Replication package for "Sustainability made elsewhere: larval connectivity and source accountability in transboundary fish stocks"

This archive is the replication package for the manuscript "Sustainability made elsewhere: larval connectivity and source accountability in transboundary fish stocks" (Yiran Li, Graduate School of Economics, Kobe University, Kobe, Japan). The study asks whether national fish-stock sustainability — measured by assessment-based B/BMSY from the RAM Legacy Stock Assessment Database — propagates across borders through directed ocean-current larval connectivity, and what this implies for SDG 14.4 accountability among states that share stocks. Treating a directed larval-connectivity network as a spatial weight, it estimates spatial-panel spillovers in stock status among sovereign coastal nations over 2001–2023, with country and year fixed effects and wild cluster bootstrap inference. A nation's stock status rises with that of its upstream larval-source nations, whereas the downstream (outflow) direction is not significant. The estimate is examined against a geographic-proximity control, random-network placebos, adult-migration exclusion, a pelagic-larval-duration (PLD) scaling check, and a set of aggregation and taxonomic-composition robustness checks; consistent with the manuscript, it is reported as characterising well-assessed, predominantly temperate demersal fisheries rather than a universal global regularity. Contents:- code/ — two self-contained Python scripts that reproduce, from the packaged data using paths relative to the archive root, the main directional-spillover table (manuscript Table 1) and the aggregation/composition robustness table (Supplementary Table S6).- data/ — processed RAM stock-status panels and stock metadata; directed larval-connectivity weights and country-coordinate support; curated PLD data and mechanism-check outputs; connectivity units and net larval-source index; species-coverage summary.- outputs/ — final main-text figures (PNG and PDF), a diagnostic supplementary figure, and result tables. Reproduction (Python 3.10+, numpy, pandas), from the archive root: python code/reproduce_table1_from_package.py python code/reproduce_robustness_tableS6.py Data provenance: all underlying sources are public — the RAM Legacy Stock Assessment Database v4.66; a global coastal ocean-connectivity product (coastalNet) for the directed larval weights; the Ramesh et al. generated larval-dynamics dataset and the SEANOE early-life traits dataset for pelagic larval duration; and Sea Around Us catch/value extracts together with Natural Earth and Exclusive Economic Zone files for supporting diagnostics and mapping.