Last week, martc03 wired up 13 government APIs as general-purpose data bridges. This week, mansurjisan (io.github.mansurjisan) goes deeper — six servers built for people who actually work with earth science data professionally.
The Six Servers
| Server | Score | What It Does |
|---|---|---|
| goes-mcp | 59 | GOES-18/19 satellite imagery — visible, infrared, water vapor, and composites via NOAA |
| recon-mcp | 59 | Hurricane reconnaissance data — HDOB flight observations, Vortex Data Messages, SFMR, and ATCF fixes |
| ww3-mcp | 59 | GFS-Wave (WAVEWATCH III) forecasts and NDBC buoy wave observations |
| usgs-mcp | 59 | USGS Water Services streamflow, flood stages, and peak events via waterservices.usgs.gov |
| adcirc-mcp | 59 | ADCIRC model configuration debugging, parameter lookup, and validation |
| schism-mcp | 59 | SCHISM model configuration debugging, parameter lookup, and validation |
All six score 59. No flags on any of them.
Who This Is For
These aren't consumer weather apps. They're bridges to the same data systems that operational meteorologists, coastal engineers, and hydrologists use daily.
GOES satellite imagery — GOES-18 covers the western hemisphere, GOES-19 covers the eastern. The server pulls visible, infrared, and water vapor channels plus composite products. This is what the National Weather Service uses for storm tracking and nowcasting.
Hurricane reconnaissance — When a hurricane threatens the US coast, the Air Force Reserve's 53rd Weather Reconnaissance Squadron (the "Hurricane Hunters") flies directly into the storm. They drop instruments called dropsondes and measure conditions at flight level. The recon-mcp server provides access to HDOB (High-Density Observations), Vortex Data Messages (the detailed center fix reports), SFMR (Stepped Frequency Microwave Radiometer — measures surface winds from the aircraft), and ATCF (Automated Tropical Cyclone Forecast) fix data. This is the raw observational backbone of hurricane intensity forecasting.
WAVEWATCH III — The global wave model run by NOAA's Environmental Modeling Center. The server bridges both the model forecasts and real-time NDBC (National Data Buoy Center) observations — the actual buoys floating in the ocean measuring wave height, period, and direction.
USGS streamflow — The USGS operates roughly 10,000 stream gauges across the United States, measuring real-time water levels and discharge. The server queries waterservices.usgs.gov for streamflow data, flood stages, and historical peak events. Essential for flood forecasting, water resource management, and dam operations.
ADCIRC and SCHISM — These are the two dominant coastal ocean circulation models used for storm surge forecasting and coastal flood prediction. ADCIRC (ADvanced CIRCulation) and SCHISM (Semi-implicit Cross-scale Hydroscience Integrated System Model) are computationally intensive and have complex configuration files with hundreds of parameters. These servers help debug and validate model configurations — parameter lookup, validation, and troubleshooting.
The Developer
The combination of servers tells a story. Someone who works with hurricane reconnaissance data, wave models, storm surge models, satellite imagery, and streamflow gauges is almost certainly in operational meteorology or coastal engineering — probably at a university, NOAA, or a consulting firm that does storm surge modeling. These servers solve their own daily workflow problems: checking satellite imagery, pulling recon data during a storm, validating model configs, cross-referencing buoy observations with wave forecasts.
This is the difference between building MCP servers for a market and building them for yourself. The specificity — HDOB flight observations, SFMR surface wind measurements, ADCIRC parameter validation — is the kind of detail you only include if you've actually needed it.
All six servers score 59. No flags.
Sources: mansurjisan — GitHub · Scorecard: io.github.mansurjisan (score 59)