Originality and significance

The manuscript titled “Sensitivity of tropical woodland savannas to El Niño droughts” provides an ecosystem-level analysis of the effects of the strongest El Niño event on record on the biomass productivity of two vegetation types (Cerrado and Cerradão) in the Brazilian Cerrado. The study synthesizes approximately eight years of productivity data, some of which were collected at an intra-annual frequency. This research is valuable as it expands our understanding of a critically important and endangered biome in South America.

General comments

Overall, the manuscript is well-written, though certain sections could benefit from further clarification, and organizing content into subsections may help enhance readability. For example, the discussion is currently presented as a single section that addresses multiple aspects of ecosystem productivity: stems, canopy, fine root production, biomass allocation, and ecosystem-level NPP. The authors might consider restructuring these sections to better highlight their results. This suggestion also applies to the results section. Additionally, some passages could be refined to improve clarity, and certain sentences might benefit from adjustments in English style and grammar to enhance overall readability. Some of my general comments include:

• Introduction & main hypothesis: What are the expected differences in how these plant communities respond to ENSO events, and what are the underlying reasons for these differences? The introduction currently lacks sufficient information to propose a well-developed hypothesis in this direction. It mainly describes basic characteristics of the plant communities without clearly linking this information to the research objectives, making it difficult for the reader to grasp the relevance of these details. A critical missing element is a discussion on biomass allocation patterns in response to drought and their significance for productivity in the Cerrado, which is essential given that much of the discussion revolves around these patterns. Additionally, a more nuanced description of physiological differences beyond those briefly mentioned in lines 85-96 could better illustrate how these differences influence sensitivity to climate change, providing stronger context for the study’s hypotheses. Addressing these gaps would require a comprehensive revision and restructuring of the introduction, including clearer research questions. Breaking down the introduction, results, and discussion into distinct sections could also improve the flow and clarity of the manuscript.
• NPP estimated metrics: It is unclear why the authors estimate the contribution of herbivory to NPP and coarse root production, as these estimates are not used in any analysis or discussion. Including these estimates adds confusion and does not contribute significantly to the overall narrative, especially since these fluxes were not directly measured. Inferring their sensitivity to ENSO based on estimations with prescribed uncertainty could be highly misleading and does not add anything to the manuscript.
• Cerrado 2018 productivity data: The 2018 productivity data for the Cerrado is not clearly explained, and the authors only briefly touch upon the reasons for the observed decline. Why was productivity so much lower in 2018? The authors suggest a lack of accuracy in the diameter readings (lines 367-370), raising concerns about the reliability of the data. This also prompts the question: was the data prior to 2018 accurate, or could similar measurement errors have led to overestimations and potentially flawed conclusions?
• Statistical analysis: This section lacks critical information. The statistical models used for each analysis are not clearly described, leaving the reader uncertain about which model was applied to address specific research questions. This is particularly important because the data originate from only two plots (n=1 per forest type), raising questions about the sample size and analysis approach. Did the authors use subplots, litterfall traps, or individual trees as sampling units instead? If so, how was spatial autocorrelation within subplots accounted for? None of these methodological details are explained in the statistical analysis section, making it difficult to assess the robustness of the results.
• Figures: The figures need several enhancements to improve clarity. For example, pairwise comparisons are not indicated, making it difficult for readers to discern if values are significantly different. Some figures, like Figure 1, have font sizes that are too small, and others, like Figure 6, are missing error bars. In Figure 2, letters appear within the panels (upper left corner) that are not explained in the figure legend, leading to confusion. Although the manuscript specifies that the color scheme was chosen to be accessible to color-blind readers (lines ), many bar plots still use similar shades of green and include a dashed red line, which may not be effective. A simpler color scheme, such as using white and dark gray/black, might improve readability.

Line-by-line comments below:

Lines 25-26: Perhaps you could also highlight whether these differences persisted during ENSO.

Line 42: I think the neotropics in general experience the strongest drought on record. For instance tropical dry forest in Central America suffer exacerbated mortality rates, to the point that up 30% of all individuals of a given species died (Powers et al., 2020).

Line 66: This idea repeats one item (wildfires) in the list above. Maybe delete from the list and just state it here.

Line 89: Perhaps you can use the word “indicates” instead of “means”.

Lines 89-90: Is this a hypothesis you are proposing? One could argue that the earlier the leaf loss is due to more hydraulic vulnerability in these species (Reich & Borchert, 1984; Eamus, 1999; Brodribb et al., 2002; Sobrado, 2015; Vargas G. et al., 2021).

Line 93: tree height and plant vulnerability to drought. Maybe you would like to check the work by Olson and collaborators (2018, 2020).

Line 97: Maybe state the main hypothesis of the study here. Also, what are the broader implications of studying these responses. Can they inform how we predict the fate of this ecosystem in a warmer/drier world.

Line 108: What is the size of the plots, how many subplots are there? Something missing in the site description is the water table proximity, which is really important in this region of the world (Mattos et al., 2023).

Lines 157-159: Are you referring here to the hydrological year? Or is just the period being there was ENSO? I think using the hydrological year is the most appropriate (Aragão et al., 2007; Feng et al., 2013; Schwartz et al., 2020).

Table 1: the writing in some of the descriptions is confusing (e.g., LAI).

Table 1: Loss to leaf herbivory and coarse root net primary productivity. Is the 1.37 the average of "shrubland" in Miranda et al. 2014? Since this was not measured and the systematic uncertainty seems arbitrary and do not follow the CV presented by Miranda et al. 2014, I would restrain of using this.

Lines 182-198: I would focus the estimations on the measurements you have data for. The problem of adding estimations with systematic errors, is that these can be highly inaccurate and can lead to misleading results. While I appreciate the intend to be as precise as possible in estimating the NPP. The focus of the study is to quantify the effects of ENSO on the measured fluxes. Adding all this information on unmeasured quantities is just confusing and does not seem to add much to the study.

Line 188: This is just the production of fine roots, because you are not considering the production of coarse roots.

Lines 207-210: Maybe is worth explaining here the decision making process to define these % of error propagation.

Lines 215-216: Please describe the models more. Main effects? nested effects? Interactions? Also, how did you deal with a sample size of 1 per forest type?

Lines 220-221: This is unnecessary, and shades of green with red dashed lines is not a color-blind friendly combination of colors.

Line 229: Can we say then that ENSO did not affect the Cerrado?

Line 240: What are these little letters in the upper left corners, what do they represent?

Lines 250-253: This is not correct. First, 3 b) shows dbh growth, not NPP of alive trees. Second, npp should consider the biomass lost to mortality (Anderson‐Teixeira et al., 2016).

Line 260: Can you indicate whether you observed differences in the plot?

Line 276: Again, indicate with letters pairwise comparisons.

Line 291: Indicate significance here.

Lines 310-312: This is not clear from the figures.

Line 318: Add the color scale to the plot as a legend, the axes should be black and same with the gridlines.

Line 325: This is not clear from the figures. Resistant, yes when comparing absolute values. Did you try looking at the relative change in productivity? Resilient, is impossible to compare when the 2018 data seem wrong.

Lines 335-338: This is unnecessary, and just fill space.

Lines 343-344: dominant species or most important species might be the most appropriate term.

Line 345: plants can adjust this... See Guo et al (2020) (https://doi.org/10.1111/nph.16196) and Guo et al. (2024) (https://doi.org/10.1111/nph.19805). Perhaps is not a matter of adjusting their stomatal control, but more about their hydraulic safety margin? Nowhere in the discussion or introduction hydraulic safety margins are mentioned. Also, in Jankoski et al. 2022, Tachigali vulgaris seems to be partially isohydric (Fig. 3), which suggests some degree of stomatal regulation.

Lines 349-351: How does the amount of embolisms link to wind resistance? In your previous article (Reis et al. 2022), you assessed mortality but did not gather enough information to suggest that drough+heat decreased the capacity of stems to resist winds. Please explain?

Lines 354-355: higher productivity when?

Lines 362-364: This is not clear from that reference....

Lines 367-370: plants were competing also before ENSO. This statement does not make sense. Was there a fire event after or during ENSO???? The one problem is the magnitude of the inaccuracy, that is concerning.

Lines 377-378: This could also be root phenological patterns associated to water availability (Kummerow et al., 1990; Kavanagh & Kellman, 1992).

Lines 388-390: Yes, but this was not measured. Something the authors have not developed is whether plants are deciduous or evergreen or brevideciduous and how the underlying physiological differences link to observed responses. There are only a couple of lines about that, but that might have implications in the different responses of the two plant communities. For example:

• It might be important in the regulation of water transport and photosynthesis (Brodribb et al., 2002).
• It might also be related with rooting depth (Smith-Martin et al., 2019, 2020).

Line 396: this terminology (wood-fine root trade-offs) is confusing as is not defined. The manuscript will benefit from better staging of these concepts in the introduction.

Lines 397-400: This is an interesting proposition, but it will benefit from more background information. Basically, can we expect different biomass allocation patterns among biomes?

Line 407: Maybe “important” is more appropriate than “vital”.

Line 411: Are you referring to ENSO or drought?

Line 413: this was not quantified.

Lines 414-415: Why is this? It just appear out of the bleu.

References

Anderson‐Teixeira KJ, Wang MMH, McGarvey JC, LeBauer DS. 2016. Carbon dynamics of mature and regrowth tropical forests derived from a pantropical database (TropForC-db). Global Change Biology 22: 1690–1709.

Aragão LEOC, Malhi Y, Roman‐Cuesta RM, Saatchi S, Anderson LO, Shimabukuro YE. 2007. Spatial patterns and fire response of recent Amazonian droughts. Geophysical Research Letters 34.

Brodribb TJ, Holbrook NM, Gutiérrez MV. 2002. Hydraulic and photosynthetic co-ordination in seasonally dry tropical forest trees. Plant, Cell & Environment 25: 1435–1444.

Eamus D. 1999. Ecophysiological traits of deciduous and evergreen woody species in the seasonally dry tropics. Trends in Ecology & Evolution 14: 11–16.

Feng X, Porporato A, Rodriguez-Iturbe I. 2013. Changes in rainfall seasonality in the tropics. Nature Climate Change 3: 811–815.

Kavanagh T, Kellman M. 1992. Seasonal Pattern of Fine Root Proliferation in a Tropical Dry Forest. Biotropica 24: 157.

Kummerow J, Castillanos J, Maas M, Larigauderie A. 1990. Production of fine roots and the seasonality of their growth in a Mexican deciduous dry forest. Vegetatio 90: 73–80.

Mattos CRC, Mazzochini GG, Rius BF, Penha D, Giacomin LL, Flores BM, Silva MC, Xavier RO, Nehemy MF, Petroni AR, et al. 2023. Rainfall and topographic position determine tree embolism resistance in Amazônia and Cerrado sites. Environmental Research Letters 18: 114009.

Olson ME, Anfodillo T, Rosell JA, Martínez‐Méndez N. 2020. Across climates and species, higher vapour pressure deficit is associated with wider vessels for plants of the same height. Plant, Cell & Environment 43: 3068–3080.

Olson ME, Soriano D, Rosell JA, Anfodillo T, Donoghue MJ, Edwards EJ, León-Gómez C, Dawson T, Camarero Martínez JJ, Castorena M, et al. 2018. Plant height and hydraulic vulnerability to drought and cold. Proceedings of the National Academy of Sciences 115: 7551–7556.

Powers JS, Vargas G. G, Brodribb TJ, Schwartz NB, Pérez‐Aviles D, Smith‐Martin CM, Becknell JM, Aureli F, Blanco R, Calderón‐Morales E, et al. 2020. A catastrophic tropical drought kills hydraulically vulnerable tree species. Global Change Biology 26: 3122–3133.

Reich PB, Borchert R. 1984. Water Stress and Tree Phenology in a Tropical Dry Forest in the Lowlands of Costa Rica. Journal of Ecology 72: 61–74.

Schwartz NB, Lintner BR, Feng X, Powers JS. 2020. Beyond MAP: A guide to dimensions of rainfall variability for tropical ecology. Biotropica 52: 1319–1332.

Smith-Martin CM, Bastos CL, Lopez OR, Powers JS, Schnitzer SA. 2019. Effects of dry-season irrigation on leaf physiology and biomass allocation in tropical lianas and trees. Ecology 100: e02827.

Smith-Martin CM, Xu X, Medvigy D, Schnitzer SA, Powers JS. 2020. Allometric scaling laws linking biomass and rooting depth vary across ontogeny and functional groups in tropical dry forest lianas and trees. New Phytologist 226: 714–726.

Sobrado MA. 2015. Leaf Tissue Water Relations Are Associated with Drought-Induced Leaf Shedding in Tropical Montane Habitats. American Journal of Plant Sciences 6: 2128–2135.

Vargas G. G, Brodribb TJ, Dupuy JM, González-M. R, Hulshof CM, Medvigy D, Allerton TAP, Pizano C, Salgado-Negret B, Schwartz NB, et al. 2021. Beyond leaf habit: generalities in plant function across 97 tropical dry forest tree species. New Phytologist 232: 148–161.

In this manuscript, the authors examined how the vegetation in the Cerrado responded to the 2015-16 El Niño event, using field measurements collected between 2014 and 2019. They discovered contrasting patterns between the cerrado and cerradão ecosystems. The dataset employed in this study is particularly valuable and unique, with intriguing results. Nevertheless, there are two major issues that the authors need to address during the revision.

1. The ‘Results’ and ‘Discussion’ sections must be better structured. At present, the authors present all the information without using sub-headings, particularly in the ‘Discussion’ section.

Below are the suggestions for the structure of the ‘Results’ section.

3.1 Total NPP and its allocation

3.2 Canopy NPP  

3.3 Stem NPP and mortality

3.4 Root NPP

3.5 Dynamics among canopy, stem and root NPP

The ‘Discussion’ should adopt a similar structure.

2. The authors defined the 12 months from May 2015 to April 2016 as the 2015-16 El Niño event (lines 157-158) and presented temperature, precipitation, and MCWD anomalies during this period compared to other years (Fig. 1). Why not use this same May-to-April definition for the remaining figures in the study? This would provide a clearer understanding of the 2015-16 El Niño event's impact on vegetation.