Canopy litterfall and soil respiration under rainfall and fog reduction in tropical montane cloud forests

Huaraca Huasco, Walter; Ezquera Condori, Maria Edith; Corrales Alvarez, Daniela; Espinoza Otazu, Blanca Rosa; Galiano Cabrera, Darcy Fernando; Puma Vilca, Beisit; Sacatuma Cruz, Roxana; Soto Quispe, Yulisa Sari; Alvarez Mayorga, Luis Miguel; Chambi Paucar, Jimmy Ronald; Horwath, Aline Barbara; Bartholomew, David C.; Metcalfe, Daniel B.

doi:10.5194/egusphere-2026-2882

Preprints

https://doi.org/10.5194/egusphere-2026-2882

Preprints

11 Jun 2026

| 11 Jun 2026

Canopy litterfall and soil respiration under rainfall and fog reduction in tropical montane cloud forests

Walter Huaraca Huasco, Maria Edith Ezquera Condori, Daniela Corrales Alvarez, Blanca Rosa Espinoza Otazu, Darcy Fernando Galiano Cabrera, Beisit Puma Vilca, Roxana Sacatuma Cruz, Yulisa Sari Soto Quispe, Luis Miguel Alvarez Mayorga, Jimmy Ronald Chambi Paucar, Aline Barbara Horwath, David C. Bartholomew, and Daniel B. Metcalfe

Abstract. Tropical montane cloud forests (TMCFs) are globally important ecosystems that act as large carbon sinks. However, climate-driven declines in rainfall and cloud immersion threaten these forests, and their responses to declines in these distinct water sources remain poorly understood. Two separate large-scale experiments to reduce throughfall (TE) and fog (FE) in a Peruvian TMCF were conducted, to compare the temporal patterns and drivers of canopy litterfall and soil respiration, with nearby control (CON) plots.

Litterfall and soil respiration declined during the relatively dry season on the CON plots. Seasonal patterns of soil respiration were related to soil moisture, while litterfall was related to air temperature. Litterfall and soil respiration on the TE plot were suppressed overall and aseasonal, although litterfall pattern was offset FE-induced increases in litterfall from fine wood structures and epiphytes. On the FE plot, the only sign of altered seasonality was elevated litterfall from reproductive structures in the late dry season relative to the CON plot. There was little consistent interannual trend in either litterfall or respiration over time under the FE treatment. By contrast, the TE treatment was associated with a consistent decline in total litterfall, mainly caused by leaf litterfall while soil respiration initially declined for the first four years of the treatment followed by a rise likely associated with a concomitant increase in soil moisture. TE and FE appear to alter the amount and seasonality of reproductive activity in the canopy which could have major consequences for stand structure in a drier future climate.

Received: 19 May 2026 – Discussion started: 11 Jun 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 1485 KB)

Supplement (365 KB)

Download & links

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-2882', Anonymous Referee #1, 06 Jul 2026
This study investigated the effects of fog and throughfall reduction on canopy litterfall and soil respiration in a tropical montane cloud forest. The authors established a fog manipulation experiment and a throughfall reduction experiment, monitored canopy litterfall and soil respiration over an eight-year period, and evaluated the influences of climatic variables on these ecosystem processes. The results suggest that throughfall reduction exerted much stronger effects than fog manipulation, with litterfall being primarily regulated by temperature and soil respiration mainly controlled by soil moisture.
The study addresses an important topic in a unique and understudied ecosystem and provides a valuable long-term dataset. However, I have a major concern regarding the experimental design that substantially limits the validity of the conclusions.
The most critical limitation is the lack of true treatment replication. In both the fog manipulation and throughfall reduction experiments, each treatment was applied to only a single plot with one corresponding control plot. Consequently, treatment effects are completely confounded with plot effects, making it impossible to distinguish treatment effects from inherent differences between plots. The multiple sampling units within each plot represent subsamples rather than independent experimental replicates and therefore cannot be treated as true replication for statistical inference. Under this design, the linear mixed-effects model and subsequent regression analyses can appropriately describe temporal dynamics and relationships between environmental variables and ecosystem responses within individual plots, but they cannot provide a statistically valid test of treatment effects, which is one of the primary objectives of this study. Because this limitation cannot be remedied through statistical analysis, I do not believe the current experimental design adequately supports the treatment comparisons presented in the manuscript. Therefore, I recommend rejection.
Specific comments
L80: Did the authors measure canopy productivity? If not, please clarify why only litterfall was used as an indicator of canopy productivity.

L96: Please state the scientific hypotheses more explicitly at the end of the Introduction.

L112: Change "where" to "when."

L113: There is a formatting issue ("around12 °C"); please insert a space.

L122–125: The throughfall reduction experiment lacks true replication. Only one treatment plot and one control plot were established, so treatment and plot effects are confounded.

L127–130: The same concern applies to the fog manipulation experiment, where only one treatment plot and one control plot were used.

Figure 1: Because the experiment began in 2017, I suggest displaying climate data only for the study period (2017–2025). The current figure contains too many lines and is difficult to interpret.

L182: "Linear mixed-fixed effects model" should be revised to "linear mixed-effects model."

L218: Change "on TE-EXP" to "in TE-EXP" (and make this correction consistently throughout the manuscript).

L227: Please describe how the annual total soil respiration was calculated or estimated.

Figure 2: The manuscript describes increases and decreases among treatments, but statistical significance is not clearly indicated in the figure. In addition, the reported n = 7 represents repeated sampling units within a treatment rather than independent treatment replicates and should not be interpreted as the sample size for treatment comparisons.

L275–276: Please revise this sentence. The statement referring to the SE is inconsistent with the information presented in the "Signif." column.
Citation: https://doi.org/10.5194/egusphere-2026-2882-RC1
RC2:
'Interesting study but methodological issues and unclear stats', Anonymous Referee #2, 22 Jul 2026
General comments
This manuscript describes an experiment in a Peruvian tropical montane cloud forest (TMCF) in which throughfall and fog ingress were (putatively at least) manipulated in large forest plots. The introduction does a nice job of laying out why this is an important and little-studied question in an era of ongoing climate change—there’s little data from TMCFs and they have unique characteristics. The experiment ran for seven years with monthly measurements of litterfall and soil respiration. The ms is generally well written and frequently interesting.
Unfortunately, there are some significant problems. First, the experimental design is formally unreplicated (one treatment and one control plot for each manipulation), but that’s okay—these are ecosystem-scale experiments in a remote location. More troubling is that no pre-treatment data are presented. Since all the authors’ inferences depend on treatment relative to control, this makes it impossible to know whether differences existed before the treatments were implemented. In addition, the efficacy of the throughfall exclusion treatment is given as 95%, but it’s unclear where this number comes from, and there’s no documentation at all on whether the fog-reduction treatment worked, and if so to what degree. Finally, it appears (lines 521-522) that the plastic sheeting of the TE treatment developed significant failures over time. I understand how difficult it is to maintain such equipment in remote areas, but this significantly affects interpretation of the results—the soil respiration ‘curve’ is disquieting!—and needs to be noted much more prominently in the methods and elsewhere!
Second, the analytical methods are far too unclear. I appreciate that the authors deposited their analytical code, but the statistical methods are vague and poorly described; it’s often unclear what test has been performed, or why. Protected (e.g. Bonferroni) tests should probably be used in some cases? Some of figure captions are unclear, and there are some cross-referencing problems (see specific comments below).
Finally, the discussion and conclusions are limited, with the latter in particular simply restating results. Both the results and discussion assert causality with regard to observational variables, and this needs to be tempered.
In summary, I’m sorry to be so negative, but between the severe limitations of the study design, lack of analytical clarity, and text problems, I think that this ms should not be considered further.

Specific comments
Line 37: “was offset by “?

66-67: awkward transition; consider removing the first sentence of this paragraph

81: “However, experiments have”

92-95: this seems tacked on and out of place; consider integrating into final sentence of previous paragraph

117: “are the dominant”

123: what is this 95% based on? Measurements? Ocular estimation? A guess?

127-130: does it work?

141: why remove these data? OK, it’s anomalously high compared to the other traps, but your inferences about treatment effects depend on changing differences between EXP and CON, not on the absolute values, right?

152: 120 seconds?

170: be more explicit. What are “conventional quadrature rules”?

176-: document version numbers of all packages used

Figure 1: why not simply show the monthly average across years? Seems much simpler and clearer than “marginal predictions from a GAM”. Also, define all acronyms in figure caption

219: “notably suppressed” – stats to back this up? Similar comment applies to l. 226, 228, and elsewhere

240: Figure 2a?

241: what’s the test being performed here?

252: “main driver” – careful about asserting causality of things you’re not manipulating

260-: there are lots of individual correlation tests being performed here; it seems like a protected test of some sort would be more robust? Clarify

283-284: this mean and se is based on trap-to-trap variability? Clarify

316: Figure 5a?

344-350 and elsewhere: try not to restate results in discussion

430: the conclusions simply restate the results; remove
Citation: https://doi.org/10.5194/egusphere-2026-2882-RC2

Supplement

https://doi.org/10.5194/egusphere-2026-2882-supplement

Viewed

Total article views: 137 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
103	24	10	137	17	6	6

HTML: 103
PDF: 24
XML: 10
Total: 137
Supplement: 17
BibTeX: 6
EndNote: 6

Views and downloads (calculated since 11 Jun 2026)

Month	HTML	PDF	XML	Total
Jun 2026	75	10	7	92
Jul 2026	28	14	3	45

Cumulative views and downloads (calculated since 11 Jun 2026)

Month	HTML	PDF	XML	Total
Jun 2026	75	10	7	92
Jul 2026	28	14	3	45

Viewed (geographical distribution)

Total article views: 122 (including HTML, PDF, and XML) Thereof 122 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 25 Jul 2026

Short summary

Tropical montane cloud forests are globally important ecosystems threatened by drought, but the impacts of these climate changes remain unclear. We experimentally reduced rainfall and fog in two large patches of cloud forest in Peru, then measured key carbon fluxes in these plots and nearby controls, to provide insights into the likely effects of future drought on carbon cycling in these systems.


Total:	0
HTML:	0
PDF:	0
XML:	0