Estimating additional flood warning time using raindrops intensity data captured by x-band mp radar

Abstract

The current monitoring system of flood warnings on Mt. Merapi slope, Yogyakarta, Indonesia relies on rainfall gages and visual observation. Most flood events occurred with a fast-moving front resulting in a very short available warning time. An X-Band Multi-Parameter Radar installed at the southwest slope of Mt. Merapi measures raindrop intensity high above the ground and provides rainfall information earlier. This paper presents the effort to estimate the additional warning time available by extracting binary radar data, transforming it into 3D grid data, and visualize it into meaningful charts. This study compared the IDW and NN interpolation methods and verified the grid resolution and the limiting distance, r, of the IDW method for obtaining optimal performance. This study also conducted a qualitative review on the sequences of horizontal contour images and the comparison of horizontal and vertical grid resolutions. The estimation of time lag of peak raindrop intensity of different elevations was evaluated qualitatively by overlaying time-series data at four elevations and quantitatively by conducting a cross-correlation analysis of pairs of time series data of +3000 m and +1300 m elevations with horizontal offset for considering wind drift. The result shows that the optimal value of r is 400 m, and the optimal grid resolution is 100x100x100 m. It found that the approximate additional warning time was 6 minutes. The conclusion is that the developed interpolation method is reliable for raindrop analysis. The cross-correlation analysis gives a conservative estimate of the additional warning time.