Factors controlling the productivity of tropical Andean forests: climate and soil are more important than tree diversity

Homeier, Jürgen; Leuschner, Christoph

doi:https://doi.org/10.5194/bg-18-1525-2021

Articles | Volume 18, issue 4

https://doi.org/10.5194/bg-18-1525-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/bg-18-1525-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 18, issue 4

Research article

|

03 Mar 2021

Research article |

| 03 Mar 2021

Factors controlling the productivity of tropical Andean forests: climate and soil are more important than tree diversity

Jürgen Homeier and Christoph Leuschner

Download

Final revised paper (published on 03 Mar 2021)
Supplement to the final revised paper
Preprint (discussion started on 28 Sep 2020)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review of bg-2020-332', Anonymous Referee #1, 07 Oct 2020
- AC1: 'Response to Referee #1', Jürgen Homeier, 21 Dec 2020
RC2: 'Review', Anonymous Referee #2, 08 Dec 2020
- AC2: 'Reply on RC2', Jürgen Homeier, 23 Dec 2020

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (04 Jan 2021) by Sara Vicca

AR by Jürgen Homeier on behalf of the Authors (05 Jan 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (06 Jan 2021) by Sara Vicca

RR by Anonymous Referee #2 (08 Jan 2021)

Suggestions for revision or reasons for rejection

The revision and response letter addresses all the points raised; thanks very much for the effort, and also for the analysis of an important and fascinating dataset. Overall the ms is improved, but there are some key issues that have not really been elucidated enough here for publication, leaving the analysis much less complete than it should or could be. There is the odd useful but missing reference that I think would be helpful to add, but the main conceptual gap appears to be around the connection of temperature with mechanism and the relationships to traits and diversity.

The authors mention a plot size effect on analyzing diversity/productivity relationships. I thought I had mentioned the paper of Sullivan 2016 (Sci Reps) before, but perhaps not (apologies if not). This is relevant to the introductory material around line 28 and in the discussion, around line 345-355: Sullivan et al. 2016 find (small) plot size can influence inferences from their analysis of diversity and AGB/C storage in tropical forests. This needs to be included/discussed, I think.

As for the productivity-temperature-elevation discussion this seems to (i) miss some of the earlierst references to the subject (eg S Bruijnzeel's classic work) and (ii) be a bit simplified to the point that key issues are obscured (or inadvertently lost) in the current text. There is significant advantage in being specific here – eg TMF or tropical lowland forest studies often focus on NPP (or just tree growth, as here), but others also consider GPP, both from an empirical and modelling perspective, and the papers in the introduction (eg Line 44-60) refer to several different versions of ‘productivity’.

One advantage of being specific is that it allows the author to dissect better the drivers of reduced productivity (at higher elevation), especially when linking tree growth and gross primary productivity (=gross photosynthesis, GPP). This is can be powerfully done using a combination of data and mechanistic modelling. Line 46 introduces modelling work but refers only to that of Fyllas 2017. The Fyllas study uses a mechanistic model, but validates only based on annual-scale estimates of summed components of the carbon cycle. The modelling study of van der Weg 2014 is not mentioned except in reference to LMA, and yet this earlier modelling paper (the first to use a mechanistic analysis to identify temperature as the main driver of differences in GPP with elevation in TMF) validates model results based on high-frequency sap flux data as well as annual-scale component-summed carbon cycle analysis. That is, it provides validation data at a timescale relevant to the processes being driven by variation in the environment, and via the relevant mechanisms. Both the Fyllas and van der Weg papers use photosynthetic physiology at the core of their models but they come up with different interpretations of the main driver of change productivity (GPP) with elevation. As indicated in the first review, this needs to be discussed because the van der Weg findings are consistent with those in this submitted paper (ie that temperature is a dominant driver) but the Fyllas paper suggests that any effects of temperature are expressed through trait variation.

Taking the analysis presented in this submitted manuscript on Ecuadorian data a step further, the work of Wittich 2012 is cited as showing that light-saturated assimilation rates (called Amax) are lower at higher elevation, ie this is consistent with the results presented in this submitted paper (of lower wood production (WP)) at higher elevation...but I don’t know if the Wittich values are temperature-corrected, or if they are cited at ambient temperature (ie T differing by elevation).

However, this outcome (Wittich) appears to be inconsistent with the results presented by Fyllas et al 2017 who argue that Amax is constant with elevation and that this constancy is achieved through species turnover. In the Fyllas argument, the changes in species composition with elevation/ temperatures lead, on average, to a higher photosynthetic capacity at higher elevation (ie biochemical capacity, Vcmax25 – see Bahar et al. 2016 New Phytologist and van der Weg 2012 Oecologia for the data on Vcmax25), and whilst the temperature is lower, the effect on Amax (via higher Vcmax) is for Amax to be constant across elevations. For this reason, Fyllas et al argue that radiation and leaf photosynthetic traits drive productivity, not temperature (temperature is inferred to be an indirect driver of average trait variation). The data used by Fyllas are taken at leaf temperatures that do not go very cold even though we know leaves at high elevation spend much time well below 20 C (see van der Weg 2014), so the Wittich analysis might be of much interest to the analysis presented in this paper from Ecuadoran sites, depending on how temperature is handled (remember that 24 hr (/12 hr!) photosynthesis is what determines overall assimilation totals not the maximum observed value of net photosynthesis, Amax, or the maximum photosynthetic capacity, Vcmax25) .

Finally, there is a diversity question. The authors show that species diversity is very weakly related to WP in their data. The analysis also shows (and states) that variation in WP is large at each site (possibly larger than overall mean change in WP with elevation?...as also seen in Vcmax25 data, Bahar 2016, referred to above); and the argument is presented that this variation in WP reflects local variation in the environment/diversity. However, their discussion could be enhanced quite a bit by considering species composition and relations to diversity in mean traits (apols if this distinction is made and I've missed it). The Fyllas argument suggests that changes in mean traits do strongly drive variation in WP, and that the traits change with species turnover, ie species composition. Thus, it may be that raw species diversity does not explain variation in WP very well, but other elements of that diversity, ie mean trait change, conceivably might.

Overall, it seems that there is a key element missing here in synthesizing evidence for the mechanistic explanation of the observed variation in WP with elevation (ie, the argument that temperature drives the observed differences – ie, a/the principal focus of the paper). One part of it is explaining the different modelling approaches/conclusions and the other is relating this to evidence from the cited leaf physiology studies and the related mechanistic (/modelling) analysis, both in terms of the full variation in temperature at each elevation and in terms of the effect on diversity in traits (as well as species). I hope this extended comment is useful to the authors, as the data they have are fascinating and have the capacity to shed additional light on this overall discussion about a large fundamental question in forest ecology, and especially so given their additional nutrient information.

Other small comments
Line 48. Should Tanner et al. 1998 be cited here in addition to the other place(s) where it is already?

Line 51. Should van der Weg 2014 be cited here in addition to the other place(s) where it is already?

Line 53, 69, elsewhere. It’s not clear where mechanism and correlation are separated. This may be a timescale issue. The fertility discussion seems to be mainly by correlation but the impacts on photosynthesis need to be mechanistic; they are linked of course.

Line 80. The ‘predestined’ term might be replaced by ‘...which makes them attractive for the study of..’.

Line 89. The mention of assumed diversity effects is good, as we are not sure if they are real or if they only seem to occur in small plots. Perhaps this can refer to the preceding discussion as well, to tie up the idea of ‘assumed diversity effects’?

Line 262. Zimmermann 2010 (Glob Biogeochem Cycl) reports soil moisture data for a similar Andean elevation gradient that suggest little moisture limitation across elevations because of high rainfall, as here – it looks like it would be useful for you to cite here to substantiate your (reasonable) assumption.

Line 288. Need to refer to more modelling/data studies than just Fyllas here?

Line 318. How does WSG mediate this effect on productivity? Can the authors suggest a process or mechanism? Leaving it open like this seems insufficient. Variation in WSG is often associated with variation in moisture constraints because of the link between WSG and hydraulic vulnerability (in some studies)…but the authors don’t mention this. In this wet TMF environment, is it nutrients/growth rate, or even herbivory pressure, or just taxonomic identity that are related to the variance in WSG?

Line 329. Can you shorten the text here by saying that the difficulty of LAI estimation may lead to both under-estimation where leaf clumping is dominant, or over-estimation where stem density is high?

Line 349. Useful to refer here to the Sullivan 2016 work here on plot size/diversity/carbon storage.

Line 368. The analysis here is clear in that it interprets temperature as a key driver of WP-elevation variance. But in the element where future needs are considered it could usefully also refer to a need to understand variation in overall traits – either trait diversity or change in mean trait values, and how they affect major ecosystem processes including productivity, in relation to temperature effects on the core driving processes themselves (ie as well as the species diversity question).

Hide

ED: Publish subject to minor revisions (review by editor) (08 Jan 2021) by Sara Vicca

AR by Jürgen Homeier on behalf of the Authors (21 Jan 2021) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (22 Jan 2021) by Sara Vicca

AR by Jürgen Homeier on behalf of the Authors (25 Jan 2021) Author's response Manuscript

Download

Article (2293 KB)
Full-text XML

Short summary

We studied aboveground productivity in humid tropical montane old-growth forests in two highly diverse Andean regions with large geological and topographic heterogeneity and related productivity to tree diversity and climatic, edaphic and stand structural factors. From our results we conclude that the productivity of highly diverse Neotropical montane forests is primarily controlled by thermal and edaphic factors and stand structural properties, while tree diversity is of minor importance.