the range of receiver noise floor above which it can be identified as signal or an atmospheric echo. However it should be noted that the signal above the noise floor may be contaminated by the air-borne non-meteorological targets such as insects, birds, or airplanes. The second step in this analysis statistically reviews the continual radar echoes to determine the signal de-correlation period. Cloud echoes are observed to be temporally more coherent, homogenous and have a longer de-correlation period than insects and noise. This step critically helps in separating the clouds from insects and noise which show shorter de-correlation periods. The above two steps ensure the identification and removal of non-hydrometeor contributions from the cloud radar reflectivity profile which can then be used for inferring unbiased vertical cloud structure. However these two steps are insufficient for recovering the weakly echoing cloud boundaries associated with the sharp reduction in cloud droplet size and concentrations. In the final step in order to obtain intact cloud height information, identified cloud echo peak(s) needs to be backtracked along the either sides on the reflectivity profile till its value falls close to the mean noise floor. The proposed algorithm potentially identify cloud height solely through the characterization of high resolution cloud radar reflectivity measurements with the theoretical echo sensitivity curves and observed echo statistics for the cloud tracking (TEST). This technique is found to be more robust in identifying and filtering out the contributions due to insects and noise which may contaminate a cloud reflectivity profile. With this algorithm it is possible to improve monsoon tropical cloud characterization using cloud radar.