Zhibo Shao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zhibo Shao has a total of 101 co-authors. Here are the top co-authors by number of shared publications:
- Ya-Wei Luo
State Key Laboratory of Marine Environmental Science and College of Ocean and Earth Sciences, Xiamen University, Xiamen, China
6 shared publications Hua Wang
State Key Laboratory of Marine Environmental Science and College of Ocean and Earth Sciences, Xiamen University, Xiamen, Fujian, China
3 shared publicationsLice Wang
State Key Laboratory of Marine Environmental Science and College of Ocean and Earth Sciences, Xiamen University, Xiamen, Fujian, China
3 shared publicationsWeicheng Luo
State Key Laboratory of Marine Environmental Science and College of Ocean and Earth Sciences, Xiamen University, Xiamen, Fujian, China
3 shared publications- Yangchun Xu
State Key Laboratory of Marine Environmental Science, College of Ocean and Earth Sciences, Xiamen University, Xiamen, 361102, China
3 shared publications Yuhong Huang
State Key Laboratory of Marine Environmental Science and College of Ocean and Earth Sciences, Xiamen University, Xiamen, Fujian, China
3 shared publications
Latest Publications
- Open Access
- 29 April 2026
- Published by California Digital Library (CDL) in EarthArXiv
Evaluating Sampling Bias and Model Uncertainty in Species Distribution Models of Marine Plankton Using Virtual Ecosystem Data
Understanding the biodiversity and biogeography of plankton in the ocean is essential for predicting responses to environmental changes and informing ocean conservation and management strategies. Species distribution models (SDMs) are a pivotal tool in this regard. This study used data from a global marine ecosystem model as a testbed to assess the reliability of various SDMs, including Generalized Linear Model (GLM), Generalized Additive Model (GAM), Random Forest (RF), Boosted Regression Trees (BRT) and Artificial Neural Network (ANN). We used artificial datasets to replicate the sampling patterns of three datasets: a compiled dataset of global scope, the Tara Ocean dataset, and the Atlantic Meridional Transect (AMT) project. Our findings indicate that tree-based algorithms, RF and BRT, exhibit better predictive accuracy and stability compared to GLM, GAM, and ANN, especially when trained with more spatially resolved datasets. We highlight the significant influence of sampling bias on model performance, with models trained on more comprehensive global datasets outperforming those trained on more latitudinally and longitudinally biased data respectively (Tara and AMT). Furthermore, we demonstrate that broad spatial coverage is a more critical determinant of predictive skill than sample size alone, as simply increasing sampling density within a biased region is insufficient to overcome poor spatial representation. Overall, this research underscores the necessity of careful consideration of sampling strategies and model selection in plankton species distribution modelling.
- 29 April 2026
- Published by California Digital Library (CDL) in EarthArXiv
Open AccessEvaluating Sampling Bias and Model Uncertainty in Species Distribution Models of Marine Plankton Using Virtual Ecosystem Data
Understanding the biodiversity and biogeography of plankton in the ocean is essential for predicting responses to environmental changes and informing ocean conservation and management strategies. Species distribution models (SDMs) are a pivotal tool in this regard. This study used data from a global marine ecosystem model as a testbed to assess the reliability of various SDMs, including Generalized Linear Model (GLM), Generalized Additive Model (GAM), Random Forest (RF), Boosted Regression Trees (BRT) and Artificial Neural Network (ANN). We used artificial datasets to replicate the sampling patterns of three datasets: a compiled dataset of global scope, the Tara Ocean dataset, and the Atlantic Meridional Transect (AMT) project. Our findings indicate that tree-based algorithms, RF and BRT, exhibit better predictive accuracy and stability compared to GLM, GAM, and ANN, especially when trained with more spatially resolved datasets. We highlight the significant influence of sampling bias on model performance, with models trained on more comprehensive global datasets outperforming those trained on more latitudinally and longitudinally biased data respectively (Tara and AMT). Furthermore, we demonstrate that broad spatial coverage is a more critical determinant of predictive skill than sample size alone, as simply increasing sampling density within a biased region is insufficient to overcome poor spatial representation. Overall, this research underscores the necessity of careful consideration of sampling strategies and model selection in plankton species distribution modelling.
- Top Cited PapersOpen Access
- 15 August 2023
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 15 (8) , 3673-3709
Global oceanic diazotroph database version 2 and elevated estimate of global oceanic N2 fixation
Marine diazotrophs convert dinitrogen (N2) gas into bioavailable nitrogen (N), supporting life in the global ocean. In 2012, the first version of the global oceanic diazotroph database (version 1) was published. Here, we present an updated version of the database (version 2), significantly increasing the number of in situ diazotrophic measurements from 13 565 to 55 286. Data points for N2 fixation rates, diazotrophic cell abundance, and nifH gene copy abundance have increased by 184 %, 86 %, and 809 %, respectively. Version 2 includes two new data sheets for the nifH gene copy abundance of non-cyanobacterial diazotrophs and cell-specific N2 fixation rates. The measurements of N2 fixation rates approximately follow a log-normal distribution in both version 1 and version 2. However, version 2 considerably extends both the left and right tails of the distribution. Consequently, when estimating global oceanic N2 fixation rates using the geometric means of different ocean basins, version 1 and version 2 yield similar rates (43–57 versus 45–63 Tg N yr−1; ranges based on one geometric standard error). In contrast, when using arithmetic means, version 2 suggests a significantly higher rate of 223±30 Tg N yr−1 (mean ± standard error; same hereafter) compared to version 1 (74±7 Tg N yr−1). Specifically, substantial rate increases are estimated for the South Pacific Ocean (88±23 versus 20±2 Tg N yr−1), primarily driven by measurements in the southwestern subtropics, and for the North Atlantic Ocean (40±9 versus 10±2 Tg N yr−1). Moreover, version 2 estimates the N2 fixation rate in the Indian Ocean to be 35±14 Tg N yr−1, which could not be estimated using version 1 due to limited data availability. Furthermore, a comparison of N2 fixation rates obtained through different measurement methods at the same months, locations, and depths reveals that the conventional 15N2 bubble method yields lower rates in 69 % cases compared to the new 15N2 dissolution method. This updated version of the database can facilitate future studies in marine ecology and biogeochemistry. The database is stored at the Figshare repository (https://doi.org/10.6084/m9.figshare.21677687; Shao et al., 2022).
- 15 August 2023
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 15 (8) , 3673-3709
Top Cited PapersOpen AccessGlobal oceanic diazotroph database version 2 and elevated estimate of global oceanic N2 fixation
Marine diazotrophs convert dinitrogen (N2) gas into bioavailable nitrogen (N), supporting life in the global ocean. In 2012, the first version of the global oceanic diazotroph database (version 1) was published. Here, we present an updated version of the database (version 2), significantly increasing the number of in situ diazotrophic measurements from 13 565 to 55 286. Data points for N2 fixation rates, diazotrophic cell abundance, and nifH gene copy abundance have increased by 184 %, 86 %, and 809 %, respectively. Version 2 includes two new data sheets for the nifH gene copy abundance of non-cyanobacterial diazotrophs and cell-specific N2 fixation rates. The measurements of N2 fixation rates approximately follow a log-normal distribution in both version 1 and version 2. However, version 2 considerably extends both the left and right tails of the distribution. Consequently, when estimating global oceanic N2 fixation rates using the geometric means of different ocean basins, version 1 and version 2 yield similar rates (43–57 versus 45–63 Tg N yr−1; ranges based on one geometric standard error). In contrast, when using arithmetic means, version 2 suggests a significantly higher rate of 223±30 Tg N yr−1 (mean ± standard error; same hereafter) compared to version 1 (74±7 Tg N yr−1). Specifically, substantial rate increases are estimated for the South Pacific Ocean (88±23 versus 20±2 Tg N yr−1), primarily driven by measurements in the southwestern subtropics, and for the North Atlantic Ocean (40±9 versus 10±2 Tg N yr−1). Moreover, version 2 estimates the N2 fixation rate in the Indian Ocean to be 35±14 Tg N yr−1, which could not be estimated using version 1 due to limited data availability. Furthermore, a comparison of N2 fixation rates obtained through different measurement methods at the same months, locations, and depths reveals that the conventional 15N2 bubble method yields lower rates in 69 % cases compared to the new 15N2 dissolution method. This updated version of the database can facilitate future studies in marine ecology and biogeochemistry. The database is stored at the Figshare repository (https://doi.org/10.6084/m9.figshare.21677687; Shao et al., 2022).
- Open Access
- 19 January 2023
- Published by Copernicus GmbH in EGUsphere
- Vol. 2023, 1-44
Version 2 of the global oceanic diazotroph database
Marine diazotrophs convert dinitrogen (N2) in seawater into bioavailable nitrogen (N), contributing approximately half of the external input of bioavailable N to the global ocean. A global oceanic diazotroph database was previously published in 2012. Here, we compiled version 2 of the database by adding 23,095 in situ measurements of marine diazotrophic abundance and N2 fixation rates published in the past decade, increasing the number of N2 fixation rates and microscopic and qPCR-based diazotrophic abundance data by 140 %, 26 % and 443 %, respectively. Although the updated database expanded spatial coverage considerably, particularly in the Indian Ocean, the data distribution was still not uniform and most data were sampled in the surface Pacific and Atlantic Oceans. By summing the arithmetic means of the N2 fixation rates in each ocean basin, the updated database substantially increased the estimate of global oceanic N2 fixation from 137 ± 9 Tg N yr-1 using the old database to 260 ± 20 Tg N yr-1 (mean ± standard error). However, using geometric means instead, the updated database gave an estimate of global oceanic N2 fixation (60 Tg N yr-1) similar to that estimated from the old database (62 Tg N yr-1), while the new estimate had a larger uncertainty (confidence intervals based on one standard error: 47 – 107 Tg N yr-1 versus 52 – 73 Tg N yr-1), mostly attributable to elevated uncertainties in the Pacific Ocean. An analysis comparing N2 fixation rates measured at the same months and location (1° × 1° grids) showed that the new 15N2 dissolution method obtained N2 fixation rates higher than the conventional 15N2 bubble method in 65 % of cases, with this percentage increasing when the N2 fixation rates were high (> approximately 3 μmol N m-3 d-1 using the 15N2 dissolution method). With greatly increased data points, this version 2 of the global oceanic diazotrophic database can support future studies in marine ecology and biogeochemistry. The database is stored at the Figshare repository (https://doi.org/10.6084/m9.figshare.21677687) (Shao et al., 2022).
- 19 January 2023
- Published by Copernicus GmbH in EGUsphere
- Vol. 2023, 1-44
Open AccessVersion 2 of the global oceanic diazotroph database
Marine diazotrophs convert dinitrogen (N2) in seawater into bioavailable nitrogen (N), contributing approximately half of the external input of bioavailable N to the global ocean. A global oceanic diazotroph database was previously published in 2012. Here, we compiled version 2 of the database by adding 23,095 in situ measurements of marine diazotrophic abundance and N2 fixation rates published in the past decade, increasing the number of N2 fixation rates and microscopic and qPCR-based diazotrophic abundance data by 140 %, 26 % and 443 %, respectively. Although the updated database expanded spatial coverage considerably, particularly in the Indian Ocean, the data distribution was still not uniform and most data were sampled in the surface Pacific and Atlantic Oceans. By summing the arithmetic means of the N2 fixation rates in each ocean basin, the updated database substantially increased the estimate of global oceanic N2 fixation from 137 ± 9 Tg N yr-1 using the old database to 260 ± 20 Tg N yr-1 (mean ± standard error). However, using geometric means instead, the updated database gave an estimate of global oceanic N2 fixation (60 Tg N yr-1) similar to that estimated from the old database (62 Tg N yr-1), while the new estimate had a larger uncertainty (confidence intervals based on one standard error: 47 – 107 Tg N yr-1 versus 52 – 73 Tg N yr-1), mostly attributable to elevated uncertainties in the Pacific Ocean. An analysis comparing N2 fixation rates measured at the same months and location (1° × 1° grids) showed that the new 15N2 dissolution method obtained N2 fixation rates higher than the conventional 15N2 bubble method in 65 % of cases, with this percentage increasing when the N2 fixation rates were high (> approximately 3 μmol N m-3 d-1 using the 15N2 dissolution method). With greatly increased data points, this version 2 of the global oceanic diazotrophic database can support future studies in marine ecology and biogeochemistry. The database is stored at the Figshare repository (https://doi.org/10.6084/m9.figshare.21677687) (Shao et al., 2022).
- Open Access
- 19 January 2023
- Published by Copernicus GmbH
Supplementary material to "Version 2 of the global oceanic diazotroph database"
- 19 January 2023
- Published by Copernicus GmbH
Open AccessSupplementary material to "Version 2 of the global oceanic diazotroph database"
- Open Access
- 16 June 2022
- Published by Copernicus GmbH in Biogeosciences
- Vol. 19 (11) , 2939-2952
Controlling factors on the global distribution of a representative marine non-cyanobacterial diazotroph phylotype (Gamma A)
Non-cyanobacterial diazotrophs may be contributors to global marine N2 fixation, although the factors controlling their distribution are unclear. Here, we explored what controls the distribution of the most sampled non-cyanobacterial diazotroph phylotype, Gamma A, in the global ocean. First, we represented Gamma A abundance by its nifH quantitative polymerase chain reaction (qPCR) copies reported in the literature and analyzed its relationship to climatological biological and environmental conditions. There was a positive correlation between the Gamma A abundance and local net primary production (NPP), and the maximal observed Gamma A abundance increased with NPP and became saturated when NPP reached ∼ 400 mg C m−2 d−1. Additionally, an analysis using a multivariate generalized additive model (GAM) revealed that the Gamma A abundance increased with light intensity but decreased with increasing iron concentration. The GAM also showed a weak but significant positive relationship between Gamma A abundance and silicate concentration, as well as a substantial elevation of Gamma A abundance when the nitrate concentration was very high (≳ 10 µM). Using the GAM, these climatological factors together explained 43 % of the variance in the Gamma A abundance. Second, in addition to the climatological background, we found that Gamma A abundance was elevated in mesoscale cyclonic eddies in high-productivity (climatological NPP > 400 mg m−2 d−1) regions, implying that Gamma A can respond to mesoscale features and benefit from nutrient inputs. Overall, our results suggest that Gamma A tends to inhabit ocean environments with high productivity and low iron concentrations and therefore provide insight into the niche differentiation of Gamma A from cyanobacterial diazotrophs, which are generally most active in oligotrophic ocean regions and need a sufficient iron supply, although both groups prefer well-lit surface waters. More sampling on Gamma A and other non-cyanobacterial diazotroph phylotypes is needed to reveal the controlling mechanisms of heterotrophic N2 fixation in the ocean.
- 16 June 2022
- Published by Copernicus GmbH in Biogeosciences
- Vol. 19 (11) , 2939-2952
Open AccessControlling factors on the global distribution of a representative marine non-cyanobacterial diazotroph phylotype (Gamma A)
Non-cyanobacterial diazotrophs may be contributors to global marine N2 fixation, although the factors controlling their distribution are unclear. Here, we explored what controls the distribution of the most sampled non-cyanobacterial diazotroph phylotype, Gamma A, in the global ocean. First, we represented Gamma A abundance by its nifH quantitative polymerase chain reaction (qPCR) copies reported in the literature and analyzed its relationship to climatological biological and environmental conditions. There was a positive correlation between the Gamma A abundance and local net primary production (NPP), and the maximal observed Gamma A abundance increased with NPP and became saturated when NPP reached ∼ 400 mg C m−2 d−1. Additionally, an analysis using a multivariate generalized additive model (GAM) revealed that the Gamma A abundance increased with light intensity but decreased with increasing iron concentration. The GAM also showed a weak but significant positive relationship between Gamma A abundance and silicate concentration, as well as a substantial elevation of Gamma A abundance when the nitrate concentration was very high (≳ 10 µM). Using the GAM, these climatological factors together explained 43 % of the variance in the Gamma A abundance. Second, in addition to the climatological background, we found that Gamma A abundance was elevated in mesoscale cyclonic eddies in high-productivity (climatological NPP > 400 mg m−2 d−1) regions, implying that Gamma A can respond to mesoscale features and benefit from nutrient inputs. Overall, our results suggest that Gamma A tends to inhabit ocean environments with high productivity and low iron concentrations and therefore provide insight into the niche differentiation of Gamma A from cyanobacterial diazotrophs, which are generally most active in oligotrophic ocean regions and need a sufficient iron supply, although both groups prefer well-lit surface waters. More sampling on Gamma A and other non-cyanobacterial diazotroph phylotypes is needed to reveal the controlling mechanisms of heterotrophic N2 fixation in the ocean.
- Open Access
- 7 January 2022
- Published by Copernicus GmbH in EGUsphere
- Vol. 2022, 1-30
Controlling factors on the global distribution of a representative marine heterotrophic diazotroph phylotype (Gamma A)
Heterotrophic diazotrophs emerge as a potentially important contributor to the global marine N2 fixation, while the factors controlling their distribution are unclear. Here, we explored what controls the distribution of the most sampled heterotrophic diazotroph phylotype, Gamma A, in the global ocean. First, we analyzed the relationship between nifH-based Gamma A abundance and climatological biological and environmental conditions. The carrying capacity of Gamma A abundance increased with net primary production (NPP) and saturated when NPP reached ~400 mg C m−2 d−1. The reduction in Gamma A abundance from its carrying capacity was mostly related to low temperature, which possibly slowed the decomposition of organic matter, and high concentration of dissolved iron, to which the explanation was elusive but could result from the competition with autotrophic diazotrophs. Using a generalized additive model, these climatological factors together explained 41 % of the variance in the Gamma A abundance. Second, in additional to the climatological background, we found that mesoscale cyclonic eddies can substantially elevate Gamma A abundance, implying that Gamma A can respond to short-term features and benefit from stimulated primary production by nutrient inputs. Overall, our results suggest that the distribution of Gamma A is most likely determined by the supply of organic matters, not by those factors controlling autotrophic diazotrophs, and therefore insight a niche differentiation between the heterotrophic and autotrophic N2 fixation. More samplings on Gamma A and other heterotrophic diazotroph phylotypes are needed to better reveal the controlling mechanisms of heterotrophic N2 fixation in the ocean.
- 7 January 2022
- Published by Copernicus GmbH in EGUsphere
- Vol. 2022, 1-30
Open AccessControlling factors on the global distribution of a representative marine heterotrophic diazotroph phylotype (Gamma A)
Heterotrophic diazotrophs emerge as a potentially important contributor to the global marine N2 fixation, while the factors controlling their distribution are unclear. Here, we explored what controls the distribution of the most sampled heterotrophic diazotroph phylotype, Gamma A, in the global ocean. First, we analyzed the relationship between nifH-based Gamma A abundance and climatological biological and environmental conditions. The carrying capacity of Gamma A abundance increased with net primary production (NPP) and saturated when NPP reached ~400 mg C m−2 d−1. The reduction in Gamma A abundance from its carrying capacity was mostly related to low temperature, which possibly slowed the decomposition of organic matter, and high concentration of dissolved iron, to which the explanation was elusive but could result from the competition with autotrophic diazotrophs. Using a generalized additive model, these climatological factors together explained 41 % of the variance in the Gamma A abundance. Second, in additional to the climatological background, we found that mesoscale cyclonic eddies can substantially elevate Gamma A abundance, implying that Gamma A can respond to short-term features and benefit from stimulated primary production by nutrient inputs. Overall, our results suggest that the distribution of Gamma A is most likely determined by the supply of organic matters, not by those factors controlling autotrophic diazotrophs, and therefore insight a niche differentiation between the heterotrophic and autotrophic N2 fixation. More samplings on Gamma A and other heterotrophic diazotroph phylotypes are needed to better reveal the controlling mechanisms of heterotrophic N2 fixation in the ocean.