Coastal marine eutrophication footprint of reactive nitrogen: global emission-to-impact model chain and GLORIA MRIO satellites (2015–2022)
Year: 2026
Type: dataset
Abstract: Global model chain and dataset quantifying the coastal marine eutrophication footprint of reactive nitrogen (2015–2022) — from nitrogen emissions, through spatially explicit fate and effect modelling, to the final consumption ultimately driving the impacts. Three pathways are covered: agricultural leaching of waterborne N, airborne NH₃, and airborne NOₓ. Impacts are expressed in kg O₂-eq (marine oxygen depletion) and resolved to 305 coastal Large Marine Ecosystem sub-segments (cLME-ss), carrying the emitting country and the driving activity (crops, grazing, IPCC sectors). The three pathways are harmonised into one comparable database, converted into GLORIA v60 MRIO satellite matrices (305 LME-ss rows × 39,360 region–sector columns, per year and substance), and traced to final demand with a Leontief demand-pull model — including product-level detail for seven coral-reef LME sub-segments. Underlying models and data. The framework builds on and couples: the spatially differentiated marine-eutrophication assessment models of Cosme & Hauschild (2017), Cosme, Mayorga & Hauschild (2018) and Vea et al. (2022, 2024), with atmospheric characterisation following Bjørn et al. (2020); the IMAGE-GNM global nutrient model (Beusen et al., 2015; Beusen & Bouwman, 2022) for soil nitrogen budgets; EDGAR v8.1 gridded NH₃/NOₓ emissions (Crippa et al., 2024); the crop-specific nitrogen fertilisation dataset of Adalibieke et al. (2023); the GLORIA global environmentally-extended MRIO database, release 060 (University of Sydney IELab; cf. Lenzen et al., 2013; Malik et al., 2022). Contents: Shared GIS layers and grid–LME correspondences (folder 0); the crop and grazing leaching pipelines (A); the NH₃/NOₓ emission preparation and impact pipelines, including the characterisation-matrix builders CF_NHx.m / CF_NOx.m (B, C); the harmonised all-pathway database in Parquet (D); the GLORIA v60 satellite builder with row/column labels, exclusion tables and a conservation report (E); Every pipeline is a single linear, cell-structured script with built-in conservation checks; PIPELINE.md documents the execution order and run_pipeline.bat runs the whole chain. Modelling choices a reviewer should see are flagged REVIEW in the scripts and reports (receptor-country attribution assumption, the NOₓ 3C1b proxy split key, high-seas/Antarctica exclusions). Units & conventions. Impacts: kg O₂-eq. Emissions: kg NH₃, kg NOₓ (as NO₂), kg N (leaching), with all releases also normalised to kg N (NH₃ ×14/17, NOₓ ×14/46). Grids: agriculture 5 arcmin, EDGAR 0.1°, atmospheric fate 2°; receptors: cLME-ss labels <LME>_<sub-segment>. Environment. Python ≥3.10 (pandas, geopandas, numpy, scipy, xarray, h5py, rasterio, pyarrow, openpyxl); MATLAB R2021a+ for folder F. Not redistributed — with pointer files inside the archive (_DATA_NOT_INCLUDED_README.txt) explaining exactly how to obtain each: the TPDC crop-specific fertilisation HDF5 files (personal download credentials; doi:10.11888/Terre.tpdc.300446) — folder 2. N_application_rate; the IMAGE-GNM budget netCDFs (INMS data portal, registered users) — folder 1. Input_IMAGE_GNM; the GLORIA core matrices (T, Y, satellite accounts; IELab registration). Third-party data remain under their original licenses and citation requirements; cite the original sources when reusing them.
Language: English
Subfield: Environmental Engineering
Field: Environmental Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7204128958