Multi-scale assessment of a grassland productivity model

Taylor, Shawn D.; Browning, Dawn M.

doi:https://doi.org/10.5194/bg-18-2213-2021

Articles | Volume 18, issue 6

https://doi.org/10.5194/bg-18-2213-2021

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

https://doi.org/10.5194/bg-18-2213-2021

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

Articles | Volume 18, issue 6

Research article

|

01 Apr 2021

Research article |

| 01 Apr 2021

Multi-scale assessment of a grassland productivity model

Shawn D. Taylor and Dawn M. Browning

Download

Final revised paper (published on 01 Apr 2021)
Supplement to the final revised paper
Preprint (discussion started on 09 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 comments', Anonymous Referee #1, 13 Oct 2020
- AC2: 'Response to Referee 1', Shawn Taylor, 17 Nov 2020
RC2: 'comments', Anonymous Referee #2, 27 Oct 2020
- AC1: 'Response to Referee 2', Shawn Taylor, 17 Nov 2020

Peer-review completion

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

ED: Reconsider after major revisions (22 Nov 2020) by Ben Bond-Lamberty

AR by Shawn Taylor on behalf of the Authors (28 Dec 2020) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (29 Dec 2020) by Ben Bond-Lamberty

RR by Anonymous Referee #1 (15 Jan 2021)

Suggestions for revision or reasons for rejection

I find the revised manuscript “Multi-scale assessment of a grassland productivity model“ to be much improved. I find the new introduction paragraph about “low-dimensional” models interesting and the expanded method section on estimating optimal parameter values (as well as the appendix) helpful.

• The authors emphasize in the responses to individual comments of both reviewers that the methods (e.g., model version, the input datasets, model evaluation) were carried out exactly as by Hufkens et al. 2016 – however, this point (and the rationale for why this was done) remains unclear in the manuscript. The method section in the manuscript should spell out that this work exactly re-created the methods used by Hufkens et al. 2016 (in terms of model formulation, input data, evaluation, etc.) but with more data, more sites, and more vegetation types. For instance, the section “2.3 Environmental Data” should mention that these datasets were chosen because these are the ones used by Hufkens et al. 2016 (and not because they are the most adequate), etc.
• I can understand that the authors chose to the Global Soil Data Task Group dataset because it re-creates the Hufkens et al. 2016 analysis – however, this reason needs to be explained in the manuscript (see above). In the response, the authors incorrectly state that none of the datasets suggested by reviewer 1 (see comment to original page 3 line 11) provide required variables (field capacity and wilting point). For instance, both NRCS soil data (gNATSGO, gSSURGO) and POLARIS provide water content at field capacity and wilting point (references in original comment of reviewer 1).
• I do not follow the authors’ response to reviewer 2 comment asking why “GPP data observed at fluxnet sites” were not used to evaluate PhenoGrass. The authors argue in their response that PhenoGrass predicts fCover and not GPP and that therefore GPP data cannot be used. However, the manuscript asserts that fCover is derived from Gcc by a scaling factor (supplement and page 2 line 23) and that “Gcc … is highly correlated … flux tower derived primary productivity (Yan et al., 2019; Toomey et al., 2015)”. So, I do not understand why fCover cannot be compared against observed GPP from fluxnet sites.
• The revised version clarified that R2 was the coefficient of determination and named it unambiguously NSE. The last paragraph of the discussion section draws comparisons with Hufkens et al. 2016 (page 10 lines 6ff) “they had an average R2 of 0.71” – please clarify in the manuscript whether that is NSE or some other variant/interpretation of R2.

Hide

ED: Publish subject to minor revisions (review by editor) (16 Jan 2021) by Ben Bond-Lamberty

AR by Shawn Taylor on behalf of the Authors (10 Feb 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (11 Feb 2021) by Ben Bond-Lamberty

Download

Article (2837 KB)
Full-text XML

Short summary

Grasslands in North America provide multiple ecosystem services and drive the production of a lot of grain, beef, and other staples. We evaluated a grassland productivity model using nearly 500 years of grassland camera data and found the areas where the model worked well and locations where it did not. Long-term productivity projections for the suitable locations can be made immediately with the current model, while other areas, such as the southwest, will need further model development.