A distributed simple dynamical systems approach (dS2 v1.0) for computationally efficient hydrological modelling at high spatio-temporal resolution

Buitink, Joost; Melsen, Lieke A.; Kirchner, James W.; Teuling, Adriaan J.

doi:https://doi.org/10.5194/gmd-13-6093-2020

Articles | Volume 13, issue 12

https://doi.org/10.5194/gmd-13-6093-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-13-6093-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 13, issue 12

Model description paper

|

02 Dec 2020

Model description paper |

| 02 Dec 2020

A distributed simple dynamical systems approach (dS2 v1.0) for computationally efficient hydrological modelling at high spatio-temporal resolution

Joost Buitink, Lieke A. Melsen, James W. Kirchner, and Adriaan J. Teuling

Download

Final revised paper (published on 02 Dec 2020)
Supplement to the final revised paper
Preprint (discussion started on 14 Jun 2019)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Interactive comment on “A distributed simple dynamical systems approach (dS2 v1.0) for computationally efficient hydrological modelling” by Joost Buitink et al.', Anonymous Referee #1, 16 Aug 2019
- AC1: 'Response to Anonymous Referee #1', Joost Buitink, 22 Aug 2019
RC2: 'Yet another conceptual model?', Anonymous Referee #2, 06 Dec 2019
- AC2: 'Response to Anonymous Referee #2', Joost Buitink, 19 Dec 2019

Peer-review completion

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

AR by Joost Buitink on behalf of the Authors (10 Jan 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (20 Apr 2020) by Didier Roche

RR by Anonymous Referee #2 (29 Apr 2020)

Suggestions for revision or reasons for rejection

After having read the authors' response and the revised manuscript I think the authors well addressed the main concerns. Their dS2 model incorporates a distinct conception (in comparison to widely used bucket-based conceptual models) and is also an advancement in comparison to the SIMPLEFLOOD model, which is (as it seems) the only other known model employing the simple dynamical system approach. In my opinion this justifies the publication of their work, even though I still do not think we could learn more from the model than from other models. Besides, many aspects in the paper are now better understandable or justified (e.g. with respect to sensitivity analysis).

The example provided by the authors helps to understand their model approach and I was able to reproduce the example. However, some efforts were necessary: As it seems the model assumes backslashes when analyzing paths (typical for Windows) and I had to adjust some lines of the model's source code (occurrences of f.split('\\')[-1].split("_")) to run it on my Linux system. From a user's point of view this is rather annoying and could be made more user friendly. Furthermore, one still has to dive into the source code, e.g., to find out what output the model provides. While I think these are not arguments against a publication of the paper, the model is still lacking a bit of user-friendliness to attract more attention.

I found three remaining minor issues that should be addressed / amended before publication:

- Evaporation input: Does the model require actual or potential evapotranspiration as input? I thought the latter but now I am not sure anymore due to page 4, lines 28 ff. (is E_input potential evaporation?).

- Adaptive time stepping: "The most simple method is the explicit Euler scheme. This is a first order estimation, meaning that the value at tn+1 is based on the values given at tn." (page 11 lines 9 f.). This sentence seems misleading to me. Also in case of higher order methods the value at tn+1 is based on the value given at tn (except for linear multistep methods). The distinction is the number of intermediate stages / midpoints between tn and tn+1. Besides, first order means that the error of the numerical solution is directly proportional to the time step size. I.e. increasing the time step size (h=2) increases the error by a factor of two (2^1), in case of a 2nd order scheme even by four (2^2), and so on.

- Fig. 10: (e) Is the dashed green line (Rietholzbach, no routing) missing or just not visible due to overlay? (a) The small overview graphic is overlaying the scale bar.

Hide

ED: Publish subject to technical corrections (28 Sep 2020) by Richard Neale

AR by Joost Buitink on behalf of the Authors (02 Oct 2020) Author's response Manuscript

Short summary

This paper presents a new distributed hydrological model: the distributed simple dynamical systems (dS2) model. The model is built with a focus on computational efficiency and is therefore able to simulate basins at high spatial and temporal resolution at a low computational cost. Despite the simplicity of the model concept, it is able to correctly simulate discharge in both small and mesoscale basins.