Attenuation correction effects on quantitative precipitation estimation (QPE) at Sidoarjo weather radar

Abstract

Weather radar waves experience attenuation due to meteorological objects in their path or referred to as wave attenuation. Attenuation can cause a large enough error to reflect the weather radar as input data for the calculation of QPE through the Z-R relation which eventually also causes an error in the QPE value. Attenuation can occur along the wave propagation path that contains meteorological objects, so that the farther from the weather radar location, the higher the potential for QPE error caused as the consequence of unapplied attenuation correction. Weather radar applies a local Z-R relation model where the backscatter of meteorological objects received by the radar is processed into a reflectivity value (Z) and then processed through an empirical equation to produce an estimated rainfall (R). There has been no research on attenuation correction using the Z-R relation model which was developed in Indonesia to determine the effect of attenuation correction from QPE in Sidoarjo weather radar. This research is intended to apply the attenuation correction preprocessing based on Kraemer Verworn preprocessing algorithm and Z-R relation model processing algorithm, Arida et al., developed in Indonesia for the QPE calculation considering the radius from the weather radar location. QPE that is incorrected for attenuation will be used as a comparison in this study to determine the differences that occur before and after the attenuation correction is applied. Furthermore, validation is carried out with rainfall observations from rain gauges through correlation analysis and RMSE. The results show that rainfall without attenuation correction tends to be of lower value, especially at greater distances from the weather radar. A higher correlation with a lower RMSE is generated by QPE with attenuation-corrected raw data input and at the relatively close distance to the weather radar.