PDO Index Details

[SOLV-Code]

NOTE: This issue is about the source data used for the monthly PDO index. There's a separate issue here regarding alternative covariates created from that source data, as well as a note comparing alternative pdo-based covariates

Data Source

• Why does the PDO source data here end in Sep 2018? Another source here, called ersstv5 goes to Jan 2024, but has different values than the source provided in the 2021 FC paper. This source called OI SST v2 has the same values as the one linked in the paper, and goes up to 2022, so that's the one we used.

• the pacea package uses the other version called ersstv5

Warming Trend Removed?

The PDO index source file states that the underlying warming trend has been removed from the index to focus on anomalies.

• [ ] For the purpose of salmon forecasting, should a separate warming index be included as a stand-alone environmental covariate?
• [ ] Or is the warming captured in the more localized SST covariate based on lighthouse data?
• [ ] Is the warming trend handled differently in OI SST v2 vs. ER SST v5?

Started a note re: Trends vs. Anomalies

[SOLV-Code]

Found some materials that might help picking one over the other:

• NOAA Blog Post : Exactly the same, but completely different has 3 key points:
  • Use the ER SST version for historical comparison of "event strength"
  • Cut off in 1950 given earlier estimates are unreliable
• SST Data Sets Overview
• Publication using multiple versions
• PDO definitions

[SOLV-Code]

Next Steps

• [ ] reach out to forecast folks and pacea folks.
  • Try to agree on a single version for moving forward
  • discuss re: warming trend removed, and implications
  • then make sure choice of version and detrending issue are documented in all 3 places (FC document, pacea helpfile, PSEC data entry for both)