Challenges and opportunities in land surface modelling of savanna ecosystems

Whitley, Rhys; Beringer, Jason; Hutley, Lindsay B.; Abramowitz, Gabriel; De Kauwe, Martin G.; Evans, Bradley; Haverd, Vanessa; Li, Longhui; Moore, Caitlin; Ryu, Youngryel; Scheiter, Simon; Schymanski, Stanislaus J.; Smith, Benjamin; Wang, Ying-Ping; Williams, Mathew; Yu, Qiang

doi:https://doi.org/10.5194/bg-14-4711-2017

Articles | Volume 14, issue 20

https://doi.org/10.5194/bg-14-4711-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Special issue:

OzFlux: a network for the study of ecosystem carbon and water...

https://doi.org/10.5194/bg-14-4711-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 14, issue 20

Reviews and syntheses

|

24 Oct 2017

Reviews and syntheses |

| 24 Oct 2017

Challenges and opportunities in land surface modelling of savanna ecosystems

Rhys Whitley, Jason Beringer, Lindsay B. Hutley, Gabriel Abramowitz, Martin G. De Kauwe, Bradley Evans, Vanessa Haverd, Longhui Li, Caitlin Moore, Youngryel Ryu, Simon Scheiter, Stanislaus J. Schymanski, Benjamin Smith, Ying-Ping Wang, Mathew Williams, and Qiang Yu

Download

Final revised paper (published on 24 Oct 2017)
Preprint (discussion started on 11 May 2016)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Regarding "Challenges and opportunities in modelling savanna ecosystems"', Anonymous Referee #1, 06 Jun 2016
RC2: 'Comments', Anonymous Referee #2, 09 Jun 2016
RC3: 'Good Highlights of Improvements Needed for TBMs But Omits Some Important Processes and Theory', Anonymous Referee #3, 08 Jul 2016
AC1: 'Response to reviewer comments', Rhys Whitley, 15 Aug 2016

Peer-review completion

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

ED: Reconsider after major revisions (19 Aug 2016) by Dario Papale

AR by Jason Beringer on behalf of the Authors (22 Feb 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (22 Feb 2017) by Dario Papale

RR by Anonymous Referee #3 (08 Mar 2017)

Suggestions for revision or reasons for rejection

This is a good paper with something to add and the revisions help. The paper still has inconsistencies throughout regarding framing and scope and this should be fixed.

The authors have done a good job of highlighting some priorities for improving the capacity of TBMs to simulate carbon, water, and energy exchanges between savannas and the atmosphere on hourly to yearly time scales. They have also highlighted several opportunities and needs for datasets that could be brought to bear on this challenge. However, the paper does not do a good job of making a case for how this will build predictive capacity of savanna ecological dynamics, writ large, and as they change in response to global change factors. This appears as a motivating factor woven throughout the paper, and meaningful progress on this broad theme is promised to result from improving TBM capacity to simulate fast time scale mass and energy exchanges with the atmosphere. However, this would seem to be a misleading hope.

1) Improvements of Modeled Savanna Ecology and Dynamics or Just Water and Carbon Fluxes?

The title and many elements scattered throughout the paper suggest that this review will highlight the key model advances needed to improve predictive capacity of savanna ecology (structure and composition) as it responds to the changing climate, atmospheric composition, and land use. In other parts of the paper, the aim is suggested to be to improve simulations of ecosystem scale water and carbon dioxide fluxes. The response letter also narrows the scope to “improve the predictive capability of models in simulating the turbulent fluxes of savannas”, and “first-order controls on savanna water and carbon exchange”. These two framings reveal a cognitive dissonance that appears throughout the paper. Addressing them requires different priorities. This needs to be fixed.

Section 2.2 in the manuscript (L 297) opens up with a sweeping context for TBMs, suggesting that TBMs need to be developed to capture effects of climate change, including changes in the structure and diversity of global terrestrial ecosystems and biogeochemical fluxes. If the aim is really meant to be as broad as the prediction of savanna ecological dynamics, there is a mismatch with what a wide swath of TBMs would be capable of achieving. For this, TBMs are arguably not the right tool unless they have a wide range of additional features that capture vegetation dynamics, not just half-hourly to daily fluxes of water and carbon.

The paper also flip-flops on whether DGVMs fall under the umbrella of TBMs (L 358) or not (e.g. L 101 and elsewhere), with conflicting classification in different parts of the paper. At the very least, the class of TBMs that have imposed phenology from data inputs on vegetation cover, type, and leaf area, need to be questioned at their core regarding applicability to prediction of savanna ecology. Futzing with those, or even the others, to match EC-measured CO2 and H2O fluxes would be expected to do little if anything to help develop their capacity to predict savanna ecological dynamics. Getting models to simulate ET and GPP that matches hourly to monthly flux tower estimates is not going to transform TBMs into tools with predictive capability for adequately simulating savanna dynamics and their response to future global change (L 106). The paper needs to be clear about its scope and the scope of the range of relevant models.

Is the paper about adjusting TBMs to do a better job of simulating hourly to seasonal land-atmosphere fluxes of water and carbon dioxide, or is it about developing models that can predict vegetation changes with climate and other forcings (DGVM needed here)? The authors need to be specific about the simulated processes they hope TBMs could represent (e.g. L 135), and about the limitations of using TBMs absent of vegetation dynamics for prediction of the future of savanna ecology and this biome. Predicting changes in vegetation conditions (composition, structure, functional capacity) with various forcings such as climate change or atmospheric CO2 would seem to require a model with vegetation dynamics.

It would also be worth discussing (questioning) whether the competitive interactions that are implicit to at least some DGVMs are adequately representative of what happens in the real world, and can accurately reproduce tree-grass mixtures, C3-C4 distributions, and such. Are improvements in phenology, root water uptake, and disturbance expected to get this right? Will the data priorities ensure this? It’s not really clear. TBMs, and even DGVMs, are generally not well equipped to characterize changes in vegetation cover (L 305) and species composition. Neither are eddy flux measurements or phenocams for that matter (L132-134). Vegetation cover, structure, and composition (species of PFT) can vary widely for the same GPP in semi-arid systems, given strong influences of competitive interactions, disturbances, allocation and allometric attributes, and other factors. It would help if the paper could be clearer on all of this.

In sum, it remains questionable whether the three improvements (phenology, root zone water, and disturbances) would be adequate for transforming TBMs to become tools that can predict changes in ecosystem structure and diversity as they respond climate change, and make them capable of “answer(ing) questions on the future of this biome”, as suggested at the end of the introduction (L 129). Therefore, I would argue for edits that narrow the framing and scope to what is more realistic and in line with the tools being emphasized in this paper (TBMs and fast-timescale fluxes).

Hide

ED: Reconsider after major revisions (13 Mar 2017) by Dario Papale

AR by Jason Beringer on behalf of the Authors (06 Aug 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (15 Aug 2017) by Dario Papale

RR by Anonymous Referee #3 (28 Aug 2017)

ED: Publish as is (28 Aug 2017) by Dario Papale

AR by Jason Beringer on behalf of the Authors (06 Sep 2017)

Short summary

This paper attempts to review some of the current challenges faced by the modelling community in simulating the behaviour of savanna ecosystems. We provide a particular focus on three dynamic processes (phenology, root-water access, and fire) that are characteristic of savannas, which we believe are not adequately represented in current-generation terrestrial biosphere models. We highlight reasons for these misrepresentations, possible solutions and a future direction for research in this area.