Primary Contact
abigail.tyrell@noaa.gov
Choose a tech doc page to suggest edits
13 Zooplankton
Page Edit 1
This edit is for current chapter 12, Zooplankton Indices
Update contact information:
Contributor(s): Abigail Tyrell, Adelle Molina, Sarah Gaichas, and Harvey Walsh
Data steward: Abigail Tyrell abigail.tyrell@noaa.gov
Point of contact: Abigail Tyrell abigail.tyrell@noaa.gov
Page Edit 2 (if applicable)
Section: 12.1.2.1 VAST models
New text -- to replace lines 36-57 in current Rmd
Model selection was conducted in 2024 using this script:
https://github.com/NOAA-EDAB/zooplanktonindex/blob/main/VASTscripts/VASTunivariate_zoopindex_modselection.R
Model selection results are reported at this link: https://noaa-edab.github.io/zooplanktonindex/CopeModSelection.html
In the second stage of model selection, the day of year covariate had mixed success. Results of the best models which both converged and had the lowest AIC are reported here: https://noaa-edab.github.io/zooplanktonindex/CopeModResults.html
Models were run using REML and without bias correction. Two different observation models were applied. The default VAST index standardization (purpose = “index2” in make_settings) uses a Gamma distribution for positive catches and an alternative “Poisson-link delta-model” using log-link for numbers-density and log-link for biomass per number (ObsModel= c(2,1)). For datasets with years where the group of interest was sampled at all stations, we needed to use a different link function, the Poisson-link fixing encounter probability=1 for any year where all samples encounter the species (James T. Thorson 2019).. We kept all other settings for index standardization the same, but set (ObsModel= c(2,4)).
We applied the default observation model to Calanus finmarchicus, euphausiids, and large copepod datasets. We applied the alternative observation model to small copepods and zooplankton volume. Based on the model selection results linked above, a day of year catchability covariate was applied when the addition of the covariate had reduced the model AIC and produced a model likely to converge. The two models that incorporated day of year covariates were the spring euphausiid and spring large copepod models.
The script used to run the VAST models is available here: https://github.com/NEFSC/READ-EDAB-VAST_indices/blob/main/zooplankton/zooindex_run_model.R
Page Edit 3 (if applicable)
Section: 12.1.3.3 Years were selected for analysis
New text (to replace existing text)
SOE published in 2026: 1982 - 2024
The script that processes input data is available here: https://github.com/NEFSC/READ-EDAB-VAST_indices/blob/main/zooplankton/zooindex_process_input_data.R
SOEs published in 2023-2025:
The script that processes input data is available here: https://github.com/NOAA-EDAB/zooplanktonindex/blob/main/data/VASTzoopindex_processinputs.R
Primary Contact
abigail.tyrell@noaa.gov
Choose a tech doc page to suggest edits
13 Zooplankton
Page Edit 1
This edit is for current chapter 12, Zooplankton Indices
Update contact information:
Contributor(s): Abigail Tyrell, Adelle Molina, Sarah Gaichas, and Harvey Walsh
Data steward: Abigail Tyrell abigail.tyrell@noaa.gov
Point of contact: Abigail Tyrell abigail.tyrell@noaa.gov
Page Edit 2 (if applicable)
Section: 12.1.2.1 VAST models
New text -- to replace lines 36-57 in current Rmd
Model selection was conducted in 2024 using this script:
https://github.com/NOAA-EDAB/zooplanktonindex/blob/main/VASTscripts/VASTunivariate_zoopindex_modselection.R
Model selection results are reported at this link: https://noaa-edab.github.io/zooplanktonindex/CopeModSelection.html
In the second stage of model selection, the day of year covariate had mixed success. Results of the best models which both converged and had the lowest AIC are reported here: https://noaa-edab.github.io/zooplanktonindex/CopeModResults.html
Models were run using REML and without bias correction. Two different observation models were applied. The default VAST index standardization (purpose = “index2” in make_settings) uses a Gamma distribution for positive catches and an alternative “Poisson-link delta-model” using log-link for numbers-density and log-link for biomass per number (ObsModel= c(2,1)). For datasets with years where the group of interest was sampled at all stations, we needed to use a different link function, the Poisson-link fixing encounter probability=1 for any year where all samples encounter the species (James T. Thorson 2019).. We kept all other settings for index standardization the same, but set (ObsModel= c(2,4)).
We applied the default observation model to Calanus finmarchicus, euphausiids, and large copepod datasets. We applied the alternative observation model to small copepods and zooplankton volume. Based on the model selection results linked above, a day of year catchability covariate was applied when the addition of the covariate had reduced the model AIC and produced a model likely to converge. The two models that incorporated day of year covariates were the spring euphausiid and spring large copepod models.
The script used to run the VAST models is available here: https://github.com/NEFSC/READ-EDAB-VAST_indices/blob/main/zooplankton/zooindex_run_model.R
Page Edit 3 (if applicable)
Section: 12.1.3.3 Years were selected for analysis
New text (to replace existing text)
SOE published in 2026: 1982 - 2024
The script that processes input data is available here: https://github.com/NEFSC/READ-EDAB-VAST_indices/blob/main/zooplankton/zooindex_process_input_data.R
SOEs published in 2023-2025:
The script that processes input data is available here: https://github.com/NOAA-EDAB/zooplanktonindex/blob/main/data/VASTzoopindex_processinputs.R