Xiaolin Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Xiaolin Li has a total of 128 co-authors. Here are the top co-authors by number of shared publications:
- Minhan Dai
State Key Laboratory of Marine Environmental Science, College of Ocean and Earth Sciences, Xiamen University, Xiamen 361102, China
12 shared publications - Dalin Shi
State Key Laboratory of Marine Environmental Science Xiamen University Xiamen Fujian China
5 shared publications Kai Wu
State Key Laboratory of Marine Environmental Science College of Ocean and Earth Sciences Xiamen University Xiamen China
5 shared publicationsMinhan Dai
State Key Laboratory of Marine Environmental Science and College of Ocean and Earth Sciences, Xiamen University, Xiamen, China
5 shared publicationsZhimian Cao
State Key Laboratory of Marine Environmental Science, College of Ocean and Earth Sciences, Xiamen University, Xiamen 361102, China
5 shared publicationsDalin Shi
State Key Laboratory of Marine Environmental Science, College of Ocean and Earth Sciences, Xiamen University, Xiamen 361102, China
4 shared publications
Latest Publications
- Open Access
- 28 April 2026
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 18 (4) , 2951-2969
A historical nutrient dataset (1895–2024) for the North Pacific: reconstructed from machine learning and hydrographic observations
Nutrients play a critical role in oceanic primary productivity and the biological pump. However, compared to hydrographic parameters such as temperature and salinity, nutrient observations are limited due to their labor-intensive and costly measurements. Thus, nutrient observations are several orders of magnitude sparser than hydrographic observations. In this study, we first established a rigorous data quality control procedure to clean the hydrographic and nutrient (including NO , NO , DIP, and Si(OH)4) observations collected from World Ocean Database (WOD) and CLIVAR and Carbon Hydrographic Data Office (CCHDO) in the North Pacific. Subsequently, the cleaned and high-quality CCHDO dataset was used to train three machine learning models – Random Forest, Light Gradient Boosting Machine (LightGBM), and Gaussian Process Regression – to establish relationships between nutrient concentrations and key variables, including space coordinates (longitude, latitude, and depth), time variables (year and month), and water mass properties (indexed by potential temperature and salinity). Validation shows that the reconstruction closely matches the observations, with Root Mean Squared Errors (RMSEs) of <1.41, <0.071, <0.089 and <3.07 µmol kg−1 for NO , NO , DIP, and Si(OH)4, respectively. The validated models were then applied to reconstruct nutrient concentrations from the hydrographic observations in WOD, most of which lacked direct nutrient measurements. This resulted in ∼473 million reconstructed nutrient data points across 1.92 million stations for each nutrient, spanning from 1895 to 2024, representing a 2127- to 2393-fold increase compared to the original nutrient observations in the North Pacific (197 539 to 222 234). This new dataset will be valuable for studying nutrient transport and budgets, spinning up and validating ocean biogeochemical models, assessing long-term nutrients and their stoichiometric changes driven by anthropogenic forcing and climate change. The dataset generated in this study is openly available via Zenodo (https://doi.org/10.5281/zenodo.17451417) (Du et al., 2025).- 28 April 2026
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 18 (4) , 2951-2969
Open AccessA historical nutrient dataset (1895–2024) for the North Pacific: reconstructed from machine learning and hydrographic observations
Nutrients play a critical role in oceanic primary productivity and the biological pump. However, compared to hydrographic parameters such as temperature and salinity, nutrient observations are limited due to their labor-intensive and costly measurements. Thus, nutrient observations are several orders of magnitude sparser than hydrographic observations. In this study, we first established a rigorous data quality control procedure to clean the hydrographic and nutrient (including NO , NO , DIP, and Si(OH)4) observations collected from World Ocean Database (WOD) and CLIVAR and Carbon Hydrographic Data Office (CCHDO) in the North Pacific. Subsequently, the cleaned and high-quality CCHDO dataset was used to train three machine learning models – Random Forest, Light Gradient Boosting Machine (LightGBM), and Gaussian Process Regression – to establish relationships between nutrient concentrations and key variables, including space coordinates (longitude, latitude, and depth), time variables (year and month), and water mass properties (indexed by potential temperature and salinity). Validation shows that the reconstruction closely matches the observations, with Root Mean Squared Errors (RMSEs) of <1.41, <0.071, <0.089 and <3.07 µmol kg−1 for NO , NO , DIP, and Si(OH)4, respectively. The validated models were then applied to reconstruct nutrient concentrations from the hydrographic observations in WOD, most of which lacked direct nutrient measurements. This resulted in ∼473 million reconstructed nutrient data points across 1.92 million stations for each nutrient, spanning from 1895 to 2024, representing a 2127- to 2393-fold increase compared to the original nutrient observations in the North Pacific (197 539 to 222 234). This new dataset will be valuable for studying nutrient transport and budgets, spinning up and validating ocean biogeochemical models, assessing long-term nutrients and their stoichiometric changes driven by anthropogenic forcing and climate change. The dataset generated in this study is openly available via Zenodo (https://doi.org/10.5281/zenodo.17451417) (Du et al., 2025). - Open Access
- 20 February 2026
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 53 (4)
Seasonal Shift of Dissolved Organic Matter Stoichiometry in the Surface North Pacific Subtropical Gyre: Regulation by Enhanced N 2 Fixation and DOP Utilization
Dissolved organic matter (DOM) pool acts as both nutrient source and sink in the oligotrophic surface ocean. However, the drivers of seasonal variations in DOM stoichiometry and their relationship to nutrient cycling remain poorly understood. Here, we present a seasonal synthesis of dissolved organic nitrogen (DON) and phosphorus (DOP) distributions across the surface North Pacific Subtropical Gyre (NPSG) with associate biogeochemical measurements. In the central NPSG, summer exhibited elevated DON and depleted DOP, coinciding with a fourfold increase in N 2 ‐fixation rates compared with winter. This seasonal shift reflects preferential DOP utilization and accumulation of DON via N 2 ‐fixation. Mass and isotopic balance models confirm that DOP can supply about 40%–100% of summer phosphorus demand while about 50% of fixed nitrogen partitions into DON. These findings demonstrate that seasonal DOM stoichiometric flexibility encode dynamic nutrient cycling, highlighting the fundamental role of DOM transformation in regulating ocean biogeochemical processes.
- 20 February 2026
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 53 (4)
Open AccessSeasonal Shift of Dissolved Organic Matter Stoichiometry in the Surface North Pacific Subtropical Gyre: Regulation by Enhanced N 2 Fixation and DOP Utilization
Dissolved organic matter (DOM) pool acts as both nutrient source and sink in the oligotrophic surface ocean. However, the drivers of seasonal variations in DOM stoichiometry and their relationship to nutrient cycling remain poorly understood. Here, we present a seasonal synthesis of dissolved organic nitrogen (DON) and phosphorus (DOP) distributions across the surface North Pacific Subtropical Gyre (NPSG) with associate biogeochemical measurements. In the central NPSG, summer exhibited elevated DON and depleted DOP, coinciding with a fourfold increase in N 2 ‐fixation rates compared with winter. This seasonal shift reflects preferential DOP utilization and accumulation of DON via N 2 ‐fixation. Mass and isotopic balance models confirm that DOP can supply about 40%–100% of summer phosphorus demand while about 50% of fixed nitrogen partitions into DON. These findings demonstrate that seasonal DOM stoichiometric flexibility encode dynamic nutrient cycling, highlighting the fundamental role of DOM transformation in regulating ocean biogeochemical processes.
- Open Access
- 17 January 2026
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 53 (2)
Hidden Bioavailable Dissolved Organic Matter in the Deep Northwestern Pacific Ocean
The deep ocean is widely viewed as a stable reservoir of dissolved organic carbon (DOC) that is highly resistant to microbial degradation. However, this paradigm may be oversimplified. In a cross‐latitude survey conducted in the northwestern Pacific Ocean (155°E, 28.5°N–41.5°N), bulk DOC concentrations were found to be relatively constant below 1,000 m with little variation across latitude. Despite this stability, compound‐specific analyses revealed unexpectedly high abundance of hydrolyzable amino acids, up to 1.1% of DOC, at depths of 1,000–4,000 m between 30°N and 36°N. This amino acid enrichment, doubling the typical levels observed in other deep‐sea basins and comparable to upper ocean values, reflected episodic inputs of semi‐labile DOC likely derived from actively settling fresh particles. These findings reveal hidden reservoirs of bioavailable DOC in the deep ocean, a feature not captured by bulk analysis, challenging the conventional view of deep‐sea DOC as predominantly refractory and stable.
- 17 January 2026
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 53 (2)
Open AccessHidden Bioavailable Dissolved Organic Matter in the Deep Northwestern Pacific Ocean
The deep ocean is widely viewed as a stable reservoir of dissolved organic carbon (DOC) that is highly resistant to microbial degradation. However, this paradigm may be oversimplified. In a cross‐latitude survey conducted in the northwestern Pacific Ocean (155°E, 28.5°N–41.5°N), bulk DOC concentrations were found to be relatively constant below 1,000 m with little variation across latitude. Despite this stability, compound‐specific analyses revealed unexpectedly high abundance of hydrolyzable amino acids, up to 1.1% of DOC, at depths of 1,000–4,000 m between 30°N and 36°N. This amino acid enrichment, doubling the typical levels observed in other deep‐sea basins and comparable to upper ocean values, reflected episodic inputs of semi‐labile DOC likely derived from actively settling fresh particles. These findings reveal hidden reservoirs of bioavailable DOC in the deep ocean, a feature not captured by bulk analysis, challenging the conventional view of deep‐sea DOC as predominantly refractory and stable.
- Open Access
- 16 January 2026
- Published by Copernicus GmbH
- p. 1-34
Comment on essd-2025-654
Nutrients play a critical role in oceanic primary productivity and the biological pump. However, compared to hydrographic parameters such as temperature and salinity, nutrient observations are limited due to their labor-intensive and costly measurements. Thus, nutrient observations are several orders of magnitude sparser than hydrographic observations. In this study, we first established a rigorous data quality control procedure to clean the hydrographic and nutrient (including NO₃⁻, NO₂⁻, DIP, and Si(OH)₄) observations collected from World Ocean Database (WOD) and CLIVAR and Carbon Hydrographic Data Office (CCHDO) in the North Pacific. Subsequently, the cleaned and high-quality CCHDO dataset was used to train three machine learning models – Random Forest, Light Gradient Boosting Machine (LightGBM), and Gaussian Process Regression – to establish relationships between nutrient concentrations and key variables, including space coordinates (longitude, latitude, and depth), time variables (year and month), and water mass properties (indexed by potential temperature and salinity). Validation shows that the reconstruction closely matches the observations, with RMSEs of -1 for NO₃⁻, NO₂⁻, DIP, and Si(OH)₄, respectively. The validated models were then applied to reconstruct nutrient concentrations from the hydrographic observations in WOD, most of which lacked direct nutrient measurements. This resulted in ~473 million reconstructed nutrient data points across 1.92 million stations for each nutrient, spanning from 1895 to 2024, representing a 2,127 to 2,393-fold increase compared to the original nutrient observations in the North Pacific (197,539 to 222,234). This new dataset will be valuable for studying nutrient variability under climate change and anthropogenic influences, and for providing transient boundary conditions in ocean biogeochemical models. The dataset generated in this study is openly available on Zenodo at https://zenodo.org/records/17451417.
- 16 January 2026
- Published by Copernicus GmbH
- p. 1-34
Open AccessComment on essd-2025-654
Nutrients play a critical role in oceanic primary productivity and the biological pump. However, compared to hydrographic parameters such as temperature and salinity, nutrient observations are limited due to their labor-intensive and costly measurements. Thus, nutrient observations are several orders of magnitude sparser than hydrographic observations. In this study, we first established a rigorous data quality control procedure to clean the hydrographic and nutrient (including NO₃⁻, NO₂⁻, DIP, and Si(OH)₄) observations collected from World Ocean Database (WOD) and CLIVAR and Carbon Hydrographic Data Office (CCHDO) in the North Pacific. Subsequently, the cleaned and high-quality CCHDO dataset was used to train three machine learning models – Random Forest, Light Gradient Boosting Machine (LightGBM), and Gaussian Process Regression – to establish relationships between nutrient concentrations and key variables, including space coordinates (longitude, latitude, and depth), time variables (year and month), and water mass properties (indexed by potential temperature and salinity). Validation shows that the reconstruction closely matches the observations, with RMSEs of -1 for NO₃⁻, NO₂⁻, DIP, and Si(OH)₄, respectively. The validated models were then applied to reconstruct nutrient concentrations from the hydrographic observations in WOD, most of which lacked direct nutrient measurements. This resulted in ~473 million reconstructed nutrient data points across 1.92 million stations for each nutrient, spanning from 1895 to 2024, representing a 2,127 to 2,393-fold increase compared to the original nutrient observations in the North Pacific (197,539 to 222,234). This new dataset will be valuable for studying nutrient variability under climate change and anthropogenic influences, and for providing transient boundary conditions in ocean biogeochemical models. The dataset generated in this study is openly available on Zenodo at https://zenodo.org/records/17451417.
- Open Access
- 12 November 2025
- Published by Copernicus GmbH in EGUsphere
- Vol. 2025, 1-34
A historical nutrient dataset (1895–2024) for the North Pacific: reconstructed from machine learning and hydrographic observations
Nutrients play a critical role in oceanic primary productivity and the biological pump. However, compared to hydrographic parameters such as temperature and salinity, nutrient observations are limited due to their labor-intensive and costly measurements. Thus, nutrient observations are several orders of magnitude sparser than hydrographic observations. In this study, we first established a rigorous data quality control procedure to clean the hydrographic and nutrient (including NO₃⁻, NO₂⁻, DIP, and Si(OH)₄) observations collected from World Ocean Database (WOD) and CLIVAR and Carbon Hydrographic Data Office (CCHDO) in the North Pacific. Subsequently, the cleaned and high-quality CCHDO dataset was used to train three machine learning models – Random Forest, Light Gradient Boosting Machine (LightGBM), and Gaussian Process Regression – to establish relationships between nutrient concentrations and key variables, including space coordinates (longitude, latitude, and depth), time variables (year and month), and water mass properties (indexed by potential temperature and salinity). Validation shows that the reconstruction closely matches the observations, with RMSEs of -1 for NO₃⁻, NO₂⁻, DIP, and Si(OH)₄, respectively. The validated models were then applied to reconstruct nutrient concentrations from the hydrographic observations in WOD, most of which lacked direct nutrient measurements. This resulted in ~473 million reconstructed nutrient data points across 1.92 million stations for each nutrient, spanning from 1895 to 2024, representing a 2,127 to 2,393-fold increase compared to the original nutrient observations in the North Pacific (197,539 to 222,234). This new dataset will be valuable for studying nutrient variability under climate change and anthropogenic influences, and for providing transient boundary conditions in ocean biogeochemical models. The dataset generated in this study is openly available on Zenodo at https://zenodo.org/records/17451417.
- 12 November 2025
- Published by Copernicus GmbH in EGUsphere
- Vol. 2025, 1-34
Open AccessA historical nutrient dataset (1895–2024) for the North Pacific: reconstructed from machine learning and hydrographic observations
Nutrients play a critical role in oceanic primary productivity and the biological pump. However, compared to hydrographic parameters such as temperature and salinity, nutrient observations are limited due to their labor-intensive and costly measurements. Thus, nutrient observations are several orders of magnitude sparser than hydrographic observations. In this study, we first established a rigorous data quality control procedure to clean the hydrographic and nutrient (including NO₃⁻, NO₂⁻, DIP, and Si(OH)₄) observations collected from World Ocean Database (WOD) and CLIVAR and Carbon Hydrographic Data Office (CCHDO) in the North Pacific. Subsequently, the cleaned and high-quality CCHDO dataset was used to train three machine learning models – Random Forest, Light Gradient Boosting Machine (LightGBM), and Gaussian Process Regression – to establish relationships between nutrient concentrations and key variables, including space coordinates (longitude, latitude, and depth), time variables (year and month), and water mass properties (indexed by potential temperature and salinity). Validation shows that the reconstruction closely matches the observations, with RMSEs of -1 for NO₃⁻, NO₂⁻, DIP, and Si(OH)₄, respectively. The validated models were then applied to reconstruct nutrient concentrations from the hydrographic observations in WOD, most of which lacked direct nutrient measurements. This resulted in ~473 million reconstructed nutrient data points across 1.92 million stations for each nutrient, spanning from 1895 to 2024, representing a 2,127 to 2,393-fold increase compared to the original nutrient observations in the North Pacific (197,539 to 222,234). This new dataset will be valuable for studying nutrient variability under climate change and anthropogenic influences, and for providing transient boundary conditions in ocean biogeochemical models. The dataset generated in this study is openly available on Zenodo at https://zenodo.org/records/17451417.
- Open Access
- 25 November 2024
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Oceans
- Vol. 129 (11)
Elevated Heterotrophic Prokaryotic Production Supported by Low Nucleic Acid Prokaryotes at a Cyclonic Eddy Edge in the Northwest Pacific
The cyclonic eddy uplifts nutrient‐rich seawater into the euphotic zone, typically directly enhancing phytoplankton abundance and primary production. However, its impact on heterotrophic prokaryotic production (HPP) remains unclear due to the complex interplay of multiple indirect factors governing this process. Here, we conducted a comprehensive investigation of the distribution of picophytoplankton and heterotrophic prokaryotes, prokaryotic community structure, and HPP within a cyclonic eddy in the western North Pacific subtropical gyre. The results indicated that despite the higher abundance of picophytoplankton accompanied by nutrient upwelling at the eddy core compared to the edge, higher levels of HPP were observed at the eddy edge between 100 and 200 m, consistent with the distribution of the low nucleic acid content (LNA) prokaryotes. The significant positive correlation between HPP and the proportion of LNA group in total heterotrophic prokaryotes suggested a primary contribution from the LNA group over the high nucleic acid content (HNA) group. SAR11, a typical member of the LNA group, may primarily contribute to the elevated HPP observed at the eddy edge. The changes in temperature, nutrients, and light intensity induced by the cyclonic eddy may significantly influence the distribution and activity of HNA and LNA groups, potentially exerting a greater impact on HPP compared to phytoplankton‐related factors. These findings contribute to understanding the underlying mechanisms of HPP responses to cyclonic eddies in the oligotrophic open ocean.
- 25 November 2024
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Oceans
- Vol. 129 (11)
Open AccessElevated Heterotrophic Prokaryotic Production Supported by Low Nucleic Acid Prokaryotes at a Cyclonic Eddy Edge in the Northwest Pacific
The cyclonic eddy uplifts nutrient‐rich seawater into the euphotic zone, typically directly enhancing phytoplankton abundance and primary production. However, its impact on heterotrophic prokaryotic production (HPP) remains unclear due to the complex interplay of multiple indirect factors governing this process. Here, we conducted a comprehensive investigation of the distribution of picophytoplankton and heterotrophic prokaryotes, prokaryotic community structure, and HPP within a cyclonic eddy in the western North Pacific subtropical gyre. The results indicated that despite the higher abundance of picophytoplankton accompanied by nutrient upwelling at the eddy core compared to the edge, higher levels of HPP were observed at the eddy edge between 100 and 200 m, consistent with the distribution of the low nucleic acid content (LNA) prokaryotes. The significant positive correlation between HPP and the proportion of LNA group in total heterotrophic prokaryotes suggested a primary contribution from the LNA group over the high nucleic acid content (HNA) group. SAR11, a typical member of the LNA group, may primarily contribute to the elevated HPP observed at the eddy edge. The changes in temperature, nutrients, and light intensity induced by the cyclonic eddy may significantly influence the distribution and activity of HNA and LNA groups, potentially exerting a greater impact on HPP compared to phytoplankton‐related factors. These findings contribute to understanding the underlying mechanisms of HPP responses to cyclonic eddies in the oligotrophic open ocean.
- 17 January 2024
- Published by Elsevier in Chemosphere
- Vol. 351, 141225
The photo- and microbial degradation kinetics and pathways of sulfadoxine in seawater elucidated by liquid chromatography coupled with time-of-flight mass spectrometry
- 17 January 2024
- Published by Elsevier in Chemosphere
- Vol. 351, 141225
The photo- and microbial degradation kinetics and pathways of sulfadoxine in seawater elucidated by liquid chromatography coupled with time-of-flight mass spectrometry
- Open Access
- 18 August 2023
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 9 (33) , eade2078
Reduced nitrite accumulation at the primary nitrite maximum in the cyclonic eddies in the western North Pacific subtropical gyre
Nitrite, an intermediate product of the oxidation of ammonia to nitrate (nitrification), accumulates in upper oceans, forming the primary nitrite maximum (PNM). Nitrite concentrations in the PNM are relatively low in the western North Pacific subtropical gyre (wNPSG), where eddies are frequent and intense. To explain these low nitrite concentrations, we investigated nitrification in cyclonic eddies in the wNPSG. We detected relatively low half-saturation constants (i.e., high substrate affinities) for ammonia and nitrite oxidation at 150 to 200 meter water depth. Eddy-induced displacement of high-affinity nitrifiers and increased substrate supply enhanced ammonia and nitrite oxidation, depleting ambient substrate concentrations in the euphotic zone. Nitrite oxidation is more strongly enhanced by the cyclonic eddies than ammonia oxidation, reducing concentrations and accelerating the turnover of nitrite in the PNM. These findings demonstrate a spatial decoupling of the two steps of nitrification in response to mesoscale processes and provide insights into physical-ecological controls on the PNM.
- 18 August 2023
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 9 (33) , eade2078
Open AccessReduced nitrite accumulation at the primary nitrite maximum in the cyclonic eddies in the western North Pacific subtropical gyre
Nitrite, an intermediate product of the oxidation of ammonia to nitrate (nitrification), accumulates in upper oceans, forming the primary nitrite maximum (PNM). Nitrite concentrations in the PNM are relatively low in the western North Pacific subtropical gyre (wNPSG), where eddies are frequent and intense. To explain these low nitrite concentrations, we investigated nitrification in cyclonic eddies in the wNPSG. We detected relatively low half-saturation constants (i.e., high substrate affinities) for ammonia and nitrite oxidation at 150 to 200 meter water depth. Eddy-induced displacement of high-affinity nitrifiers and increased substrate supply enhanced ammonia and nitrite oxidation, depleting ambient substrate concentrations in the euphotic zone. Nitrite oxidation is more strongly enhanced by the cyclonic eddies than ammonia oxidation, reducing concentrations and accelerating the turnover of nitrite in the PNM. These findings demonstrate a spatial decoupling of the two steps of nitrification in response to mesoscale processes and provide insights into physical-ecological controls on the PNM.