Source Data and Code of Salinity Thresholds Shape Global Patterns of Freshwater Biodiversity.

This document comprises two parts: the dataset compiled for this study and the corresponding code and source data used for the analysis.The <b>code files</b> directory contains the source data and R scripts (in Tinn-R format) for each section of the analysis. Please set the working directory and install the required R packages before running the code.Code file details as follow:Step1_Calculation of rarefied richness; Step2_Classfication of salinity gradient; Step3_Biodiversity model;Step4_Salinity threshold of diversity metrics and Beta diveristy;Step5_Relative abundance along salinity gradient.The <b>dataset</b> compiles benthic macroinvertebrate community data and corresponding salinity measurements from 701 inland lakes sampled between 1972 and 2022, across multiple continents. It includes sampling metadata and SAD (species abundance distribution) data. Salinity (or conductivity) values measured at the time of biological sampling. The dataset was assembled from publicly available databases (e.g., the US EPA's National Lakes Assessment, PANGAEA), published literature, and co-author contributions. All data sources were selected based on consistent and comparable sampling methodologies, ensuring robustness and compatibility across studies.Each sample includes information on the sampling method, date, location, SAD data, and physicochemical conditions. A rarefaction-based standardization procedure was applied to ensure comparability across studies with different sampling effects.This dataset enables researchers to explore macroinvertebrate diversity responses to freshwater salinization, assess biodiversity loss under anthropogenic pressure, and support ecological modeling or conservation planning. It is particularly relevant for global change ecology, freshwater biodiversity research, and ecological threshold detection.The dataset is shared under the Creative Commons Attribution 4.0 International License (CC BY 4.0). Users are free to reuse the data with proper citation. Ethical and legal considerations were fully addressed: all data were either publicly available or shared with permission by contributing authors. Users should cite the original paper and this dataset if used in derivative work. Some data sources may have specific citation requirements; detailed references and access information for each data source are provided in Supplementary Table1. Users are encouraged to cite original sources where indicated. No sensitive or personally identifiable information is included.For more detailed information or any questions please contact <b>email</b>: <b>[e-mail]</b>