dhw: an R package for calculating various SST metrics (Maximum Monthly Mean Climatology, SST Anomalies, Coral Bleaching HotSpots, Degree Heating Weeks)
The NOAA Coral Reef Watch (CRW) daily global 5km satellite coral bleaching Degree Heating Week (DHW) product shows accumulated heat stress during the most recent 12-week period, and ranges from 0 to 20 °C-weeks. DHW is directly related to the timing and intensity of coral bleaching and is the most common metric in assessing coral bleaching events (see here for further details).
Due to changes in the datasets used by the CRW over the years (see Skirving et al. 2020 for details), the CoralTemp degree heating weeks (and other SST metrics) are calculated against a specific baseline climatology (1985–1990 and 1993) to allow for internally consistent anomaly products.
The
dhw package provides a series of functions to recreate the algorithms for the CRW metrics (Maximum Monthly Mean Climatology, SST Anomalies, Coral Bleaching HotSpots, Degree Heating Weeks) using any SST product (e.g., OISST) that spans the baseline period of 1985 to 1993.
The package is uses the `terra` package to process the `raster` data and is (potentially) slower than other R packages (e.g. `stars`) and very much slower than other alternatives (e.g. `cdo`) for large datasets. Updated versions of the package will switch to `stars` spatiotemporal arrays and raster data cubes, and provide alternative functions for `cdo` to speed up large datasets.
Also included are a series of functions for visualising SST metrics and functions to reliably download common datasets (e.g. CRW, OISST) via the [NOAA NCEI thredds server](http://www.ncei.noaa.gov).
Please file any issues and comments and suggestions to george.roff@csiro.au.
the current development version of dhw can be installed from github:
#install.packages("remotes")
devtools::install_github("https://github.com/marine-ecologist/dhw", force=TRUE)
Skirving et al (2020) Remote Sensing 12(3856) doi:10.3390/rs12233856