[Paper Review] IowaRain: A Statewide Rain Event Dataset Based on Weather Radars and Quantitative Precipitation Estimation

Muhammed Sit, Bong‐Chul Seo+1 more|arXiv (Cornell University)|Jul 7, 2021

Meteorological Phenomena and SimulationsEarth and Planetary Sciences10 references35 citations

TL;DR

IowaRain is a rainfall event dataset for Iowa (2016–2019) derived from the NEXRAD radar network and IFC QPE system, providing 5-minute, 500 m precipitation maps and 1242 categorized rain events.

ABSTRACT

Effective environmental planning and management to address climate change could be achieved through extensive environmental modeling with machine learning and conventional physical models. In order to develop and improve these models, practitioners and researchers need comprehensive benchmark datasets that are prepared and processed with environmental expertise that they can rely on. This study presents an extensive dataset of rainfall events for the state of Iowa (2016-2019) acquired from the National Weather Service Next Generation Weather Radar (NEXRAD) system and processed by a quantitative precipitation estimation system. The dataset presented in this study could be used for better disaster monitoring, response and recovery by paving the way for both predictive and prescriptive modeling.

Motivation & Objective

• Provide a comprehensive, pre-processed rainfall event dataset for Iowa to support climate change modeling, disaster planning, and data-driven hydrology research.
• Describe end-to-end data acquisition, processing, quality control, and event extraction from NEXRAD-based QPE for Iowa.
• Offer a benchmark resource enabling predictive and prescriptive modeling in environmental science and disaster response.

Proposed method

• Collect raw radar observables from seven WSR-88D radars covering Iowa.
• Process data with the IFC QPE system using attenuation corrections and melting-layer analyses to generate 5-minute, 500 m rain rate maps.
• Classify precipitation into convective and stratiform using melting layer information and elevation-based QPE products.
• Composite 2D rain rate maps from individual elevations and radars into a single Iowa-domain product with spatial/temporal synchronization.
• Serialize and normalize rainfall snapshots into gzip archives, and define rainfall events based on consecutive snapshots meeting a 0.5 mm/hr and 10% domain criteria.

Experimental results

Research questions

• RQ1 How can a statewide, radar-based rainfall event dataset be constructed from multi-radar QPE products?
• RQ2 What are the characteristics (duration, coverage, intensity) of Iowa rainfall events from 2016–2019 using this processing pipeline?
• RQ3 How can the dataset be structured to support data-driven tasks such as frame-prediction and rainfall-runoff research while remaining adaptable for extension?

Key findings

• The IowaRain dataset comprises 1242 rainfall events from 2016–2019.
• Events are extracted from 5-minute, 500 m resolution QPE maps generated by seven WSR-88D radars.
• Each event is stored as a gzip archive containing start date, number of snapshots, and the list of snapshots.
• The dataset includes yearly variations in event counts and durations, with average event lengths around 6–8 hours and monthly coverage statistics provided.
• The data processing pipeline includes quality control, melting-layer analysis, radar data fusion, and normalization steps to produce consistent 2D rain rate maps for the Iowa domain.