<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Forecasting | Donghui Xu</title><link>https://xdongh.github.io/tags/forecasting/</link><atom:link href="https://xdongh.github.io/tags/forecasting/index.xml" rel="self" type="application/rss+xml"/><description>Forecasting</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Tue, 19 May 2026 00:00:00 +0000</lastBuildDate><image><url>https://xdongh.github.io/media/icon_hu7729264130191091259.png</url><title>Forecasting</title><link>https://xdongh.github.io/tags/forecasting/</link></image><item><title>Co-authored paper featured as EOS Research Spotlight: Keeping Humans in the Loop Improves Flood Forecasting</title><link>https://xdongh.github.io/post/eos-spotlight-2026-flood-forecasting/</link><pubDate>Tue, 19 May 2026 00:00:00 +0000</pubDate><guid>https://xdongh.github.io/post/eos-spotlight-2026-flood-forecasting/</guid><description>&lt;p>Our paper &lt;a href="https://doi.org/10.1029/2025GL118317" target="_blank" rel="noopener">&lt;strong>&amp;ldquo;The Value of Forecasters-in-the-Loop in Real-Time Flood Forecasting in the Age of Machine Learning&amp;rdquo;&lt;/strong>&lt;/a> was featured as a Research Spotlight on &lt;em>Eos&lt;/em>, the science news magazine of AGU.&lt;/p>
&lt;p>The spotlight highlights how human forecasters using traditional hydrologic models continue to outperform machine learning approaches for flood prediction, particularly for extreme events with long lead times. This work was led by first author &lt;a href="https://www.ornl.gov/staff-profile/vinh-tran" target="_blank" rel="noopener">Vinh Tran&lt;/a> (ORNL).&lt;/p>
&lt;p>Read the spotlight: &lt;a href="https://eos.org/research-spotlights/keeping-humans-in-the-loop-improves-flood-forecasting" target="_blank" rel="noopener">Keeping Humans in the Loop Improves Flood Forecasting&lt;/a>&lt;/p></description></item></channel></rss>