![](https://thumbs.gfycat.com/BlushingJubilantImperialeagle-size_restricted.gif)
The artificial neural network (ANN) regression has been applied to the radar reflectivity data to estimate monsoon rainfall using parametric Z-R models. The 10-min reflectivity data recorded in Kota Bahru radar station (in Malaysia) and hourly rain record in nearby 58 gauge stations during 2013–2015 were used.
![](https://people.utm.my/nadzri/files/2021/12/Study_Area_Hani-1024x777.png)
![](https://people.utm.my/nadzri/files/2021/12/3D_Hani.png)
The three-dimensional nearest neighbor interpolation with altitude correction was applied for pixel matching. The non-linear Levenberg Marquardt (LM) regression, integrated with ANN regression minimized the spatiotemporal variability of the proposed Z-R model.
3D NN model is independent to the complexity of radar coefficients (alpha & beta) thus minimizing the geometric bias. Highest rain = 3 mm/h & Low rain = 0.3 mm/h (accuracy). Massive flood episode in Kelantan (2014), 91-hour continuous downpour and ~20 mm/h (highest) measured by weather radar.
![](https://people.utm.my/nadzri/files/2021/12/PPI_Hani-1024x555.png)