What is this? Click to learn more
While Stavrocast predicts future rainfall, Hineycast looks to the rear with one question: how much did it rain? Hineycast provides coupled map and timeseries visualizations that show cumulative and incremental precipitation over time, with data available from recent minutes to previous years.
If you want to see how much it rained but you don’t have rain gauge records handy, then Hineycast is for you. Its primary goal is to help us make faster and easier, data-informed decisions around topics such as:
- Emergency management
- Storm severity assessment (i.e. was this a 100-yr storm?)
- Construction delays for rain days
- Localized erosion vs. localized rainfall
- Local drought severity
- Calibration of hydrologic and hydraulic models
While the gridded rainfall estimates provided by Hineycast are far more reliable than future rainfall forecasts, it’s important to understand that they are still just estimates. Therefore, an overarching goal of this product is to help users intuitively visualize the pros, cons, and use-cases for different types of historic rainfall estimates.
The data sources available on Hineycast are also described below, but I’d like to offer my specific awe and appreciation to Iowa State University, their Department of Agronomy, and the Iowa Environmental Mesonet Reanalysis (IEMRE) products that they publish. The basic features that Hineycast offers are possible only because a handful of brilliant Iowans got the green light to spend their careers cataloging, improving, and serving one of the world’s highest quality continental scale radar archives.
Hineycast examples
EventRecord Drought Date2026 LocationCharlottesville, VA TypeDrought
The chart shows how little it rained in Charlottesville for most of 2026 compared to normal.
EventStorm Severity Assessment DateAugust 2026 LocationCharlottesville, VA TypeFrequency
Record drought in Charlottesville finally broke around August 2, 2026. Here’s a fun pop quiz — what storm frequency does this rain event look like? Sure, it rained a lot that day… but was this a 1-year rainfall event? A 50-year one? 100-year? Compare against NOAA Atlas 14 and see what you think. Spoiler — through Hineycast, we can infer with relative confidence that this rainfall event was no greater than a ~1-year storm for all durations equal to and exceeding one hour, depending on one’s location in the City. To assess 5-minute storms, though, we would need local rain gauge measurements of at least 5-minute frequency. This granularity is much harder to come by! Coming soon :).
EventCamp Mystic Flooding DateJuly 2025 LocationTexas Hill Country TypeFlash Flood
This rainfall event became one of the deadliest inland flooding disasters in the mainland US in decades. A slow-moving convective storm system fed by tropical moisture stalled over a normally dry drainage basin, bringing 10 inches of rain in roughly three hours. Thunderstorms were in the forecast, but the most advanced weather-forecast models still struggle to predict the placement and severity of summertime storms. The deluge here was far greater than forecasted.
EventHurricane Helene DateSept 2024 LocationAsheville, NC TypeTropical Cyclone
Hurricane Helene was the deadliest hurricane to strike the US mainland since Hurricane Katrina. Scroll through the map images to see rainfall totals for each hour. Notice the radar artifacts just northeast of Knoxville. These radar artifacts represent one of the most significant limitations of radar-based precipitation estimates.
EventHurricane Harvey DateAugust 2017 LocationHouston, TX TypeTropical Cyclone
Over four days, Harvey dropped more than a year’s worth of precipitation across the Houston metro and upper Texas coast. View hourly or daily data on the map to track the sustained, relentless rainfall bands that shattered continental U.S. tropical rainfall records.
The data sources
SourceIEM Radar Mosaic TemporalHourly & daily Grid0.01° (~1 km) AvailableLive · archive Daily ~2011 / Hourly ~2013 FeedsMap & pin AccuracyModerate
IEM Radar Mosaic: The map rasters. This product is a catalog of NOAA’s Multi-Radar Multi-Sensor (MRMS) mosaic of radar only rainfall estimates, with no gauge correction, in 0.01° (~1 km) squares. It is the best answer to where rain fell and how hard — it can resolve a single storm cell a mile or two across. Against a physical gauge it runs roughly 12–20% low overall, more in mountains and the cold season. Every view is labelled with the exact local clock time its total runs from and to. The archived hourly products reach back to about 2013 and the archived daily view to about 2011; earlier dates predate the archive. A 2-minute product is also archived.
Drop a pin and its popup reads the exact ~1 km value straight from the color under it in the map image, so the pin and the map can never disagree. It is radar-only, with the same ~12–20% low bias as the map; for a gauge-corrected total use the chart line below.
SourceIEMRE (gauge-corrected) TemporalHourly & daily Grid0.125° (~12 km) Available1981–present* FeedsChart (≤90 d) AccuracyHighest
IEMRE is Iowa State's hourly repackaging of NOAA's Stage IV rainfall analysis. IEMRE averages it onto a 0.125° (~12 km) grid, then rescales each day's hours so they add up to Oregon State's gauge-based PRISM total. The benefit of this approach is that the chart data are much more reflective of actual hourly rainfall amounts, but averaged over a larger area. Every point in the same ~12 km cell returns one value per hour.
While IEM's own page calls IEMRE “nothing special” and points to ERA5-Land, Daymet, gridMET and PRISM instead, I beg to differ. As far as I can tell, none of those four sources provide gauge-corrected hourly rain at all. IEMRE is the only free source that serves gauge-corrected hourly rain for any point in the lower 48, updated through the current day.
*While the IEM Reanalysis archive itself reaches back to 1893, Hineycast standardises on 1981 to present for all charts.
SourcePRISM (via ACIS) TemporalDaily (24h ending 12 UTC) Grid2.5 arc-min (~4 km) Available1981–present FeedsChart (>30 d) AccuracyHigh
For spans longer than 30 days the chart switches to PRISM, Oregon State’s gauge-based analysis, on a 2.5 arc-min (~4 km) grid back to 1981 — this is what gives Hineycast its ~45-year reach. One caveat: PRISM’s “day” is fixed at 12 UTC (a 24-hour period ending at 12:00 UTC), so the local hour shifts by one between summer and winter because the boundary does not move when the clocks do. An evening storm can be filed under the next date. Season totals land within about 5%, but individual days can be offset. Like the shorter-range chart, it shows as a rising line for the running total with bars for each day’s rain, with daily bars and vertices plotted at the 12 UTC boundary (e.g. 8:00 AM Eastern in summer / 7:00 AM in winter) so each bar reflects the exact 24-hour window measured, with hover details showing the exact span.
Radar-based accumulation estimate(MRMS, 1 km resolution)
Get a rainfall accumulation chart by double‑clicking on a map point. Click the crosshairs for your current location.
More options are in map and chart filters.
Gauge-corrected time series(IEMRE, 12 km resolution)
Chart will appear here once a location is selected.