Pooling locations raised the estimated overlap gain for 25 plankton codes at 17 North Sea cells, 1958–1980 to 1997–2019

We tested whether a reported increase in monthly plankton overlap also occurred when the analysis kept location. We used public CPR records for 14 diatom codes and 11 copepod codes at 17 fixed one-degree North Sea cells. The comparison used 7,099 samples in 1958–1980 and 9,402 in 1997–2019. Each code had a normalized distribution of mean recorded counts across cells and months. Mean pair overlap increased by 0.0937 after cells were pooled within months. It increased by 0.0260 when cells remained separate. The nominal 95% interval for the latter change was −0.0136 to +0.0749, from shared three-year calendar blocks. The difference between the gains was 0.0677, with an interval of 0.0242 to 0.0912. An exact identity links this difference to cancellation across cells. Equal cell-month sample counts gave the same pattern. The difference was less clear with occurrence records, log counts, or the wider 43-code list. Thus the supported result concerns the selected codes and counted records. Joint overlap still pools days and years and does not measure encounters or feeding. The package preserves source data, provenance, a local plan, code, all results, and figures. An independent source rebuild passed. A complete offline repeat matched 16 numerical files.

plankton · Continuous Plankton Recorder · North Sea · spatial aggregation · seasonal overlap · diatoms · copepods · reproducible analysis