Largest connected component of the Mid-Columbia stream network — columbia_main

A prepared full-graph river component for the directional and symmetric Whittle–Matérn examples. The original 9,521 source rows were imported as 2,758 unique ungrouped graph locations; the largest component retains 18,668 edges and 2,080 observations. The transformation preserves stable source-row identifiers in columbia_obs_id and records all selection counts and a content fingerprint in summary.

Usage

columbia_main_component

Format

`columbia_main_component`

A columbia_main_component list with five elements:

edges: An sf object containing the retained stream edges, h2oAreaKm2 directional weights, and edge geometry.
observations: A data.frame containing normalized edge positions, STREAM_AUG, ELEV, SLOPE, PRECIP, and stable columbia_obs_id values.
summary: Component-selection counts and the audited fingerprint.
weights_name: The edge attribute used for directional weights.
fingerprint: A SHA-256 fingerprint of the prepared component.

Source

Jay Ver Hoef (2023), MidColumbia.zip - a large spatial stream network data set, Figshare, doi:10.6084/m9.figshare.24132840.v1 . The source data are licensed under the Creative Commons Attribution 4.0 International license.

References

Isaak, D.J. et al. (2017). The NorWeST Database and Modeled Summer Temperature Scenarios. Water Resources Research, 53(11), 9181–9205. doi:10.1002/2017WR020969 .

Ver Hoef, J.M., Dumelle, M., Higham, M., Peterson, E.E., and Isaak, D.J. (2023). Indexing and Partitioning the Spatial Linear Model for Large Data Sets. PLOS ONE, 18(11), e0291906.