Join a section to a climatology and difference it

anomaly = value - clim_mean, matched on station, the cruise's month and depth bin. The month is the one cruise_key designates (YYYY-MM-NODC), not the date a station was occupied: a cruise often starts in the last days of the month before (CalCOFI 2607 worked line 93.3 inshore of station 50 on 30 June), and the release's climatology files every cast under its cruise's month. The date is used only when the key does not parse. Cells with no baseline come back NA rather than 0 — an unsampled baseline is not a zero anomaly, and collapsing the two is how a map ends up claiming "normal" for somewhere never measured.

Usage

cc_anomaly(section, clim, stations)

Arguments

section

from cc_transect_section().

clim

from cc_climatology().

stations

from cc_transect_stations() — supplies grid_key and the month, which section does not carry.

Value

section plus clim_mean, clim_sd, clim_n, anomaly, anomaly_sd; baseline attribute carried through.

Details

anomaly_sd expresses the departure in baseline standard deviations, which is what makes a 1 °C anomaly interpretable: large in the deep, unremarkable at the surface in spring.