Environmental Drivers

Many environmental drivers were compared to the spring bloom date to investigate which factors had the most influence on interannual variability of the bloom timing. For each driver, the monthly averages for January, February and March were extracted.

The environmental drivers considered:

  • Wind speed (as wind speed cubed)

  • Solar radiation at ocean surface

  • Surface temperature

  • Surface salinity

  • Fraser river flow

  • Mid-depth nitrate concentration (30-90m)

  • Deep nitrate concentration (below 250m)

  • Halocline depth

  • Turbocline depth

  • Average eddy diffusivity in the upper 15m and 30m

  • Density difference from surface to a series of depths (5m, 10m, 15m, 20m, 25m , 30m)

  • Surface zooplankton concentration

  • Surface mesozooplankton concentration

  • Surface microzooplankton concentration

  • Depth-integrated zooplankton concentration

  • Depth-integrated mesozooplankton concentration

  • Depth-integrated microzooplankton concentration

For efficient data usage, four pickle files for each year were created for different variables using this notebook. As with bloom timing variables, these files are stored in /ocean/aisabell/MEOPAR/extracted_files. The format is loc being the location of interest (eg. S3), modver being the model run (eg. 201905) and year being the year.

There is a file for each year for:

  • Location-specific variables JanToMarch_TimeSeries_year_loc_modver.pkl

  • Non-location-specific variables JanToMarch_TimeSeries_year_modver.pkl

  • Location-specific mixing variables JanToMarch_Mixing_year_loc_modver.pkl

Biological variables, such as nitrate, diatom, flagellate, ciliate, microzooplankton and mesozooplankton concentrations as well as time were all extracted from daily .ptrc_T files. Temperature and salinity were extracted from daily .grid_T files. Eddy diffusivity values were taken from hourly .grid_W files. Here “T” and “W” specify the vertical model grid on which the variables in each file are defined. Wind and solar variables were extracted from model forcing files. Fraser flow data is from Environment Canada and comes as daily flow averages, measured at Hope. Deep nitrate concentrations (below 250m) were also calculated from .ptrc_T files, but these are not location specific and rather represent the average across the entire Strait of Georgia region.

Some key factors that relate to mixing and stratification were calculated to compare to bloom timing and to other environmental drivers. This includes the monthly averages of halocline depth and turbocline depth, the density difference from the surface to a series of depths from 5m to 30m, in increments of 5m, and the average eddy diffusivity over the upper 15m and 30m. Halocline depth was defined as the depth with the maximum gradient in salinity between two grid cells. Turbocline depth was defined as the depth above which the eddy diffusivity passes a threshold value of 1 x 10-3m2s--1. This threshold value was determined by examining depth profiles of eddy diffusivity, and is consistent with a turbocline definition by Luneva et. al (2015) [1].The code for these functions can be found here.

To recreate extracting these variables for a different location:

  1. Open the makePickles201905 notebook

  2. Follow the instructions in the second code cell of the notebook