Marine heatwaves and cold spells

Flags periods when a cell was unusually warm — or unusually cold — for the time of year, and measures how unusual and for how long. Follows the definition of Hobday et al. (2016): a threshold set by a high percentile of the cell's own seasonally varying climatology, an event being a run of consecutive steps above it, and an intensity measured from the climatology rather than from the threshold.

Usage

marine_heatwave(
  env_dat,
  var = "SST",
  percentile = 0.9,
  min_steps = 1L,
  max_gap = 0L,
  direction = c("warm", "cold"),
  measures = c("event", "intensity", "category", "duration", "cumulative"),
  prefix = NULL
)

Arguments

env_dat: an sf POINT object with one row per location and time step, as datamatch's access functions return
var: the covariate to examine, typically sea surface temperature
percentile: threshold percentile, between 0 and 1. Hobday et al. use 0.9 for heatwaves; the cold-spell equivalent is 0.1, which direction applies for you
min_steps: shortest run of consecutive steps counted as an event. Use 5 on daily data for the published definition
max_gap: runs separated by at most this many non-exceeding steps are joined into one event. Use 2 on daily data for the published definition
direction: "warm" for heatwaves, "cold" for cold spells
measures: which of "event", "intensity", "category", "duration" and "cumulative" to add
prefix: stem for the new column names. Defaults to "mhw", or "mcs" when direction = "cold"

Value

env_dat with one column per requested measure

Details

The Gulf of Maine is one of the fastest-warming shelf seas on record, so whether an animal was there in an ordinary summer or a heatwave summer is often a more useful covariate than the temperature itself.

What each measure is

event is TRUE where the cell is in an event that met min_steps.
intensity is the departure from the climatological mean, in the units of var, and is NA outside an event. Measured from the climatology, not from the threshold, so it is comparable between cells whose thresholds differ.
category is 1 to 4 — moderate, strong, severe, extreme — from how many multiples of the threshold's excess over the climatology the value reaches.
duration is the length of the event in time steps, repeated on every row belonging to it.
cumulative is the summed intensity over the whole event, which distinguishes a long mild event from a short violent one.

Time steps, not days

Hobday et al. define an event as five or more consecutive days. This works in whatever step the data is in, because datamatch's access functions serve monthly, daily and sub-daily archives alike. On daily data min_steps = 5 and max_gap = 2 reproduce the published definition. On monthly data they do not — five consecutive months is a far rarer and larger thing — so the defaults here are deliberately permissive and the choice is left to you.

Runs are counted over consecutive entries in the sequence of time steps present in the data, which has two consequences worth knowing.

A cell absent from a step that other cells have breaks that cell's event: it cannot be shown to have stayed warm through a step it has no value for.

A step missing from the record entirely is a different matter. It is not in the sequence at all, so the steps either side of it are adjacent and an event runs straight through. On a record with holes in it that can join what were two events, and the duration will be in steps you have rather than in elapsed time. Check the series is complete before reading duration as a period.

Why it needs a long series

A 90th percentile taken over three values is the largest of the three, and every third step will then be a heatwave. The published definition uses a 30-year baseline. This warns when the groups behind the threshold are too thin for the percentile to mean anything, but it cannot tell you that a 10-year baseline is short for your purpose.

Because the climatology is computed from the data given, a warming trend within the series raises the threshold and later heatwaves are measured against a warmer baseline. That is a choice, not an oversight: it makes events relative to recent conditions. If you want a fixed baseline, compute the threshold on a subset and apply it.

References

Hobday AJ, Alexander LV, Perkins SE, Smale DA, Straub SC, Oliver ECJ, Benthuysen JA, Burrows MT, Donat MG, Feng M, Holbrook NJ, Moore PJ, Scannell HA, Sen Gupta A, Wernberg T (2016). A hierarchical approach to defining marine heatwaves. Progress in Oceanography 141, 227-238. doi:10.1016/j.pocean.2015.12.014

Hobday AJ, Oliver ECJ, Sen Gupta A, Benthuysen JA, Burrows MT, Donat MG, Holbrook NJ, Moore PJ, Thomsen MS, Wernberg T, Smale DA (2018). Categorizing and naming marine heatwaves. Oceanography 31(2), 162-173. doi:10.5670/oceanog.2018.205

Examples

if (FALSE) { # \dontrun{
env <- marine_heatwave(env, "SST")

# The published definition, on daily data.
env <- marine_heatwave(env, "SST", min_steps = 5, max_gap = 2)

# Cold spells instead.
env <- marine_heatwave(env, "SST", direction = "cold")
} # }