Skip to contents

Match PSCIS assessments to modelled crossings

Source: R/lnk_match_pscis.R
lnk_match_pscis.Rd

Convenience wrapper around lnk_match_sources() with BC PSCIS defaults. Optionally applies a hand-curated cross-reference CSV first — these known matches take priority over spatial matching.

Usage

lnk_match_pscis(
  conn,
  crossings = "bcfishpass.modelled_stream_crossings",
  pscis = "whse_fish.pscis_assessment_svw",
  xref_csv = NULL,
  distance = 100,
  to = "working.matched_pscis",
  verbose = TRUE
)

Arguments

conn

A DBI::DBIConnection object.

crossings

Character. Modelled crossings table.

pscis

Character. PSCIS assessment table.

xref_csv

Character. Optional path to a CSV of known PSCIS-to-modelled matches (GPS error corrections from field work). Must have columns stream_crossing_id and modelled_crossing_id. Applied first; remaining unmatched records go through spatial matching.

distance

Numeric. Maximum network distance (metres) for spatial matching.

to

Character. Output table name.

verbose

Logical. Report match statistics.

Value

The output table name (invisibly).

Details

Why matching matters: PSCIS assessments have field measurements (outlet drop, culvert slope, channel width). Modelled crossings have precise network position (blue_line_key, downstream_route_measure). Matching links the measurements to the network so lnk_score_severity() can classify crossings using real data.

xref CSV priority: the most valuable part of this function. Hand-curated matches from field work represent thousands of hours of GPS correction. When provided, these override spatial matching — if a PSCIS crossing is in the xref, it won't be re-matched spatially.

Examples

# --- Zero-config for BC ---
if (FALSE) { # \dontrun{
conn <- lnk_db_conn()

# Default tables — just works
lnk_match_pscis(conn)
# Matched 4,231 pairs: PSCIS <-> modelled within 100m
#
# Now your crossings table has both stream_crossing_id AND
# modelled_crossing_id — the bridge between field data and network.

# With hand-curated corrections from field work
lnk_match_pscis(conn,
  xref_csv = "data/overrides/pscis_modelled_xref.csv")
# Applied 892 known matches from xref
# Matched 3,339 additional pairs spatially
# Total: 4,231 matches

# Then score using the linked measurements
lnk_score_severity(conn, "working.crossings")
} # }