شناسایی نوفه‌های موجود در رادارهای داپلر هواشناسی بوشهر و حذف آنها

The word Radar from the term Radio Detection and Ranging means target radio tracking and ranging, an electromagnetic system that is used to detect and determine the position and speed of a target. Most meteorological radars are of Pulse-Doppler type, which have the ability to detect the movement of raindrops and the intensity of precipitation. Both types of this information can be analyzed to determine the structure of the storm and its potential for producing severe weather. Also, Doppler type radars measure rainfall and wind direction and speed at certain times and in a wide area. These types of radars display the position of the target and reflectivity with the help of a computer by combining colors on the display screens (Reinhart, 1997). Ground echo is a challenge for meteorological radar data analysis, especially in hydrology and precipitation estimation. Therefore, the removal of ground clutter is a prerequisite for the use of meteorological radar for both quantitative and qualitative purposes. An important part in the quality control of radar data is the removal of clutter.
Several researches have been conducted to provide different algorithms for removing ground clutter and abnormal echoes emitted by radar data (Gabela and Notarpietro, 2002; Layton et al., 2013; Zheng et al., 2016; Kaman et al., 2018; Proft et al., 2019; Adaba et al., 2021; Xie et al., 2021; Wang et al., 2022; Yan et al., 2020).
The aim of the current research is to present a simple and fast algorithm by selecting the most appropriate thresholds to remove ground clutter and unusual echoes presented in radar reflection data, so that it has the least impact on showing meteorological phenomena.
In the present study, the effectiveness of removing ground clutter as well as determining the best thresholds has been quantitatively evaluated on the meteorological Doppler radar of Bushehr, located in the southwest of Iran.
This algorithm has been used in rainy condition (squall line event On March 19, 2017 at 04:00UTC) using the Bushehr Doppler Weather Radar data which includes spatial-proximity filter and compactness test. The spatial-continuity filter that is applied to each pixel eliminates data that are weakly and spatially correlated to the surrounding. Therefore, when the difference between each pixel and the surrounding pixels is less than a certain threshold , its value is assumed to be a meteorological echo. The compactness test identifies adjacent pixels of not null intensity; thus, choosing a threshold level  slightly greater than one, will eliminate most of the unwanted clutter. To determine the best threshold,  for 3 to 8 values and  between 1.1 to 2.6 are examined.
Removal of ground noises in Bushehr weather Doppler radar reflection data equal to 3, 4, 5, 6, 7 and 8 and  with a threshold of 1.1 for 04UTC on March 19, 2017 showed that the difference in does not make a significant change in the removal of clutter.
Due to the fact that the change in  does not have much change in the removal of clutter, so in the following, is equal to 8 and the change in  is examined. The results showed that the present algorithm has a significant ability to remove the ground clutter from the Doppler radar reflectivity data. Varying the  values from 3 to 8 did not have a significant effect on noise removal. This is likely due to the saturation of  after reaching a certain threshold, which may result from the homogeneity of the area's texture (coastal region). Also, in the examined thresholds,  equal to 8 and  equal to 1.8 has the best result by removing most ground clutters and unusual echoes in the radar reflectivity data, as well as preserving storm convective cells.