plug.directory
Add my server

Climate Risk MCP Server

by towardsinnovationlabco.huggingface.towardsinnovationlab/climate_risk_monitoring

  • Remote endpoint
  • 4 tools
  • APACHE-2.0
  • Climate_Risk_Monitoring_climate_risk_report Combined climate-hazard evidence: historical data, Sentinel-2 and forecast in one call. Runs the three tools above in sequence and returns their outputs plus a hazard-focused summary. No numeric risk scores; exposure and vulnerability are not modelled. Slow (several minutes): prefer the single tools for quick questions. Returns: JSON string with all evidence, the hazard summary and the run time in seconds.
  • Climate_Risk_Monitoring_weather_data_analyst Historical climate evidence for a location from NASA POWER daily data. Compares the latest value with the seasonal climatology since 1991 and returns the historical percentile, P95/P99 thresholds, days at or above P95 in the last 30 days, the 30-day rainfall accumulation (precipitation only) and ADF/KPSS stationarity tests. Units: temperature °C, precipitation mm/day, wind speed km/h (NASA WS10M converted from m/s). Fast (about 30 seconds). Returns: JSON string with structured evidence, statistics and stationarity diagnostics.
  • Climate_Risk_Monitoring_weather_data_forecast 7-day forecast with calibrated uncertainty bands for a location. Trains the chosen model on NASA POWER history, tests it on the last 30 days, runs a 5-fold walk-forward backtest, builds adaptive conformal intervals and compares the Winkler interval score with a seasonal-naive benchmark. Takes one to a few minutes. Units: temperature °C, precipitation mm/day, wind speed km/h. Returns: JSON string with metrics, backtest, conformal diagnostics and the daily forecast.
  • Climate_Risk_Monitoring_weather_image_analyst Sentinel-2 land-surface evidence for a location (Earth Search STAC catalogue). Returns scene metadata (cloud cover, land-cover shares) and pixel-level spectral indices for up to 12 scenes: NDVI (vegetation), NDWI (surface water), NDSI (snow/ice), BSI (bare soil), with a consistency check between metadata and spectral values. Returns: JSON string with satellite metadata, spectral evidence and reliability notes.

As the server's own listing gives them. We never call a tool.