If you've been around weather enthusiast communities long enough, you remember what Weather Underground used to be.
Before the redesigns and the ownership changes and the feature cuts, Wunderground was where the serious amateurs lived. Not the casual seven-day-checkers. The people who ran personal weather stations in their backyards, wrote their own forecast discussions, argued about snow ratios on the forums, and treated a blizzard forecast like an intellectual problem worth solving in public.
That community is gone now. The platform that housed it stripped out most of the features that made it worth visiting. The people who built their presence there are scattered.
This is what happened, where the community went, and what it means for independent weather forecasting.
What Weather Underground Actually Was
Weather Underground was founded in 1995 by Jeff Masters and two other University of Michigan students — originally a project to provide internet weather data during the internet's early commercial years. It grew into something much more interesting than a weather data site.
By the 2000s, WU had built several distinct community layers that didn't exist anywhere else:
Personal weather stations. WU created and maintained one of the world's largest networks of citizen-run weather stations — personal observation points in backyards, farms, and rooftops that contributed hyperlocal data. At peak, over 250,000 stations were feeding real-time observations into the network. This was genuinely valuable science infrastructure, not just a hobby project.
Forecaster communities. The WU forums had dedicated spaces for forecast discussion — regional severe weather threads, hurricane track debates, winter storm forecasting discussions. These weren't casual chats. They were substantive meteorological discussions, often involving people with real analytical skill who didn't work for NWS or a TV station but knew what they were talking about.
Forecast blogs. WU gave users the ability to publish their own forecast blogs — written analysis pieces tied to their location and specialty. Some of these blogs developed real followings. The good ones were worth reading.
The Weather Underground name itself had a certain credibility that consumer apps never earned. If you told someone you ran a weather blog on WU, they understood it wasn't just a hobbyist vanity project.
The Ownership Chain That Killed It
The decline of WU's community is a fairly standard story of successive corporate ownership prioritizing data infrastructure over community, then stripping even that down.
2012: The Weather Channel acquisition. Jeff Masters sold Weather Underground to The Weather Channel. At the time, this seemed like it might provide stability and resources. In practice, TWC was a media company that valued the WU station network data more than the community around it.
2016: IBM acquisition. IBM acquired The Weather Channel's digital properties (including WU) as part of its Watson cloud infrastructure play. WU's citizen weather station network became a data feed for IBM's enterprise weather APIs. The community features were tolerated but no longer prioritized. Redesigns began optimizing for consumer audiences, not enthusiasts.
2024: Francisco Partners acquisition. Private equity acquired WU from IBM in early 2024. At this point, the end was visible to anyone paying attention. PE acquisition of media/community properties rarely ends with community investment.
2025: The cuts. By mid-2025, WU had removed hourly, daily, and weekly forecast products entirely. The webcam network was discontinued. Forum activity had already been declining for years, but the forecast removals eliminated the core utility for forecasters who were still using the platform. Customer service went dark. The site became, essentially, a current-conditions display with a map of weather stations — valuable data infrastructure, but no longer a community.
Jeff Masters had already seen the writing on the wall. He left WU in 2022 and moved to Yale Climate Connections. The founding vision didn't survive the ownership changes.
Where the Community Actually Went
The WU diaspora didn't go to one place. It scattered across several communities and platforms, and that fragmentation is itself part of the story.
Weather.gov and the NWS ecosystem. Some station operators and weather enthusiasts migrated toward NWS-adjacent resources and communities. Less social, more data-focused.
MesoWest / Synoptic Data. For personal weather station operators specifically, MesoWest became a significant aggregator of citizen observation data. It doesn't have a community layer, but the data quality is high.
Windy. The wind/precipitation visualization platform has developed a community feature set that some ex-WU users adopted. It's primarily focused on data visualization rather than original forecasting.
Niche forums (Stormtrack, American Weather). The severe weather and chaser communities that overlapped with WU had their own forums that predate and postdate WU's relevance. Stormtrack remains active for convective forecasting discussions.
Twitter/X and social media. The largest migration was probably just to social. Forecasters who had WU blogs moved to Twitter, where the weather community is genuinely large and active. The tradeoff: you're on a platform that could change its rules or algorithm at any time, and your content isn't organized around your forecast record in any systematic way.
YouTube. Storm chasing and weather event coverage found a natural home on YouTube. Some of the indie forecasters who cut their teeth in WU communities built significant YouTube audiences during the severe weather uptick of the early 2020s.
Substack. For the essay-and-analysis end of weather writing, some former WU bloggers found a home on Substack. The subscriber model suits long-form weather writing well.
What WU Didn't Replace
Here's the thing the fragmentation obscures: the specific function WU served — a place where you could publish structured forecasts, build a local following, and be known for the quality of your predictions — doesn't really exist in any of these successor platforms.
Twitter is for broadcast. YouTube is for video. Substack is for newsletters. None of them have a concept of a structured forecast — a prediction with geographic zones, accumulation ranges, and a timestamp that can later be compared against what actually happened.
The ex-WU forecaster community went everywhere and nowhere specific at the same time. The people who cared about weather accuracy as a craft lost the only platform that had even a rough approximation of tools for them.
That's the gap ForecasterHQ is built to fill — not as a weather station network, but as a place for independent forecasters to publish structured predictions, build a verified track record, and find an audience. If you're coming from WU or the communities adjacent to it, the discover page is where you can see what other forecasters are publishing.
And if you're ready to publish your own: Start publishing — it's free →
Already read about the platforms and tools for independent forecasters? Our weather underground alternative post covers the product landscape for displaced WU users more directly. If you're a personal weather station owner specifically looking for where to upload your station data, see weather underground PWS alternatives for station owners. If you were a WU forecaster specifically — not just an observer — Weather Underground alternatives for independent weather forecasters covers where that audience has moved and what a dedicated forecasting workflow looks like now. If you're thinking about building an independent forecasting practice from scratch, How to Become an Independent Weather Forecaster covers the full path.