the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Evaluating state-of-the-art process-based and data-driven models in simulating CO2 fluxes and their relationship with climate in western European temperate forests
Abstract. This study evaluates two process-based (LPJ-GUESS and SMAP-L4C) and two data-driven (CarbonSpace and FLUXCOM) models to capture the temporal variability of CO2 flux exchanges (GPP, RECO and NEE) of evergreen needleleaf and deciduous broadleaf forests (ENFs and DBFs) in temperate western Europe and its relationship with climate. Three sites from the FLUXNET network are considered together with two non-instrumented sites located in Burgundy (North-East France). The focus is put on the representation of the annual cycle, annual budget, interannual variability and “long-term” trend. The data-driven models are the best models for representing the mean annual cycle and mean annual budget in CO2 fluxes despite magnitude uncertainties. In particular, the models accounting for plant functional types in their outputs tend to simulate more marked annual cycle and lower annual CO2 sequestration for DBFs than ENFs in Burgundy. At the interannual timescale, the CO2 flux – climate relationshipis stronger for GPP and RECO than NEE, with increased CO2 fluxes when 2 m temperature, vapor pressure deficit and evapotranspiration increase and when precipitation and soil moisture decrease. The models forced by dynamic climate conditions clearly outperform those driven by static climate conditions. The “long-term” trend is not obvious for NEE neither in the observations nor in the simulations, partly because both GPP and RECO tend to increase in western Europe. Our results suggest that the spatial resolution of the climate drivers is likely very important for capturing spatial and temporal patterns in CO2 exchanges and point towards the need to choose the appropriate model and spatial resolution according to the scientific question to deal with.
- Preprint
(11652 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'Comment on egusphere-2024-1758', S E Quick, 21 Jun 2024
Section 1 Introduction:
Many of your references seem to be missing from this preprint or are cited incorrectly, e.g. Smith et al., 2020; Thompson et al., 2020; Yuan et al., 2009
Section 2.1 Site description:
It would be useful to know when each forest was established ( e.g. 10s,100s or 1000s of year ago.)
Citation: https://doi.org/10.5194/egusphere-2024-1758-CC1 -
AC4: 'Reply on CC1', Julien Crétat, 26 Sep 2024
Section 1 Introduction:
Many of your references seem to be missing from this preprint or are cited incorrectly, e.g. Smith et al., 2020; Thompson et al., 2020; Yuan et al., 2009
Thank you for these remarks. We will carefully check the reference list in the revised manuscript and cite the references following the journal guideline.
Section 2.1 Site description:
It would be useful to know when each forest was established ( e.g. 10s,100s or 1000s of year ago.)
We will detail, as much as possible, the age of the forests at the ICOS sites in the revised version
Citation: https://doi.org/10.5194/egusphere-2024-1758-AC4
-
AC4: 'Reply on CC1', Julien Crétat, 26 Sep 2024
-
RC1: 'Comment on egusphere-2024-1758', Anonymous Referee #1, 15 Jul 2024
The study evaluated CO2 flux outputs from models at six sites and their relationship with climate in temperate forests of Western Europe. Overall, this manuscript addresses an important topic and presents some interesting findings. However, I am not entirely convinced that the results significantly advance our understanding of CO2 fluxes at forest sites across different temporal resolutions (such as monthly and annual, as analyzed in this study).
General comments
- The temporal variability in this study is not adequately addressed and may need reorganization. For example, there are ten figures illustrating monthly timescale results. However, these findings are neither included nor highlighted in the abstract or conclusion. Although there are similar patterns between the monthly and annual results, as shown in Figures 6 and 8, the monthly findings should be incorporated. Alternatively, I suggest removing or reducing the figures or text about the monthly results and focusing on annual and interannual scales.
- While I am not an expert in statistics, the R and R² results in this study (such as those in Figures 6 and 7) seem very close. Is it necessary to present Figure 7 in the main text? Regarding the long-term evolution, which method was used? Why do the authors state that it 'does not depict any trend'? Please provide this information in the Methods section.
- The introduction of this study is not well-articulated. The discussion of the knowledge gap lacks a solid basis. Why is a finer spatial scale important? Why did the authors focus on monthly and annual timescales? The second objective, which involves the climate relationship, lacks motivation regarding the choice of variables in this study. Why were radiation or PPFD not considered?
- It is not appropriate to assess flux trends using the FLUXCOM dataset, as FLUXCOM does not account for CO2 fertilization effects. Please refer to the following paper: Jung, M., Schwalm, C., Migliavacca, M., Walther, S., Camps-Valls, G., Koirala, S., ... & Reichstein, M. (2020). Scaling carbon fluxes from eddy covariance sites to globe: synthesis and evaluation of the FLUXCOM approach. Biogeosciences, 17(5), 1343-1365.
Line-by-line comments:
Line 42: “..the best models…” sounds inaccurate. Perhaps "better models" would be more appropriate, as the comparison is only made with process-based models.
LL 45-48: could you please explain why?
LL 45-47: The sentence contains a lot of information and lacks clarity. I suggest revising it for better readability.
Line 49: How long is the ‘long-term’?
LL 50-53: The statement is too general and does not seem related to the main point of the study.
Line 61: ‘At the global scale, forest ecosystems cover about 30% of landmasses’ suggest adding a reference.
LL 71-72: What is the meaning of this number here?
Line 101: Suggest deleting ‘when available’
LL 101-113: Suggest moving these sentences to another paragraph or placing them above the objectives.
LL127-128: Redundace, and which varaible? all fluxes?
Line 132: "Extreme" can refer to both high and low conditions. To specify, consider using "extreme high" or "extreme hot" for clarity.
LL 132-133: Suggest revising the sentence.
LL147-149: Suggest deleting these sentences.
Figure 1: If the reader is not familiar with Europe, it may not be clear. Consider adding specific locations, such as "France".
Table 1: The table format needs adjustment. Please refer to the journal's guidelines, which typically suggest using horizontal lines only above and below the table, and as a separator between the table header and the main body.
Figure 9-11: Perhaps move to supplement.
LL555-556: Suggest deleting the sentence.
Line 603: “qualitatively similar” might sound speculative?
Line 724: Replace ‘best’ to ‘better’.
Line 772: Should be ‘Fig. 7’ not ‘Fig, 5’
Line 776: ‘Fig. 7’
Citation: https://doi.org/10.5194/egusphere-2024-1758-RC1 - AC1: 'Reply on RC1', Julien Crétat, 26 Sep 2024
-
RC2: 'Comment on egusphere-2024-1758', Anonymous Referee #2, 05 Sep 2024
This study evaluates the performance of various carbon flux products against eddy covariance measurements at three forest sites in France. The authors investigate the monthly, seasonal, and inter-annual variability of NEE, GPP, and RECO to assess different global products and explore their relationships with meteorological variables. While the manuscript is generally written clearly, the analysis lacks sufficient depth and significance for the scientific community. This isn't to say that evaluating existing products isn't valuable, but the limited number of eddy covariance sites and the selection of only four global products raise concerns about the comprehensiveness of the study.
1.Choice of Models (LPJ-GUESS and FLUXCOM v1): Why were LPJ-GUESS and FLUXCOM v1 selected when many other land surface models or upscaled products are available for evaluation? Including more models and products could improve relevance, especially since ensemble models are commonly used in land carbon sink studies.
2.Spatial Resolution Mismatch: The 50 km spatial resolution of LPJ-GUESS and FLUXCOM products may not align with the footprint of the eddy covariance sites, and this mismatch is not addressed in the manuscript.
3.Insufficient Number of Sites: Only three forest sites are used in the analysis, despite the availability of hundreds of FLUXNET sites globally. Using only these sites may not provide a robust basis for summarizing product performance.
4.Correlation Analysis: The correlation analysis in Figure 6 lacks a logical basis, as some variables (e.g., VPD and RECO) do not have clear biogeochemical or biophysical relationships. Also, the analysis does not account for multicollinearity among variables, which affects the validity of the results.
5.Focus on Temperature: The authors only consider temperature when analyzing carbon fluxes in Figure 6 and do not include other important variables, such as soil moisture, which were emphasized in the introduction. Given this, the use of polynomial regressions without considering other factors raises questions about interactive effects of multivariate factors of the carbon fluxes.
Specific Points:
Inconsistent Visualization (Figures 5 and 13): Figures 5 and 13 present similar data for annual and monthly scales, but the visualizations need to be consistent to enable direct comparison.
Climate Anomalies Definition: The authors should clarify how climate anomalies are defined, as the methods section only explains CO2 flux anomalies. Also, the choice of the -0.5/+0.5 thresholds for carbon flux anomalies seems arbitrary and needs further justification.
Citation: https://doi.org/10.5194/egusphere-2024-1758-RC2 - AC2: 'Reply on RC2', Julien Crétat, 26 Sep 2024
- AC3: 'Reply on RC2', Julien Crétat, 26 Sep 2024
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
205 | 58 | 114 | 377 | 14 | 14 |
- HTML: 205
- PDF: 58
- XML: 114
- Total: 377
- BibTeX: 14
- EndNote: 14
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1