Assessing vegetation structure and ANPP dynamics in a grassland–shrubland Chihuahuan ecotone using NDVI–rainfall relationships

Moreno-de las Heras, M.; Díaz-Sierra, R.; Turnbull, L.; Wainwright, J.

doi:https://doi.org/10.5194/bg-12-2907-2015

Articles | Volume 12, issue 10

https://doi.org/10.5194/bg-12-2907-2015

© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/bg-12-2907-2015

© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 12, issue 10

Research article

|

21 May 2015

Research article |

| 21 May 2015

Assessing vegetation structure and ANPP dynamics in a grassland–shrubland Chihuahuan ecotone using NDVI–rainfall relationships

M. Moreno-de las Heras, R. Díaz-Sierra, L. Turnbull, and J. Wainwright

Download

Final revised paper (published on 21 May 2015)
Supplement to the final revised paper
Preprint (discussion started on 06 Jan 2015)
Supplement to the preprint

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC C144: 'Comments to the authors', Anonymous Referee #1, 11 Feb 2015
- AC C777: 'Responses to Anonymous Referee#1's comments', Mariano Moreno de las Heras, 25 Mar 2015
RC C306: 'Comments on "Assessing vegetation structure and ANPP dynamics in a grassland-shrubland Chihuahuan ecotone using NDVI-rainfall relationships"', Anonymous Referee #2, 25 Feb 2015
- AC C801: 'Responses to Anonymous Referee#2's comments', Mariano Moreno de las Heras, 25 Mar 2015

Peer-review completion

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

AR by Mariano Moreno de las Heras on behalf of the Authors (25 Mar 2015) Author's response Manuscript

ED: Referee Nomination & Report Request started (28 Mar 2015) by Trevor Keenan

RR by Anonymous Referee #2 (12 Apr 2015)

RR by Anonymous Referee #1 (21 Apr 2015)

Suggestions for revision or reasons for rejection

The authors have provided a detailed list of answers to my comments on the first version of this manuscript. I appreciate their efforts to better explain their point of view although I am not entirely convinced by some of their choice (see below). I acknowledge that a number of clarifications have been made and that the manuscript has improved because of these changes.

There has been a misunderstanding of my comments regarding model structure. I did not call for a more complex model than the one presented in the first section, as it is obvious that the available empirical data would not allow to properly calibrating and validating it. I fully reckon the merit of low-dimensional model that capture the essentials of ecohydrological dynamics in drylands and that could be easily confronted to time series of vegetation greenness. The limitations of these models should be discussed and, above all, their overall performance to represent ecosystem functioning and dynamics should be assessed. The authors have provided evidence to meet these requirements.

In the context of this study, I understand the usefulness of introducing a simple and process-oriented model of the coupled dynamics between plant growth and soil water availability. I still regret that the presentation of this model is somewhat disconnected from the remote sensing study. Once again, I would have preferred a phenomenological analysis of greenness with a statistical approch emphasizing the effect of precedent cumulative rainfall on plant growth. This should conduct to optimize the parameters Olrs and Olrh and to provide the rationale for a NDVI decomposition in mixed stands. In a further step, it could be shown that the parameter estimates (57 and 145 days) are consistent with the Rietkerk conceptual model providing certain combination of plant growth and plant mortality rates (isolines of figure 1 c) and given a fixed set of values for the other parameters. Then it should be mentionned that published parameters (gmax and m) for grass and shrubs fall in the range of inferred values using optimized Olr but also that NDVI data alone do not provide enough constraint to narrow the domain of plausible values. I have the feeling that this would provide a tighter connection between the two sections of the manuscript. In this logic, the first estimates of optimal rainfall accumulation length (Olr) using parameters from the litterature are unnecessary.

I am not very satisfied with the answer to my comments on the performance of the NDVI decomposition. Obviously, we are talking about the assessment of model performance when considering the entire time series (not a single ti). Consider a MODIS pixel with known percentages of shrub and grass cover, reconstruct the time series of greenness/biomass for this mixed stand using the observed rainfalls and the optimized parameters Olrh and Olrs, how much of the observed NDVI dynamics are you capturing ? how does the bivariate relationship between modelled and predicted NDVI looks like ?

Hide

ED: Publish subject to technical corrections (21 Apr 2015) by Trevor Keenan

AR by Mariano Moreno de las Heras on behalf of the Authors (29 Apr 2015) Author's response Manuscript

Download

Article (4118 KB)
Full-text XML

Short summary

Exploration of NDVI-rainfall relationships provided ready biophysically based criteria to study the spatial distribution and dynamics of ANPP for herbaceous and shrub vegetation across a grassland-shrubland Chihuahuan ecotone (Sevilleta NWR, New Mexico). Overall our results suggest that shrub encroachment has not been particularly active for 2000-2013 in the area, although future reductions in summer precipitation and/or increases in winter rainfall may intensify the shrub-encroachment process.