Is Gumfreston flooded?
Is Gumfreston flooded? is a local flood awareness tool for the B4318 at Gumfreston.
It combines live environmental data with anonymous community reports to answer one simple question:
“Is the road likely to be flooded right now?”
The system uses a risk-based approach rather than a simple yes-or-no rule.
It looks at river levels, recent rainfall, tide conditions and short-term forecasts, then checks whether local reports support or contradict that picture.
From this, the situation is grouped into one of three outcomes:
NO — flooding unlikely
MAYBE — conditions worth paying attention to
YES — flooding likely or ongoing
Community reports help increase confidence when conditions are uncertain, but they never override clear environmental signals.
Flooding at Gumfreston is highly local and influenced by several overlapping factors.
Conditions can change quickly, and it’s often unclear whether the road is passable until you’re already committed. Official warnings don’t always reflect what’s happening at this specific point.
This tool exists to make that uncertainty visible, rather than pretending there’s a single definitive answer.
This tool is not an official warning service and cannot guarantee road conditions at any given moment.
Flooding can change quickly and may vary across short distances.
Always follow local signage and guidance from:
Natural Resources Wales
Traffic Wales
Pembrokeshire County Council
The system combines several independent sources of information.
Live river levels and rainfall data are provided by Natural Resources Wales. Tide height readings come from the Environment Agency gauge at Milford Haven, with daily tide predictions from TideTimes.org.uk. Short-term weather forecasts are supplied by Open-Meteo.
Alongside this, anonymous community reports add local context from people who have recently passed through the area.
No single source determines the result. All data is refreshed automatically and interpreted together.
Is Gumfreston flooded? was designed and built by Flock of Clouds as a lightweight public utility.
It’s an ongoing, locally built experiment, shaped by real-world use and improved over time.
The aim isn’t to predict flooding perfectly, but to communicate uncertainty clearly – explaining why conditions look calm or risky without overstating confidence.