rhiza-forecasting

Plugin rhiza-forecasting from marketplace weather-skills in rhiza-research/weather-skills. Commands read from the repo's manifest — today.

Skills (54)

• 3dpaws-fetch

  Fetch daily 3D-PAWS (3D-Printed Automatic Weather Station) observations (rainfall, mean, max and min temperature, humidity, pressure, wind speed and gust) from a CHORDS portal, by default the FEWS NET network of Kenya Met Department and KALRO stations, for chosen station ids or every station in a la

• aggregate-temporal

  Roll up a weather-skills Zarr along its time axis or forecast lead (step) axis into fixed windows (sub-daily such as 6 hours, daily, weekly, 10-day, calendar monthly, or any whole number of days) or a rolling window, taking the mean, min, or max. Use to turn hourly rates into 6-hour means, or daily

• arco-era5-fetch

  Fetch ERA5 hourly reanalysis (temperature, wind, precipitation, pressure-level fields, and more) for a date range and region from Google's public ARCO-ERA5 Zarr store, as a weather-skills Zarr. No credentials needed. Use when a task needs gridded reanalysis as ground truth for comparison, verificati

• cams-fetch

  Fetch CAMS global air-quality fields (PM2.5, PM10, near-surface NO2, O3, SO2, CO, aerosol optical depth, total columns) for a lat/lon box from the Copernicus Atmosphere Data Store, as a gridded weather-skills Zarr: the 5-day forecast (0.4°, 2015 on) or the EAC4 reanalysis (0.75°, 3-hourly, 2003 on).

• chirps-fetch

  Fetch CHIRPS daily rainfall observations (0.05° land grid, 1981 to a few days ago) for a date range and optional region from the dynamical.org catalog, as a weather-skills Zarr. Joins the validated final product with the preliminary one for recent days into one `precip` series. Use when a task needs

• clim-fetch

  Fetch a precomputed daily climatology (mean and standard deviation for each day of the year) of precipitation, sea-surface temperature, 2 m temperature or 10 m wind from a public Sheerwater mirror of IMERG, ERA5, CHIRPS, ECMWF IFS, GEFS, OISST or NOAA CPC RFE2 (Africa), laid onto a real calendar dat

• clip-region

  Clip a gridded or station Zarr to a lat/lon box, a named region (country, province or county, multi-country region), or a GeoJSON polygon. Use to restrict any dataset to an area of interest.

• cmip6-fetch

  Fetch one CMIP6 climate-model run (a model, experiment, and variable such as tas or pr) for a date range and region from the public Pangeo CMIP6 catalog on Google Cloud, as a weather-skills Zarr. No credentials needed. Use when a task needs climate-projection grids, either historical runs (1850-2014

• coarsen

  Put a gridded weather-skills Zarr onto a coarser or equally spaced lat/lon grid, copied from another Zarr or built from a resolution and offset, by linear interpolation or an area-weighted (conservative) mean. Use to make two grids match cell for cell before subtracting or scoring them, e.g. fine sa

• concat

  Join two or more weather-skills Zarrs into one along a named dimension, either extending an existing dim (consecutive time windows, ensemble halves, adjacent boxes) or stacking them along a new dim with optional labels (e.g. one entry per model). Use when combining ensemble members, stitching time w

• convert-calendar

  Convert a weather-skills Zarr's time axis to another CF calendar: the standard calendar of observations, or a climate-model calendar such as noleap (no Feb 29) or 360_day (twelve 30-day months). Use to put two datasets on one calendar before comparing them, e.g. a noleap model run against standard-c

• convert-to-totals

  Convert precipitation mean rates in a time-aggregated weather-skills Zarr into period totals in mm, by multiplying each interval's rate by its stamped aggregation period (calendar months by their real length). Use when you need weekly, monthly, or seasonal rainfall amounts for a map, table, or compa

• csv-io

  Export a weather-skills Zarr as it is to a CSV or Excel table (one row per time, station or grid cell, units in the headers), or read a CSV or Excel (.xlsx) sheet of station records, a gridded table, a wide table with one column per day, or a single rain gauge into a weather-skills Zarr. Use when so

• cumulus-fetch

  Fetch the Cumulus AI operational ensemble precipitation forecast (Aurora S2S run, 30 members, 46 daily leads, 0.25° or 1° global grid) for one init date or a range of inits and optional region from a private Azure Blob container, as a weather-skills Zarr. Needs an Azure read token (AZURE_STORAGE_SAS

• deaccumulate

  Convert a forecast variable stored as a running total since the forecast start (init) into per-step values along its lead (step) axis; precipitation amounts become mm day-1 rates. Use on archives whose precipitation or radiation still grows with lead time and has amount units such as mm, m, or J m-2

• difference

  Subtract one weather-skills Zarr from another (A − B), cell by cell, keeping only the coordinates both share and broadcasting over dims only A or only B has. Use for anomalies (a field minus its baseline mean), change maps (scenario minus historical), or forecast minus observations.

• downscale

  Put a gridded weather-skills Zarr onto a finer or equally spaced lat/lon grid, by linear interpolation or by interpolation plus empirical quantile mapping (q-q) against an observed climatology. The target grid comes from an integer factor, a resolution, or another Zarr's grid. Use when a task needs

• drop

  Remove named data variables or non-index coordinates from a weather-skills Zarr. Use to strip leftovers that stop two datasets from lining up, or to keep only the variables a later step needs.

• dynamical-fetch

  Fetch a forecast or analysis from the dynamical.org catalog (GFS, GEFS, ECMWF IFS-ENS and AIFS, Google WeatherNext 2 and 3, ICON-EU, MRMS, IMERG, CHIRPS) for an init date or date range over a region, as a weather-skills Zarr. No credentials needed. Use it first for any of these products.

• ecmwf-fetch

  Fetch an ECMWF S2S (extended-range, 46-day) ensemble forecast for one init date over a lat/lon box, as a weather-skills Zarr: precipitation, 2 m temperature, SST, winds, solar and thermal radiation, pressure-level fields and more. Also serves inits before 2026 and fields such as ocean variables, 6-h

• ecmwf-hres-fetch

  Fetch the deterministic ECMWF HRES (IFS high-resolution, single-member) forecast for one init date and run hour over a lat/lon box from the free ECMWF Open Data feed, as a weather-skills Zarr at 0.25°, out to 15 days. Slow: every lead is a global download. For ECMWF forecasts use dynamical-fetch fir

• edit-coords

  Rename data variables, coordinates, or dimensions in a weather-skills Zarr, or re-index a dimension by another coordinate along it. Use to make two datasets' names match before combining them, or to index stations by their ids.

• ghcn-daily-fetch

  Fetch NOAA GHCN-Daily station observations (daily precipitation and max, min, mean temperature) for a date range and lat/lon box, as a weather-skills station Zarr. No credentials needed; covers about 130,000 stations worldwide. Use when a task needs daily in-situ station data anywhere in the world,

• gsmap-fetch

  Fetch JAXA GSMaP v6 gauge-corrected hourly 0.1° precipitation (archive from April 2025, 60°N-60°S) for a region and date range from JAXA's public Earth API, as a weather-skills Zarr. No credentials needed. Use when a task needs recent hourly satellite rainfall, or an independent satellite estimate t

• imerg-fetch

  Fetch NASA IMERG daily precipitation (global 0.1° grid) for a date range from NASA GES DISC via Earthdata login, as a weather-skills Zarr. Serves the daily Late (GPM_3IMERGDL, about 2 days behind) and Final (GPM_3IMERGDF, gauge-adjusted research quality) files. Use only for IMERG Final. For IMERG La

• indicator

  Apply a yes/no weather rule, such as more than 25 mm of rain in 8 days or a rainy-season onset definition (ICPAC, CHC), to each day or week of a daily or weekly Zarr, and optionally reduce it to an onset date or an ensemble probability. Use for onset dates, wet or dry spells, and the chance that a t

• inspect-zarr

  Print a short summary of a weather-skills Zarr: dimension sizes, coordinate values, and for each data variable its dims, dtype, units, key attributes, min/max/mean, finite and NaN counts, and a small sample of values. Use to see what a Zarr holds, or to check values and units after a fetch or transf

• label-group

  Add label coords to a gridded or station Zarr: the country, admin-1 / admin-2 unit (province, state, county), subregion, continent, or GeoJSON polygon each cell or station lies in, or a time label (month, year, day of year, or month groups such as MAM,OND). Data is unchanged. Use to group, select, o

• local-extrema

  Find the local highs or lows of a gridded field (pressure centres in MSLP or geopotential, rainfall or temperature maxima, SST warm and cold spots) and write their locations and central values as a station-style point dataset. Use for H / L labels on a weather map, or to list and compare centres.

• merge

  Combine the variables of two or more weather-skills Zarrs into one dataset, keeping the coordinate values they share. Use to put related quantities side by side, e.g. a climate index and rainfall, to compare or plot one against the other.

• meteosat-fetch

  Fetch Meteosat geostationary satellite imagery (FCI or SEVIRI) over a lat/lon box at chosen times on one day from the EUMETSAT Data Store, as a weather-skills Zarr of visible reflectance (%) and infrared brightness temperature (K) on a regular lat/lon grid. Covers Africa, Europe and the Indian Ocean

• netcdf-io

  Export a weather-skills Zarr to a NetCDF (.nc) file, keeping its attributes and processing history, or read one or more NetCDF files (NetCDF-3 or NetCDF-4, such as CDS, CHIRPS or model downloads) into a weather-skills Zarr. Use when someone wants data as .nc for another tool, or has their own .nc fi

• neuralgcm-fetch

  Fetch a NeuralGCM subseasonal ensemble forecast (up to 50 members, 6-hourly leads to 60 days, ~2.8° global) for one init or a range of inits over an optional lat/lon box, as a weather-skills Zarr: IMERG-trained precipitation, or 2 m temperature and dewpoint. Reads the private Tomorrow Now 2026 realt

• oisst-fetch

  Fetch NOAA OISST v2.1 daily 0.25° sea-surface temperature (1981-09 to present) for a date range and region from NOAA PSL's public OPeNDAP server, as a weather-skills Zarr. No credentials needed. Use when a task needs gridded SST observations, e.g. for ocean analysis or comparison against forecasts o

• openaq-fetch

  Fetch OpenAQ air-quality station observations (daily mean PM2.5, PM10, NO2, O3, SO2, CO) for a date range and lat/lon box, as a weather-skills station Zarr. Needs a free OPENAQ_API_KEY. Use when a task needs in-situ air-quality or atmospheric-composition data, for example to compare against gridded

• pbc-fetch

  Fetch a StillLearning PBC (probabilistic bias correction) AI Weather Quest precipitation forecast for week 3 or week 4, for one init or a range of inits, from the private Sheerwater datalake, as a weather-skills Zarr. Values are the probability of each rainfall quintile on a 1.5° global grid, not am

• point-value

  Sample a gridded weather-skills Zarr (satellite, reanalysis, or forecast) at point locations and write a station Zarr with the same layout as station observations. Points come from a station Zarr, lat/lon pairs on the command line, or a CSV. Use to put a grid and station data on the same points befo

• provenance

  Read the processing history (weather_skills_history) recorded on a weather-skills Zarr, plot image (PNG, JPEG, GIF, HTML, PDF, SVG), GeoJSON, or CSV/Excel table and print it as a readable lineage, raw JSON, or a bash script that regenerates the file. Can also validate that history against its schema

• resolve-region

  Turn a country, multi-country region, ocean box, state/province/county, or landmark name into a N/W/S/E lat/lon bounding box, and optionally write its boundary polygon, or every county or district inside it, as GeoJSON. Use whenever a task names a place and a later step needs a bbox, a polygon mask,

• resolve-time

  Turn a relative date token (last-2w, latest, yesterday, last-dekad, this-month, now-3d) into absolute YYYY-MM-DD start/end or single-day values, from the current UTC date or a given as-of date. Use before a fetch when the user asked for a relative window instead of explicit dates.

• rfe-fetch

  Fetch NOAA CPC African Rainfall Estimate (RFE2) daily rainfall for a date range and region into a weather-skills Zarr. Credential-free, 0.1° over Africa, daily from 2001, about one day behind real time. Use when a task needs RFE2 rainfall, such as a gauge-merged satellite estimate to compare with CH

• select

  Pick entries along one dimension of a weather-skills Zarr, by integer position or by coordinate value: an ensemble member, a forecast step, a date, a station. Picking one entry removes that dimension but keeps the picked label as a scalar coordinate, so outputs from different sources line up for sta

• smap-fetch

  Fetch NASA SMAP Enhanced L3 (SPL3SMP_E) daily 9 km volumetric surface soil moisture for a region and short date range via NASA Earthdata, as a weather-skills Zarr. Needs an Earthdata login. Use when a task needs gridded soil-moisture observations, e.g. for drought or agricultural analysis, or compar

• standardize-anomaly

  Compute a standardized anomaly (z-score), (field − mean) / std, from a data Zarr and a climatology Zarr holding the mean and standard deviation. The result is dimensionless, so unusualness compares across variables and places. Use for anomaly detection and SPI/SPEI-style standardized indices.

• step-to-time

  Relabel a forecast Zarr by valid date instead of lead time: one init's step (lead) axis becomes a time axis holding init + step, or many inits at one selected lead move from init dates to valid dates. Use when a forecast must line up with observations or other date-labeled data for plotting, differe

• submit-feedback

  Turn a title and markdown body you write into a prefilled GitHub new-issue link, which the user clicks to file feedback about the skills under their own account. Nothing is sent by the skill. Use when the user asks to report a bug, a suggestion, or other feedback about a skill.

• summarize-dim

  Collapse one or more named dimensions of a weather-skills Zarr with a statistic (mean, std, min, max, sum, median, quantile), either whole or separately for each value of a label coord such as a county or season. Use for an ensemble mean, spread, or percentile band across members, disagreement acros

• tahmo-fetch

  Fetch daily TAHMO weather-station observations (rainfall, mean, max and min temperature, humidity, pressure) from the Trans-African Hydro-Meteorological Observatory network, for chosen station ids or every station in a lat/lon box, as a weather-skills station Zarr. Can also list the stations in a bo

• unit-convert

  Convert data variables in a weather-skills Zarr to target units (any UDUNITS / CF string), or put recognized air-temperature, precipitation and pressure variables in the standard units degree_Celsius, mm day-1, mm and hPa. Use to put two datasets on one units basis before comparing or combining them

• verify

  Score a forecast against observations on a shared grid or shared stations, cell by cell: event hits at a threshold, bias (forecast − obs), or mean absolute error. Writes the metric field as a Zarr and prints a regional score. Use for forecast skill maps and event verification.

• weathernext-fetch

  Fetch a Google WeatherNext AI forecast for one init or a range of inits over an optional lat/lon box, as a weather-skills Zarr: WeatherNext 2 (64-member ensemble, 6-hourly leads to 15 days, 0.25° global, from 2025) or WeatherNext 3 (64-member ensemble or its precomputed mean and percentiles, hourly

• wind-speed-direction

  Compute wind speed and wind direction (the direction it blows from and the direction it blows to, in degrees clockwise from north) from the u and v wind components of a gridded or station weather-skills Zarr. Use when a task needs wind speed, a meteorological wind direction, or arrows pointing downw

• windborne-fetch

  Fetch WindBorne data as a weather-skills Zarr: the WeatherMesh-6 AI forecast (128-member ensemble, 3-hourly to 15 days, 0.25°) as a grid (live or archived runs back to mid-2026), at lat/lon points with spread and percentiles, or as vertical soundings; plain-language site summaries; ERA5 1991–2020 ho

• zonal-moisture-transport

  Compute eastward moisture flux, specific humidity times zonal wind (q·u), from a weather-skills Zarr holding both fields on pressure levels, and by default integrate it through the column to eastward integrated vapour transport (IVT, ECMWF `viwve`, kg m-1 s-1). Use for zonal moisture-transport maps

Subagents (1)

• forecaster

  Meteorological data assistant. Composes the bundled forecasting skills to answer questions and build fetch-transform pipelines over weather and climate data.

Similar plugins