Improving soil moisture prediction of a high-resolution  land surface model by parameterising pedotransfer  functions through assimilation of SMAP satellite data

Pinnington, Ewan; Amezcua, Javier; Cooper, Elizabeth; Dadson, Simon; Ellis, Rich; Peng, Jian; Robinson, Emma; Morrison, Ross; Osborne, Simon; Quaife, Tristan

doi:https://doi.org/10.5194/hess-25-1617-2021

Articles | Volume 25, issue 3

https://doi.org/10.5194/hess-25-1617-2021

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

https://doi.org/10.5194/hess-25-1617-2021

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

Articles | Volume 25, issue 3

Research article

|

31 Mar 2021

Research article |

| 31 Mar 2021

Improving soil moisture prediction of a high-resolution land surface model by parameterising pedotransfer functions through assimilation of SMAP satellite data

Ewan Pinnington, Javier Amezcua, Elizabeth Cooper, Simon Dadson, Rich Ellis, Jian Peng, Emma Robinson, Ross Morrison, Simon Osborne, and Tristan Quaife

Download

Final revised paper (published on 31 Mar 2021)
Supplement to the final revised paper
Preprint (discussion started on 15 Jul 2020)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review', Anonymous Referee #1, 25 Aug 2020
- AC1: 'Response to Reviewer 1', Ewan Pinnington, 07 Oct 2020
RC2: 'Comments on "Improving Soil Moisture Prediction of a High–Resolution Land Surface Model by Parameterising Pedotransfer Functions through Assimilation of SMAP Satellite Data"', Anonymous Referee #2, 03 Sep 2020
- AC2: 'Response to Reviewer 2', Ewan Pinnington, 07 Oct 2020
RC3: 'Reviewer comments', Anonymous Referee #3, 11 Sep 2020
- AC3: 'Response to Reviewer 3', Ewan Pinnington, 07 Oct 2020

Peer-review completion

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

ED: Reconsider after major revisions (further review by editor and referees) (23 Oct 2020) by Harrie-Jan Hendricks Franssen

AR by Ewan Pinnington on behalf of the Authors (06 Nov 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (07 Nov 2020) by Harrie-Jan Hendricks Franssen

RR by Anonymous Referee #1 (07 Dec 2020)

RR by Anonymous Referee #2 (08 Dec 2020)

RR by Anonymous Referee #3 (25 Jan 2021)

Suggestions for revision or reasons for rejection

Dear Editor and authors,

I have read the revised version of the MS entitled “Improving Soil Moisture Prediction of a High–Resolution Land Surface Model by Parameterising Pedotransfer Functions through Assimilation of SMAP Satellite Data” by Pinnington and others. I can say that the authors have made a significant effort to improve the MS and to address all reviewers questions. This has resulted in a significantly better manuscript. However, I do still have concerns about the study, especially its lack of validation and overall impact on JULES predictability of other hydrometeorological variables (explained below). For that reason, I recommend moderate revisions for the paper either by introducing additional experiments/analysis or by carefully and clearly explaining why the decision for not to do so with an explanation for the possible consequences of such changes in the model performance.

My main point relates to the fact that SMAP data is ingested into the JULES-DA system to improve the estimate of static soil properties as well as then soil moisture dynamics in a region. The authors have shown for some few selected grid-points, a comparison against independent soil moisture from the COSMOS-UK network. However, it is unclear with that information, how one should expect the region to be improved or not. For example:

1. How do we know the updated regional maps of soil properties are realistic? Could the authors compare with independent maps (e.g., Cranfield soil database in the UK, or using the global SoilsGrid database) at least as a baseline?
2. How do we know the claimed improvements in the soil moisture will ultimately result in improvements to other hydrologically-relevant variables in JULES? My understanding is that a good number of COSMOS-UK stations provide at least sensible heat flux estimates and evapotranspiration can be estimated as a residual given the other surface energy balance components are also provided. Yes, there are some uncertainties in such approach but for, say, daily estimates this should give the authors more/less confidence in their experiment. Additionally, the authors could have also run their experiments trying to estimate streamflow at particular catchments and compare their results against UK’s NRFA streamflow data, although this can be a bit more complicated if the region were characterized by groundwater dominated catchments and I believe the operational JULES version doesn’t yet include such parameterizations.

Minor comments:

With regards to my question about the thickness of JULES layers against shallower SMAP estimates, I appreciate that the results from the 5-cm thick experiments were similar to the original 10-cm thick simulations. I am fine to accept that. However, the additional explanation that 10-cm thickness was kept to be consistent for the wider community is very weak in my opinion. The increase in horizontal resolution should be thought simultaneously with its vertical resolution, and old parameterizations or model structures need to be frequently “challenged” otherwise understanding of model structures and their limitations are hindered.

Hide

ED: Publish subject to revisions (further review by editor and referees) (27 Jan 2021) by Harrie-Jan Hendricks Franssen

AR by Ewan Pinnington on behalf of the Authors (24 Feb 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (25 Feb 2021) by Harrie-Jan Hendricks Franssen

Short summary

Land surface models are important tools for translating meteorological forecasts and reanalyses into real-world impacts at the Earth's surface. We show that the hydrological predictions, in particular soil moisture, of these models can be improved by combining them with satellite observations from the NASA SMAP mission to update uncertain parameters. We find a 22 % reduction in error at a network of in situ soil moisture sensors after combining model predictions with satellite observations.