the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Oct 2025
A synthesis of water, energy, and carbon fluxes sensitivity to climate variables in Southeast Asia
Abstract. Southeast Asia (SEA) plays an important role in the Earth’s carbon and water cycle, yet ecohydrology dynamics occurring in this region remain poorly understood due to the paucity of field observations and modelling studies. Here, we investigate water, energy, and carbon fluxes by combining existing flux tower data with mechanistic ecohydrological modelling for 20 sites. A sensitivity analysis to meteorological forcings is used to understand water and energy limitations. Results show large latitudinal differences but overall suggest a strongly energy-limited region, where evapotranspiration (ET) is tightly correlated with net radiation and is highly responsive to relative humidity. Gross primary productivity (GPP) is also correlated to net radiation and is most responsive to shortwave radiation changes. Only a few ecosystems in SEA show signs of water limitations, such as certain grasslands in the Tibetan plateau, savannas, and dry deciduous forests. We further disentangled the relative effect of warming and humidity changes in vapor pressure deficit (VPD). Sensitivity analysis indicates that climate warming-induced VPD changes – rather than pure warming – can have important effects on ET but the opposite is true for GPP with complex GPP responses to temperature based on the thermal photosynthetic optimum and phenological responses. Water use efficiency (WUE) is highly correlated with annual mean precipitation across space, but its responses to precipitation changes are less consistent and WUE changes are most sensitive to relative humidity. Carbon use efficiency (CUE) is more responsive to air temperature than other climate drivers. These insights quantify water, energy, and carbon fluxes in an underrepresented part of the Earth and enhance our understanding of how climate can modify carbon and water cycles in this region.
- Preprint
(3781 KB) - Metadata XML
-
Supplement
(1255 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-4570', Anonymous Referee #1, 23 Oct 2025
-
CC1: 'Reply on RC1', Jianning Ren, 10 Nov 2025
Dear Reviewer,
Thank you for your constructive and thoughtful comments. We appreciate the time and effort you have invested during the review process.
Your feedback has been very helpful for improving the manuscript.
We have carefully addressed all comments and will revise the manuscript accordingly.
Please see the attached PDF file for our detailed reply.
Thank you once again for your time and valuable input.
Best wishes,
Jianning Ren and Simone Fatichi
-
CC1: 'Reply on RC1', Jianning Ren, 10 Nov 2025
-
RC2: 'Comment on egusphere-2025-4570', Anonymous Referee #2, 19 Jan 2026
This manuscript describes how climatic drivers affect evapotranspiration, carbon flux, and ecophysiological indices across Southeast Asia (SEA) by numerical experiments. The authors gathered 20 tower flux data, calibrated the mechanistic ecohydrological model, and conducted the numerical experiment by perturbing the climatic drivers. They found that most vegetation in SEA is energy-limited rather than water-limited, and that the impact of perturbed climatic drivers on fluxes or indices varied across drivers and vegetation types. As synthesizing research in the region remains limited, this research focuses on the important topic. However, it has significant problems with methodology and analysis.
- Why did the authors include China and Taiwan in SEA? Typically, SEA is defined as the group of countries (Myanmar, Thailand, Laos, Cambodia, Vietnam, Malaysia, Singapore, Indonesia, Brunei, Philippines, and East Timor), and most articles follows this definition, such as Pan et al. (2024; Nature), Estoque et al. (2019; Nat. Commun.), and Pletcher et al. (2022; Ecography). According to Table 2, the annual air temperature of the CN-Dan site is 2.2 °C. Including such vegetation sites in cold environments would prevent the disentangling of the nature of the vegetation mechanism in SEA. Removing the sites in China and Taiwan, especially in the Tibetan region, from the analysis is recommended, though this removal would drastically change the results, discussion, and conclusion.
- To conduct such a synthetic analysis, describing the method for post-processing eddy covariance flux data is vital because it affects the output flux values heavily. The authors provide no information on the flux data source or the flux post-processing protocol, including conversion from raw 10-20 Hz data to half-hourly or hourly data, quality control of the hourly data, and the carbon flux partitioning method. Lacking this kind of information will reduce the reproducibility and reliability of this research.
- The authors mentioned in the Introduction section that maritime SEA has a slight seasonal variation of precipitation, but mainland SEA has clear wet and dry seasons. This implies that vegetation in the mainland SEA can be water-limited in the dry season. Nevertheless, the authors focused solely on annual values in their numerical experiments and did not discuss the importance of seasonal variations in fluxes and ecophysiological indices. Also, the climatic drivers in the SEA region exhibit interannual variability induced by the El Niño–Southern Oscillation (ENSO). Some sites experienced ENSO during the studied period. But the authors averaged the fluxes for each DOY across all available years, thereby mixing ENSO effects into the data under neutral conditions. It is recommended to clarify the extent of the interannual variation in each variable before the numerical experiments.
Specific comments:
L148: I cannot determine the significance of creating the map, as the authors did not analyze it in the Results section. If the authors use this map only to confirm that the flux sites cover representative vegetation in the SEA, I recommend moving the 2.1 section to the supplementary materials.
L169–171: Some of the annual values mentioned here do not match the numbers in Table 2.
L176: As oil palm is one of the most popular plantations in SEA, not including an oil palm plantation site in the dataset is problematic. The AsiaFlux Database includes at least one dataset at an oil palm plantation site (JOP: Jambi Oil Palm Plantation) from 2014 to 2020. How about adding this site to the analysis?
L189–190: The authors' inclusion of the southern part of the Tibetan region is unreasonable. What relates the source area of the Mekong and Salween rivers to this research?
L195–198: Usually, the energy balance closure (EC) is calculated as (H + LE) / (Rn - G), where G is ground heat flux. If the G is available, the authors should use this definition. By the way, the typical EC value is around 80% (Mauder et al., 2024; Agricultural and Forest Meteorology). It is strange that some of the EC values in Table 2 exceed 100%. The authors need to check whether the energy balance correction is already applied to the energy fluxes at these sites.
L218: Clarify which parameters in the T&C model were calibrated.
L285–286: I am not sure that the authors made an appropriate effort to find the carbon flux time series. For example, GPP of the MY-PSO site is shown as NA in Table 3 and Fig. S3, but the GPP time series in the same period (2003–2009) at the site is available in the FLUXNET2015 dataset. The dataset can be easily downloaded.
L362: It is recommended not to draw the regression line in Figs. 5, 6 if the p-value exceeds 0.05.
L444, 518, 563: Describe the meaning of the lines and shaded areas.
L680: Include the information on the flux dataset source.Technical corrections:
L134: Use “SEA” instead of “Southeast Asia”.
L182: Use the exact name of the dataset (FLUXNET2015).
L477, 482: Add “C” after “°”.Citation: https://doi.org/10.5194/egusphere-2025-4570-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 384 | 168 | 32 | 584 | 59 | 24 | 24 |
- HTML: 384
- PDF: 168
- XML: 32
- Total: 584
- Supplement: 59
- BibTeX: 24
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript by Ren et al presents an analysis of carbon and water fluxes at eddy covariance sites across South East Asia and their meteorological drivers from data and model simulations. While the study has potential, as it stand it falls extremely flat. I understand the author’s argument that such studies have not been done in SE Asia, as it stands it does not bring anything new to the field.
Even disregarding novelty, which could be argued is not an argument for publishing a paper, the manuscript is lacking in clarity and the analysis is simplistic. Specifically:
Specific comments
Table 2 – what si the source of these data?
L 218 “identify suitable parameters” more details are needed as to which parameters and how values were found
L 219 was the site level met data the only source used as input? Was the model run with repeated meteorology? Does it require spinup?
L 226 what does correlation analysis mean in this context? Assuming this si for the moel output since there are variables that cannot be measured at flux towers. More details needed
L 235 clarify, presumably each variable has the same units across sites
L 239 is this then the standard deviation of a standard deviation?
L 253 which variables were added/multiplied and which subtracted?
L 260 6 years is not enough time to initialise C pools, worth checking if all pools are actually in equilibrium
L 285 why did the authors not just partition the NEE, there out of the box packages to do this
Fig 2 worth also reporting other measures of fit, such as RMSE. It would also be useful to have the land cover for each site indicated on the plot
Fig 7-9 what are the different lines here? This is not described in the methods. Givent eh nature of the perturbation, if this is a regression line for each site it essentially amounts to drawing a line through 3 points
L 680 there should be a link to the flux data too