Freshwater ecosystems face escalating degradation worldwide, undermining ecological functions and services. Spawning habitats in large river systems are vital for sustaining fish populations, yet remain poorly characterized due to limitations in tagging-based approaches. This study reconstructed the spatial use of spawning grounds by Clupisoma yunnanensis in the upper Nu-Salween River through integration of otolith microchemical signatures with high-resolution water geochemical maps (chemoscapes). Otolith Mg:Ca, Mn:Ca, Sr:Ca, and Ba:Ca ratios were quantified via electron probe microanalysis, while corresponding water chemistry profiles were established using inductively coupled plasma mass spectrometry. Quantitative relationships between otolith and ambient element ratios were modeled using a random forest algorithm to reconstruct fish spawning ground distribution. Environmental niche analysis was employed for validation. Results demonstrated pronounced core-to-edge gradients in otolith elemental composition alongside marked spatiotemporal variation in riverine water chemistry, reflecting individual habitat trajectories across a heterogeneous chemical landscape. Downstream areas were identified as dominant spawning grounds with 98% predictive accuracy, and niche-based classification showed 83% concordance with conventional ecological data. This work establishes a novel integrated biogeochemical framework, combining otolith microchemistry, chemoscape modeling, and machine learning, for precise reconstruction of essential spawning habitats. The approach elucidates the reproductive ecology of C. yunnanensis and provides a scalable tool for habitat conservation in data-scarce freshwater systems worldwide.
Bang-Li et al. (Thu,) studied this question.