In casaultb/azmpdata: A data package to share derived metrics from AZMP research documents

Discussion of Maritimes environmental data access, June 24, 2020

B. Casault, E. Chisholm, C. Layton, C. Johnson, M. McMahon

Meeting Minutes

Discussion around Benoit's document

• The structure of azmpdata (fig 1) makes sense to all team members
• CIL and other non- depth-dependent variablesx will fit into derived/plankton category (fig 1)
• This box may need to branch off to include broad scale shelf data metrics

• This would also include climate metrics eg AMO (annual, broad scale metrics)

• Some variables will be listed as lower priority and structure can be expanded later to fit, but the structure should be robust enough to include future modifications

• Remote sensing products
• want to avoid multivaR or other packages having to grab products from ftp servers
• would fit best in blue data structure box (fig 1)

• single values for statistical boxes, no sample IDs or event IDs may need work around

• There is a certain compromise on file size which is balanced well

• varibale naming scheme

• would BODC or another standard vairable naming scheme be able to incorporate these metrics? likely only discrete..
• There is large benefit in brevity of variable names
• EC will compile a list of variable names and circulate for discussion
• any ambiguity will be resolved in data documentation
• should create look up table or style sheet
• discrete vs derived variables should be clear from naming
• look at intl standards from ICES, NOAA, BODC, etc

• potential to translate between naming schemes if needed

• data frame naming scheme

• using 'matrix'
• seperated by underscores
• first part of name defines variable type : discrete, derived, zooplankton, phytoplankton
• second part of name defines temporal scale: occupations, annual
• third part of name defines spatial scale: stations, sections, broadscale

• example name would be discrete_annual_stations.csv or derived_annual_broadscale.csv

• depth measures

• sampling depth and nominal depth should be included with data
• nominal depth is used for merging data

• sampling depth is often required for calculations eg. integrated chl

• anomolies

• will not be calculated in azmpdata
• CL and BC both have anomoly scripts
• EC will gather scripts, compare and bring any discrepancies/conflicts to the group for attention

• EC will create new anomoly function for multivaR

• Data from Peter

• Ice, NAO, AMO, RivSum

• Metadata files

• BC has created metadata file for HL2, P5 should be a simple addition
• these files contain 'nuts and bolts' detail info like sample IDs, precise lat and lon, event numbers etc
• They also include ODF filenames for reference to TS
• will be linked to data files through event and sample IDs
• would be mostly used by behind the scenes processers such as Benoit, not necessary for average user
• BC says they are easy to maintain
• Will be included for completeness and in case the info is needed for more detailed analyses

UPDATES

• CJ presented to working group
• CL and BC have been quite busy but hope to dedicate more time over the next month

• EC has been investigating questions from previous meetings and creating task list for multivaR

• BC shared RNOAA package, NOAA data access repo which team members will look at for any inspiration

NEXT STEPS

• Assembling csv's (BC/CL)
• AZMP search and vis code
• next meeting July 8th 2pm

casaultb/azmpdata documentation built on July 4, 2025, 11:04 a.m.