DARK ARs + AR-Children Catalog (ERA5, 1979–2023)

This dataset provides yearly atmospheric-river (AR) masks and parent–child tracking fields derived from the ERA5 reanalysis, produced using the DARK + Children algorithm.The workflow combines the DARK (formerly Buffet v1.0) detection algorithm for AR identification and a temporal tracking extension that links parent ARs to their “children”, secondary systems maintaining strong IVT after Antarctic landfall.Although originally developed with a polar focus, the implementation operates globally on the full ERA5 grid.All outputs follow ARTMIP Tier-1 conventions, including variable naming, attributes, dimensions, CF metadata, and filename structure. Method Input field ERA5 vertically integrated water vapor flux components: viwve (eastward) viwvn (northward) Additional fields: lsm (land–sea mask; > 0 = land/ice, defines Antarctic region) Data source: ERA5 reanalysis (ECMWF), typically 6-hourly fields. Overview of the steps 1. DARK Algorithm (AR Detection) Rolling seasonal threshold (98th percentile baseline) For each leap day-of-year (1–366), a 15-day rolling climatology of Integrated Vapor Transport (IVT) is computed over all years (1979–2023).The 98th percentile of IVT values at each grid cell defines the local, seasonally varying threshold.Non-leap years are mapped to leap-year DOYs by shifting days ≥ 29 Feb (+1).This climatological percentile file is stored aspercentile_98_IVT_rolling_15D_era5.nc. Exceedance mask A grid cell is marked 1 if its IVT exceeds the local day-of-year threshold; otherwise 0. Area filtering (< 200 000 km²) Connected regions are identified using 8-connectivity with periodic longitude boundaries (dateline continuity).Components smaller than 200 000 km² are removed.Grid-cell areas are computed on WGS-84 using the local geocentric radius and Δlat/Δlon. Length filtering (< 2 000 km) Each connected component (AR candidate) is skeletonized and represented as a graph.The endpoint-to-endpoint path best aligned with the component’s principal-component axis is selected, extended to the edges, and smoothed using a B-spline.Sub-grid geodesic segment lengths (WGS-84) are summed; components shorter than 2 000 km are discarded. Output assembly Each yearly NetCDF file contains the binary field ar_binary_tag(time, lat, lon) with 1 = AR and 0 = no AR. File pattern: ERA5.ar_tag.DARK.<freq>.<YYYYMMDD-YYYYMMDD>.nc where <freq> is inferred from the ERA5 IVT temporal resolution (e.g. 6-hourly). 2. Parent–Child Tracking Extension (DARK + Children) Purpose The DARK + Children extension identifies temporal continuity in atmospheric-river events and classifies them into parents (original ARs) and children (secondary ARs evolving after Antarctic landfall). Temporal label matching Each 6-hourly exceedance mask is labeled using 8-connected components with periodic longitude boundaries.Labels are matched across consecutive timesteps by maximum pixel overlap, building continuous AR tracks.Each persistent structure receives a unique track ID. Parent identification At any timestep, if ar_binary_tag = 1, the corresponding component is tagged as a parent AR (ar_type = 1). Child identification For each track: if it previously overlapped a parent AR, and later intersects the Antarctic mask (lsm > 0), then subsequent timesteps within that track are classified as children (ar_type = 2).Children thus represent AR fragments maintaining high IVT after parent landfall. Geographic filter (optional) Optionally, only parents “born outside Antarctica” are retained using one or both criteria: latitude ≥ −65°, or landmask = 0. This ensures parent systems originate over the ocean before landfall. Output assembly Each yearly file includes both the binary and classification fields: Variable Type Description ar_binary_tag(time, lat, lon) byte 1 = AR (parent or child), 0 = none ar_type(time, lat, lon) byte 0 = none, 1 = parent, 2 = child File pattern: ERA5.ar_children_tag.DARK_children.<freq>.<YYYYMMDD-YYYYMMDD>.nc Attributes: description = "ERA5 AR parents + children (time-continuous tracking across 1979–2023)" scheme = "DARK_children" version = "1.0" Conventions = "CF-1.6" Files & Format One NetCDF file per year: ERA5.ar_children_tag.DARK_children.<freq>.<YYYYMMDD-YYYYMMDD>.nc where <freq> is inferred from ERA5 temporal resolution (e.g. 6-hourly). Variables Variable Type Description ar_binary_tag(time, lat, lon) byte 1 = AR (parent or child), 0 = none ar_type(time, lat, lon) byte 0 = none, 1 = parent, 2 = child Attributes: description = "ERA5 AR parents + children (time-continuous tracking across 1979–2023)" scheme = "DARK_children" version = "1.0" credits = "Developed by Victoire Buffet, Vincent Favier, and Benjamin Pohl" Conventions = "CF-1.6" Coordinates: time — CF-compliant (minutes since <year>-01-01 00:00:00, calendar = "standard", axis = "T") lat — degrees_north (axis = "Y") lon — degrees_east (axis = "X") Compression: NetCDF-4 deflate (zlib = True, complevel = 1, shuffle = True); time unlimited. Spatial / Temporal Coverage Spatial: Global ERA5 native 0.25° × 0.25° grid; Antarctic sector (−90 to −15° lat).(Algorithm tuned for polar ARs but evaluates the entire globe.) Temporal: 1979 – 2023, using ERA5 6-hourly timesteps. Parameters (defaults) Rolling window: 15 daysPercentile: 98thMinimum area: 200 000 km²Minimum length: 2 000 kmLatitude band: −90° to −15°Outside latitude: −65°Time resolution: 6-hourlyTemporal coverage: 1979 – 2023 Reproducibility Inputs ERA5 uIVT /vIVT flux files ERA5 Antarctic land–sea mask Outputs Yearly AR tracking files:ERA5.ar_children_tag.DARK_children.6hourly.YYYYMMDD-YYYYMMDD.nc Software Python ≥ 3.11Key libraries: xarray, numpy, scipy, scikit-image, geopy, networkx, matplotlibGeodesic distances computed with geopy (WGS-84)Files written CF-1.6-compliant using netCDF4 Credits Developed by Victoire Buffet, Vincent Favier, and Benjamin Pohlat Institut des Géosciences de l’Environnement (Grenoble) and Biogéosciences (Dijon). Keywords Atmospheric Rivers; ERA5; DARK; DARK_children; ARTMIP; IVT; Antarctic; Polar Meteorology; Reanalysis; Climate Diagnostics