RAM Legacy Stock Assessment Database version 4.64

Extended RAM Legacy Stock Assessment Database version 4.491

Please be aware that this ‘mdl’ ('model fits included') version 4.491 of the RAM Legacy Stock Assessment Database (RAMLDB) additionally includes data not contained in the assessments themselves, but generated from assessment data. Generated data are used to fill in gaps where possible, for example, if stock assessments do not include reference point estimates. Supplemented data include calculations from assessment values, conversions of biomass variables, and reference points estimated post-hoc with surplus production models. The ‘Documents’ folder includes documentation files on the methods used for these calculations, conversions and model fitting. When assessment data are available, they are used preferentially in all cases.

Although we identify data that were supplemented from data that were derived from stock assessments, some collector variables in RAMLDB include both of these data sources (with assessment values preferred). We recommend that the average user does not use this 'mdl' version of RAMLDB because of the additional assumptions and calculations involved. Instead, to avoid confusion between assessment-derived values and supplemented values, we recommend using the ‘asmt’ ('assessment data only') version of RAMLDB version 4.491. This 'asmt' version (doi 10.5281/zenodo.3676088 ) is also available in our Zenodo repository: https://zenodo.org/communities/rlsadb/

China contributes the largest catches to global marine wild-capture fisheries. The majority of them are harvested from China Seas which are highly productive, but are facing heavy fisheries exploitation. The status of exploited fisheries stocks in China Seas have remained largely unknown due to severe data-limited conditions, which hindered their sustainable use and effective management. Although the off-the-shelf use of catch-only methods (COMs) has been cautioned because of their poor estimation performance, such methods have been increasingly applied to infer the status of exploited stocks in China Seas without performance evaluation. In this study, we established an empirical approach to evaluate the performance of a suite of COMs in predicting stock biomass status for the data-limited fisheries in the East China Sea (ECS) from data-rich stocks with similar characteristics in the RAM Legacy Stock Assessment Database (RLSADB). The results confirmed that ensemble approaches performed better than the individual COMs in estimating the mean of stock biomass status for data-rich stocks selected from RLSADB. By contrast, mechanistic COMs demonstrated more accurate estimates when predicting the trend of stock biomass status. The stock status of commercial fisheries in ECS estimated by three mechanistic COMs (Catch-MSY, CMSY, and OCOM) was likely too optimistic for most species. We suggest that China establish its national database and develop and implement regular monitoring programs to satisfy formal statistical stock assessment for its coastal fisheries.

The stock-recruitment data were extracted from the RAM Legacy Stock Assessment Database. The data set contains the spawning stock biomass (SSB) and recruit time series of the studied populations.

The S-R models were fitted to the data with the Bayesian online change point detection method (BOCPD), combined with simulation-based filtering. The data were divided into segments by calculating their most likely segmentation (MLS). The methods are described more in detail in the supplementary material of the article.