Data assimilation for continuous global assessment of severe conditions over terrestrial surfaces

Albergel, Clément; Zheng, Yongjun; Bonan, Bertrand; Dutra, Emanuel; Rodríguez-Fernández, Nemesio; Munier, Simon; Draper, Clara; de Rosnay, Patricia; Muñoz-Sabater, Joaquin; Balsamo, Gianpaolo; Fairbairn, David; Meurey, Catherine; Calvet, Jean-Christophe

doi:https://doi.org/10.5194/hess-24-4291-2020

Articles | Volume 24, issue 9

https://doi.org/10.5194/hess-24-4291-2020

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

https://doi.org/10.5194/hess-24-4291-2020

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

Articles | Volume 24, issue 9

Research article

|

02 Sep 2020

Research article |

| 02 Sep 2020

Data assimilation for continuous global assessment of severe conditions over terrestrial surfaces

Clément Albergel, Yongjun Zheng, Bertrand Bonan, Emanuel Dutra, Nemesio Rodríguez-Fernández, Simon Munier, Clara Draper, Patricia de Rosnay, Joaquin Muñoz-Sabater, Gianpaolo Balsamo, David Fairbairn, Catherine Meurey, and Jean-Christophe Calvet

Download

Final revised paper (published on 02 Sep 2020)
Supplement to the final revised paper
Preprint (discussion started on 25 Oct 2019)
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 of the paper by Albergel et al., entitled "Data assimilation for continuous global assessment of severe conditions over terrestrial surfaces"', Anonymous Referee #1, 21 Nov 2019
- AC1: 'Response to Reviewer#1', Clément Albergel, 22 Jan 2020
RC2: 'Possibly a milestone in the development of LDAS-Monde', Anonymous Referee #2, 30 Nov 2019
- AC2: 'Response to Reviewer#2', Clément Albergel, 22 Jan 2020
RC3: 'Review - LDAS-Monde application', Anonymous Referee #3, 30 Nov 2019
- AC3: 'Response to Reviewer#3', Clément Albergel, 22 Jan 2020
RC4: 'Review', Anonymous Referee #4, 07 Dec 2019
- AC4: 'Response to Reviewer#4', Clément Albergel, 22 Jan 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) (26 Jan 2020) by Harrie-Jan Hendricks Franssen

AR by Clément Albergel on behalf of the Authors (28 Jan 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (28 Jan 2020) by Harrie-Jan Hendricks Franssen

RR by Anonymous Referee #4 (19 Feb 2020)

Suggestions for revision or reasons for rejection

This is my second review of the paper, after the reviewers revised the originally
submitted manuscript.

The reviewers fixed a number of obvious errors and confusing statements, including
my original "minor" comments (#9-16) as well as my major comments #2 (re. snow
variables) and #8 (re. evaporation).

However, I am still quite disappointed with the revised manuscript, for three
key reasons:

A) My original major comments #1 and # suggested de-emphasizing the "validation"
of the LDAS-Monde output against the assimilated observations and making it
clear throughout the paper when "validation" is against the assimilated obs.

In response, the authors *added* more "validation" results against the assimilated
observations. The new figure 15 shows correlations against ASCAT SWI prior to
rescaling. Simply omitting the rescaling of the ASCAT SWI does *not* make the ASCAT
SWI data independent of the assimilated observations.

B1) My original major comment #3 asked to emphasize the validation against independent
(in situ) measurements by moving the relevant graphics from the Supplement to the
main text. The authors reply that much of this has already been published in earlier
papers (Response to 4.3), which is acceptable if explained and referenced properly.
However, this was not clear from the originally submitted manuscript, and rather
than clearly referencing the earlier validation results in the revised version,
the authors chose to keep the section on in situ validation results (section 3.1.2)
without a single reference within this section to the earlier publications, where
much of the same results have already been published, according to the authors.
If that is indeed the case, then the section has to be deleted. If there are new
results here (perhaps a more global set of validation sites), then there need to
be graphics in the main text.

B2) The authors added 95% confidence intervals for correlations, results in terms
of raw and anomaly R values, and metrics for the 4th soil layer (10-20cm). None of
this changed the fact that the in situ validation results indicate no statistically
significant improvement or degradation in soil moisture skill. The numbers in Table
S3 show that this is true for each network individually, in terms of R, anomaly R,
and ubRMSE. The results in Fig 7 over CONUS only do not seem consistent with
the numbers quoted in the text that talk about significant improvements at 186
stations and significant degradation at 58 stations (Lines 505-510). The results
in Table S3 simply do not support the language of "added value of assimilating
[..] soil moisture" (e.g., Line 26).

C) My major comment #5 identified more than a dozen instances of careless editing.
The authors mostly fixed these issues, but the manuscript still shows a
considerable lack of care in the editing of English style and grammar, to the point
where the text is difficult to understand at times. There is at least one native
English speaker on the author list, and there are at least two more who have
been living and working in the U.K. for decades. Did they read the manuscript???
Here are a few random examples from just the first few pages:

a) Abstract, lines 34-41:

"..., and we demonstrate the added value of initializing [...] forecasts [...].
This is particularly true for LAI that evolves on longer time space than SSM
and is more sensitive to initial conditions than to atmospheric forcing,
even at 8-day lead time. This confirms that slowly evolving ..."

What is "This" in each of these two sentences?

b) Lines 108-116:

"Are presented in this study: [Incomplete sentence.]

- An evaluation at global scale using diverse and complementary dataset such as [...].
It is also validated... [What is "it"???]"

c) Lines 125-126:

"The paper is organised in four sections as it [??] follows: section 2 details the various
components constituting LDAS-Monde: the ISBA..."

[Note the repeated colon.]

In summary, I find the manuscript to fall short of the standard in HESS and
recommend that the paper be rejected.

Hide

ED: Reconsider after major revisions (further review by editor and referees) (07 Mar 2020) by Harrie-Jan Hendricks Franssen

AR by Clément Albergel on behalf of the Authors (14 Apr 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (16 Apr 2020) by Harrie-Jan Hendricks Franssen

RR by Anonymous Referee #3 (01 May 2020)

RR by Anonymous Referee #4 (19 Jun 2020)

Suggestions for revision or reasons for rejection

Re-re-review

Title: Data assimilation for continuous global assessment of severe conditions
over terrestrial surfaces

Authors: Albergel, Zheng, Bonan, ..., Calvet

This is my third review of the paper, after the reviewers revised the previously revised version.

I encourage ALL authors to read my review in its entirety.

The authors responded to my first key concern (A) about the lack of clarity re. the
evaluation of the LDAS-Monde results against the assimilated observations.

The authors also responded to my second key concern (B) about the lack of
prominence of the validation vs. in situ measurements.

In both cases, I am not euphoric about the changes that have been applied,
but they are good enough for me to recommend publication in HESS, provided the
authors address whatever the Editor may decide regarding my third key
concern.

My third key concern (C) was about the careless editing of the manuscript, and
in particular the considerable lack of care in the editing of English style and grammar,
to the point where the text is difficult to understand at times.

The authors claim that they have addressed this as best they can and
point to the fact that there will be a copy-editing process that can take
care of anything they might have missed.

While there may have been some fixes, awkward sentences and very simple
grammatical mistakes still abound. Below I list a few representative
examples.

There are two native English speakers on the author list (Draper, Fairbairn),
and several who have lived in the UK for a decade or more (Balsamo, de Rosnay,
Munoz-Sabater). Given this, I have great difficulty understanding how the manuscript
can still be in such poor shape in terms of the language.

Did the co-authors actually read the revised manuscript? If not, there is clearly a problem.

If they did read the manuscript, did they simply not care? If so, there is clearly a problem.

Why do these well-respected authors choose to present a poorly edited manuscript to the
reviewers after the 2nd round of revisions? If they don't have the time to work on the
paper, they should not be a coauthors.

Note that the "Author Contributions" (Line 771) statement includes:
"Writing—review and editing, All". The authors should be embarrassed.

I urge the authors to consider that their lack of care in editing a manuscript
before it goes to a (volunteer) reviewer reflects very poorly on their
credibility. Can their code and scientific work be trusted if their
writing and editing shows such a blatant lack of care?

Do the authors really want readers to put the manuscript aside because
it is simply too tedious to read?

Would the authors happily review a poorly edited manuscript three times?

The following sentences are just a few examples of text that is difficult
to read. I do *not* expect that these would be fixed by the copy editing process.

Lines 73-75:
Satellite EOs related to the terrestrial hydrological, vegetation and energy cycles
are now available at a global scale at kilometric scale and below and with long-term
records...

Lines 104-109:
In this paper, stemming from previous studies referenced above, it is shown that LDAS-Monde,
by integrating jointly Surface Soil Moisture (SSM) and Leaf Area Index (LAI) EOs into
the ISBA (Interaction between Soil Biosphere and Atmosphere) LSM (Noilhan and Planton,
1989, Noilhan and Mahfouf, 1996) at a global scale and in offline mode, can be used to
detect, monitor and forecast the impact of extreme events on LSVs.

Lines 214-215:
For values lower than 2 m2m-2, a fixed value of 0.04 m2m-2 has been used.

Lines 254-255:
In forecasting mode, HRES forecast is also available everyday from 00:00 UTC with a 10-day lead time, but with changes in the temporal resolution.

Lines 319-320:
Forecasts with a four and height day lead time are presented, only (LDAS_fc4
and LDAS_fc8, respectively).

Lines 339-341:
Sensors at 5 cm depth (SSM) are compared with soil moisture from LDAS_ERA5 third
layer of soil (4-10 cm), sensors at 20 cm depth with the fourth layer of soil
(10-20 cm, 685 stations from 10 networks).

Lines 352-355:
However, several studies (e.g, Zhang et al., 2016, Sun et al., 2017, Leroux et al., 2018)
have found that their time dynamics investigated, highlighting the potential of SIF products
to be used as a validation support for GPP models.

The following examples are (mostly) simple typos and will likely be caught in copy
editing. However, they are all taken from just one page (page 12)!!!
Also, the paragraph in question (Lines 388-422) is far too long and should
have been broken into smaller pieces.

Lines 397-398:
The number of points considered per latitudinal stripes of 0.25° is represented, also.

Line 400:
Improvements from the analysis occurs from...

Line 408:
Score differences are computed as follow, ...

Lines 413-415:
LAI and SSM being assimilated variables, the analysis leads to a clear improvement
in both correlation and RMSD.
[Poor grammar.]

Lines 419-420:
...(to a lesser extend than LAI)...

Lines 420-421:
As seen on Figure 5 d) and i), there is little impact on variable EVAP which can
be considered negligible.

Line 424:
The Number of available data...

Hide

ED: Publish subject to minor revisions (review by editor) (20 Jun 2020) by Harrie-Jan Hendricks Franssen

AR by Clément Albergel on behalf of the Authors (08 Jul 2020) Author's response Manuscript

ED: Publish subject to technical corrections (30 Jul 2020) by Harrie-Jan Hendricks Franssen

AR by Clément Albergel on behalf of the Authors (30 Jul 2020) Author's response Manuscript

Short summary

LDAS-Monde is a global offline land data assimilation system (LDAS) that jointly assimilates satellite-derived observations of surface soil moisture (SSM) and leaf area index (LAI) into the ISBA (Interaction between Soil Biosphere and Atmosphere) land surface model (LSM). This study demonstrates that LDAS-Monde is able to detect, monitor and forecast the impact of extreme weather on land surface states.