Increased climate variability and extremes amplify risks of coastal hypoxia worldwide: Implications of enhanced basin memory effects on river dissolved nitrogen in the GFDL Earth system modeling framework

Abstract:

Spreading coastal hypoxia is a pressing global problem. One of the known precursors for the hypoxia is dissolved nitrogen (DN) loads from large rivers. In general, river DN loads increase with precipitation or river flows (i.e., flow-load relationships). However, our study, with a process-based model LM3-TAN (Terrestrial and Aquatic Nitrogen), used both reported and simulated data, and demonstrates significantly high Susquehanna River DN-load anomalies from the flow-load relationships. We found that such river DN-load anomalies are attributed to high rainfalls following multi-year dry spells (or climate variability and extremes). One of the most plausible mechanisms is accumulated nitrogen (N) throughout the dry spells, which is, in turn, washed out from the basin into rivers via the subsequent high rainfalls; we call this “basin memory effects”. To further investigate this mechanism, we developed a global implementation of LM3-TAN, which was verified by comparing reported and simulated global and regional (world’s 39 large river basins) terrestrial-aquatic N budgets (e.g., river DN loads, soil and river denitrification, harvest, vegetation and soil N storage) for preindustrial (1860) and contemporary (1995) years. Our preliminary analysis of simulated river flow and DN-load variability highlights that increased climate variability and extremes under changing climate enhance the “basin memory effects”, and which significantly increase high river DN-load anomalies and risks of coastal hypoxia worldwide.