Technical note: A simple feedforward artificial neural network for high-temporal-resolution rain event detection using signal attenuation from commercial microwave links

Two simple feedforward neural networks (multilayer perceptrons - MLPs) are trained to detect rainfall events using signal attenuation from commercial microwave links (CMLs) as predictors and high-temporal-resolution reference data as the target. MLP.sub.GA is trained against nearby rain gauges, and...

Full description

Saved in:
Bibliographic Details
Published in:Hydrology and earth system sciences 2024-11, Vol.28 (23), p.5163-5171
Main Authors: Ãydvin, Erlend, Graf, Maximilian, Chwala, Christian, Wolff, Mareile Astrid, Kitterød, Nils-Otto, Nilsen, Vegard
Format: Article
Language:English
Subjects:
Artificial neural networks
Atmospheric precipitations
Datasets
Estimates
Gauges
Meteorological radar
Meteorological stations, Radar
Methods
Microwave attenuation
Microwave communications
Multilayer perceptrons
Neural networks
Precipitation
Radar
Radar data
Radar detection
Rain
Rain and rainfall
Rain gauges
Rainfall
Temporal resolution
Time series
Weather forecasting
Weather radar
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Two simple feedforward neural networks (multilayer perceptrons - MLPs) are trained to detect rainfall events using signal attenuation from commercial microwave links (CMLs) as predictors and high-temporal-resolution reference data as the target. MLP.sub.GA is trained against nearby rain gauges, and MLP.sub.RA is trained against gauge-adjusted weather radar. Both MLPs were trained on 26 CMLs and tested on 843 CMLs, all located within 5 km of a rain gauge. Our results suggest that these MLPs outperform existing methods, effectively capturing the intermittent behaviour of rainfall. This study is the first to use both radar and rain gauges for training and testing CML rainfall detection. While previous studies have mainly focused on hourly reference data, our findings show that it is possible to classify rainy and dry time steps with a higher temporal resolution.
ISSN:1027-5606
1607-7938
DOI:10.5194/hess-28-5163-2024