the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 May 2026
A Comprehensive Global Aquatic N2O Emission Database (GANED): Unravelling N2O Emission Patterns from Different Water Bodies
Abstract. Nitrous oxide (N2O) is not only one of the main potent greenhouse gases, but also currently the dominant ozone-depleting substance. The quantification of N2O emissions from aquatic ecosystems, despite their global importance, is hindered by fragmented observations, inconsistent data reporting, and pronounced spatiotemporal variability. In this study, to improve accessibility, we introduce the Global Aquatic N2O Emission Database (GANED; https://doi.org/10.6073/pasta/4a086e49a4f308679b951293b380e7b9, Nazir et al., 2026), a consolidated dataset comprising 5130 N2O concentration records and 7386 N2O flux measurements from 3002 sites across diverse aquatic systems, including rivers, streams, lakes, reservoirs, ponds, estuaries, coastal waters, and open seas. The dataset integrates information on aquatic N2O emission from 426 peer-reviewed publications across 8 continents, covering the period 1980–2023. While the number of observations has increased substantially since 2000, spatial coverage remains uneven, with significant gaps across Africa and parts of high-latitude regions, including Antarctica and South America. Our dataset revealed a highly skewed distribution of N2O concentration and flux across aquatic ecosystems, with rivers and streams exhibiting the most significant variability and functioning as emission hotspots. Lakes and estuaries showed moderate variability and emission levels, whereas seas and coastal waters were characterized by consistently lower values. Pearson correlation coefficient revealed a strong positive relationship of N2O fluxes with ammonium (NH4+; R = 0.943, p < 0.001), nitrate (NO3-; R = 0.691, p < 0.001), and nitrite (NO2-; R = 0.807, p < 0.001). Significant negative correlations were found with dissolved oxygen (DO; R = -0.205, p < 0.05), dissolved organic carbon (DOC; R = -0.977, p < 0.05), and salinity (R = -0.636, p = 0.005), while non-significant associations were observed for water temperature, total nitrogen (TN), and total phosphorus (TP). The GANED dataset facilitates improved quantification of global aquatic N2O inventories by providing comprehensive N2O concentrations and fluxes in water bodies, as well as metadata describing sampling location, aquatic system type, and associated environmental parameters. The magnitude and patterns of N2O emissions from water bodies provided by the GANED database are essential in defining how these aquatic ecosystems shape our climate, refining emission estimates, identifying drivers, and guiding mitigation strategies.
- Preprint
(2541 KB) - Metadata XML
-
Supplement
(529 KB) - BibTeX
- EndNote
Status: open (until 21 Jul 2026)
- RC1: 'Comment on essd-2026-109', Anonymous Referee #1, 07 Jun 2026 reply
-
CC1: 'Comment on essd-2026-109', Damian Leonardo Arévalo-Martínez, 09 Jun 2026
reply
-
RC2: 'Comment on essd-2026-109', Anonymous Referee #2, 11 Jun 2026
reply
The comment was uploaded in the form of a supplement: https://essd.copernicus.org/preprints/essd-2026-109/essd-2026-109-RC2-supplement.pdf
-
RC3: 'Comment on essd-2026-109', Anonymous Referee #3, 14 Jul 2026
reply
The study titled “A Comprehensive Global Aquatic N2O Emission Database (GANED): Unravelling N2O Emission Patterns from Different Water Bodies” presents a global database of N₂O fluxes and concentrations in aquatic systems compiled from published observational data. The dataset spans a wide range of aquatic environments, including rivers, streams, ponds, lakes, reservoirs, estuaries, and oceans. Developing a standardized global database for aquatic N₂O can help improve global N₂O budget estimates and reduce uncertainties in its contribution to atmospheric radiative forcing.
While the authors have clearly invested substantial effort in assembling this database—including extensive literature compilation and communication with original data providers for data verification—there remain several significant concerns that need to be addressed:
- In the introduction, the authors provide a thorough discussion of limitations in previous studies (lines 76–93 and 284–285), including methodological inconsistencies in flux estimation and substantial uncertainties associated with sparse spatial and temporal sampling. However, the present dataset still fundamentally relies on previously published observational data. In this context, it is not entirely clear what the key advancement of this work is relative to existing global synthesis datasets. It is important to more clearly articulate the novelty and added value of this contribution, given that several recent studies have already compiled aquatic N₂O concentration and flux databases.
- The analysis of temporal trends in N₂O concentrations and fluxes across aquatic systems appears to be strongly confounded by spatial heterogeneity, as sampling locations vary substantially across years. Consequently, the reported trends likely reflect a mixture of spatial and temporal variability rather than true temporal changes. It would be helpful to explicitly account for spatial heterogeneity prior to conducting the temporal trend analyses. In addition, within this spatiotemporally mixed framework, several systems (e.g., rivers, streams, and lakes) show opposing trends between concentrations and fluxes. This inconsistency warrants further explanation.
- In the correlation analysis between N₂O fluxes and environmental variables, one particularly surprising result is the negative relationship between water temperature and N₂O fluxes (line 461). This is counterintuitive, as higher temperatures are generally expected to enhance microbial activity and thus stimulate N₂O production and emissions. Moreover, while several nitrogen species show significant positive correlations with N₂O fluxes, total nitrogen (TN) does not show a significant relationship. It is recommended that the authors further elaborate on the mechanistic or statistical explanation for these patterns.
- What is the purpose of using machine learning to predict aquatic N₂O fluxes? Model performance across aquatic systems is generally weak, with a maximum R² of only 0.24. Under these circumstances, it is difficult to see how the machine learning results substantively contribute to the main conclusions of the study. In addition, the manuscript does not clearly explain why two separate approaches were used to examine relationships between N₂O fluxes and environmental drivers. The authors are encouraged to provide a clearer justification for the need for this dual analytical framework.
- The manuscript would benefit from a more careful approach to citation. For instance, a study on riverine N₂O emissions is cited in the introduction as support for a bottom-up global N₂O emission estimate, which seems insufficient to justify such a broad generalization. Later, Tian et al. (2024) is cited as key evidence for top-down global N₂O estimates. However, as far as I am aware, that study includes both top-down and bottom-up components; it would therefore be more appropriate to cite it in a consistent and comprehensive manner. In addition, the formatting of numerical values is inconsistent throughout the manuscript and should be standardized.
In addition, I have several minor comments:
L221: The manuscript states that all data have been converted to a monthly scale; however, subsequent analyses appear to retain flux units in daily form (e.g., mmol m⁻² d⁻¹ in Figure 4). This discrepancy should be clarified.
L233: Did the authors use this equation to estimate fluxes for missing observations? If so, it should be noted that the Introduction already highlights substantial uncertainties associated with estimating fluxes from concentrations compared with direct measurements. Could the authors further justify the use of this approach?
Figure 4: The bar plot indicates that 554 N₂O flux records were available in 2020, yet the corresponding flux values in the figure appear to be zero. In addition, a data point in the offshore region near Africa appears to be misclassified as a river sampling point. The authors may wish to verify the accuracy of the spatial coordinates and data classification.
Citation: https://doi.org/10.5194/essd-2026-109-RC3
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 353 | 142 | 26 | 521 | 40 | 19 | 18 |
- HTML: 353
- PDF: 142
- XML: 26
- Total: 521
- Supplement: 40
- BibTeX: 19
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
See attached file