| 19 May 2025
Constraining the time of emergence of anthropogenic signal in the global land carbon sink
Abstract. The global land carbon sink has increased since the preindustrial period, driven by increasing atmospheric CO2 concentration and climate change. However, detecting these anthropogenic signals in the global land carbon sink is challenging due to the large year-to-year variability, which can mask or amplify long-term trends, particularly on regional and decadal scales. This study aims to detect the time it takes for long-term trends driven mostly by anthropogenic signal to dominate over natural variations, that is, the "time of emergence", in the land carbon sink.
For this, we use five large ensembles of historical simulations (1851–2014) and future scenarios (2016–2100) by Earth system models. Our results show that, firstly, the anthropogenic signal in the global net land carbon sink emerges from 26 to 66 years in the period 1960–2019 (relative to the natural variations in the period of 1930–1959), depending on the ESM considered. The time of emergence is considerably shorter for the two major gross carbon fluxes: 8–13 years for gross primary productivity and 6–10 years for total ecosystem respiration. Furthermore, we find that long-term trends of net land carbon sink on most regional scales take at least 20 years more to emerge, due to larger contributions from internal climate variability at smaller scales.
Secondly, future scenarios show delayed signal detection compared to historical trends, due to a slow-down of the increasing net land carbon sink in response to emission mitigation, compared to the higher natural variability.
Thirdly, we apply dynamical adjustment to filter out the year-to-year circulation-induced variability in both the historical and future simulations. This approach allows to substantially shorten the detection time for the global net land carbon sink: between 34–39 % for the historical period and 27–54 % for the future simulations. This approach can, in principle, be applied to observational based datasets, thereby improving our ability to detect long-term trends on land carbon sink variability. Given that long-term trends are mostly associated with human impacts on the land carbon cycle, our proposed approach can offer valuable insights on the effectiveness of policy decisions and their implementation.
Review of “Constraining the time of emergence of anthropogenic signal in the global land carbon sink” by Na Li et al
This paper considers the time of emergence of 3 key variables in the global land carbon sink and shows how dynamical adjustment can be used to shorten the detection time. While the paper is interesting and provides a important study in the literature I have a few major concerns around the methodology and lack of observational evidence. As such, I recommend major revisions.
Major comments
1. The most major issue with the paper is the methods, which use linear trends to calculate ToE. This is not the typical method found in other studies such as: https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019GB006453 for large ensembles and https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2011gl050087 for the standard method.
It clear from the Figure timeseries that a linear fit is not appropriate for this data.
Additionally presenting ToE of 800 years, when the timeseries is only 100-years is confusing and extrapolation. If the signal has not emerged by the end of the timeseries I suggest presenting this as not emerged rather than extrapolating.
2. This method is used typically to understand when we might expect to observe changes outside the noise. However, the paper has no observational evidence in it. I suggest comparing to any observations we have.
3. The % reductions from the dynamical adjustment might be good to include as years as well as % as this is easier to understand. For the dynamical adjustment it is unclear whether this method could be applied to observations to reduce the TOE or just models. I suggest adding this information in the text.
4. I am confused abut use of 2020-2050 for N and 2020-2070 for S what make this choice? This seems very arbitrary
5. Figures
Figure 1: You could easily include all NBP, GPP and TER on one figure – I would recommend this rather than push some to the supplementary. I also suggest adding a multi-ensemble mean panel. The ensemble sizes seem rather arbitrary. For example ACCESS has 40 members and CESM2 100 members – so why are they only 38 and 90? Additionally CESM2 biomass burning runs have been shown to be different to the cmip6 runs have you checked these differences?
Figure 2:
Suggest adding a multi-ensemble mean to each panel – this suggestion is for all Figures. For this figure is it showing TOE = years post 1960? can this be clear in the caption please. Additionally what does ‘the spatial domains are slightly changed’ mean?
Figure 3 – Is this really resolution? i.e. is it the same area regridded to a different resolution or is it a different spatial domain?
Figure 5: there seem to be missing bars where only the light or shaded box is there? What happened to the other box?
A lot of content discussed is in the appendix figures – I wonder if more should be in the main text
6.
Paper is missing a discussion section
7. Section 3.1 – the noise could be easily quantified rather than discussed qualitatively
Minor comments
Title should be ‘the anthropogenic signal’
Intro
Line 1 ‘the’ atmospheric CO2
Line 2 – this is not driven by climate change, more likely it is driven by processes related to climate change please be specific here
line 6 replace by with from
Line 13 – it is odd to have a 1 sentence paragraph – this sentence is also unclear can you please rephrase
Line 16 – remove ‘allows to’ replace shorten with ‘shortens’
Line 18 – replace on with ‘of’
Line 30 – should be ‘temperature, and precipitation’
Line 33 – replace by with ‘on’
Line 33 – the paragraph starting on this line is hard to parse, I suggest rewriting
Line 47 – can differ substantially from what? Please be specific
Line 50 – a good example of this is found in this paper: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-3684/
Line 53 – I would end this sentence with ‘orividing a challenge because …’
Line 65 – this section is missing two key references: https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019GB006453, https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2011gl050087
Line 73 – odd use of the word ‘need’
Methods
Line 81 – in this section please add the number of ensemble members from each model
Line 82 – CESM2 is part of CMIP6
Line 84 – this is the correct citation for MPI-ESM1-2-LR large ensemble: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023MS003790
Line 125 – this is for CESM2 what about the other models?
Line 133 – what about the SSPs
Line 134 – how do you calculate N do you pool the data? Do you take the STD of each member and average? Please provide this detail
Line 137 – please add what 2N/S means in terms of significance of the emergence
Line 147 – would 500mb geopotential height be a better proxy for circulation than SLP?
Line 168 – you interchange ‘ToE’ and ‘time to detect’ this is confusing please stick to one terminology
Results:
Line 213 – similar patters to what? Each other?
Line 230 – grammer please fix ‘takes shorter’
Line 232 – please like with ‘such as’
Line 233/234/236 – please reference the figures that you are discussing in this section
Line 233 – remove ‘is apparently’
Line 240 – is this statement true – I do not make the same conclusion from the Figure
Conclusions:
Line 293 – you should be able to quantify if this is due to larger noise or not from your results not just speculate
Line 315 – can be detected how? There are no observations in the paper