Comments on “The importance of vegetation to understand terrestrial water storage variations”

This study uses a conceptual hydrological model to consider the influence of vegetation on transpiration (via vegetation fraction), the maximum available soil water for plant (via rooting depth), and the partitioning of infiltration and runoff (via the infiltration/runoff parameter being the function of vegetation fraction [it is however not clear how p_berg partition infiltration and runoff though]). The daily climatology of EVI, and several rooting depth (and maximum soil water storage capacity and maximum plant available water capacity) datasets are used here. And then, the model is calibrated using TWS anomalies from GRACE, ESA CCI Soil Moisture, FLUXCOM-RS ET and also runoff data from GRUN. The results show that with or without vegetation dynamics, seasonal storage variations from B and VEG are not that much difference. Nevertheless, including vegetation changes dramatically the contributions of soil moisture, deep soil water and slow water storages to TWS variations. It also shows in the B simulation, soil moisture dominates TWS variations, while in the VEG simulation, the role of deeper and delayed water storage becomes prominent. Although the paper reads interesting, this reviewer finds some additions are needed to clarify/reinforce the discussions.

Major comments:

1. The current manuscript only used two numerical experiments. One with and another without vegetation. And highlight that the VEG is different than the current approach of using plant functional types or land cover classes. Nevertheless, the comparison between the VEG and the ‘traditional approach’ is not presented. This reviewer suggest the authors to add simulation results of the traditional approach. As such, the add-value of using dynamic vegetation can be demonstrated more clearly.

2. It is to note that some studies (see below and some literatures mentioned in the attachment) have dealt with the impact of dynamic vegetation on land surface processes, land-atmosphere interactions, etc. please help to discuss your novelty vs. what has been done.

Weiss, M., van den Hurk, B., Haarsma, R. et al. Impact of vegetation variability on potential predictability and skill of EC-Earth simulations. Clim Dyn 39, 2733–2746 (2012). https://doi.org/10.1007/s00382-012-1572-0

BO Christoffersen, N Restrepo-Coupe, MA Arain …,  Mechanisms of water supply and vegetation demand govern the seasonality and magnitude of evapotranspiration in Amazonia and Cerrado, Agricultural and Forest meteorology, 2014 https://doi.org/10.1016/j.agrformet.2014.02.008

Weiss, M., Miller, P. A., van den Hurk, B. J. J. M., van Noije, T., Ştefănescu, S., Haarsma, R., van Ulft, L. H., Hazeleger, W., Le Sager, P., Smith, B., & Schurgers, G. (2014). Contribution of Dynamic Vegetation Phenology to Decadal Climate Predictability, Journal of Climate, 27(22), 8563-8577.

Also, this reviewer felt that the background/literature review part could be enhanced by citing some similar studies on using spatial information for model calibration, for example, those below:

Ruiz-Pérez, G., Koch, J., Manfreda, S., Caylor, K., and Francés, F.: Calibration of a parsimonious distributed ecohydrological daily model in a data-scarce basin by exclusively using the spatio-temporal variation of NDVI, Hydrol. Earth Syst. Sci., 21, 6235–6251, https://doi.org/10.5194/hess-21-6235-2017, 2017.

Su, Z., Zeng, Y., Romano, N., Manfreda, S., Francés, F., Ben Dor, E., ... Mannaerts, C. (2020). An integrative information aqueduct to close the gaps between satellite observation of water cycle and local sustainable management of water resources. Water, 12(5), 1-36. [1495]. https://doi.org/10.3390/w12051495

3. One major concern of this reviewer is that the use of various products for model calibration are not necessarily consistent. At least, the consistency issue should be checked and discussed before their use here. Sometimes, certain bias-correction might be needed to make various products consistent, before using them with the multi-criteria calibration approach. This reviewer also noticed that the author discussed a bit this in the discussion. Nevertheless, it is not fully clear how the inconsistency between different products will impact the output of the multi-criteria calibration.

Minor comments

Please see attached.

This work studied the effect of vegetation on global hydrological simulation and on the partitioning of total water storage (TWS) variations. The findings revealed the importance of deeper moisture storages and soil moisture-vegetation interactions when understanding TWS variations. I just have a few comments that may improve this manuscript, before it can be accepted for publication in the high-quality journal.

Specific comments:

1. Line 148: Eq.2 should be Eq.3.
2. Line 230: typo, ‘Therefor’.
3. Line 410: Fig.2 should be Fig.3.
4. Figure 4: please add the label of y axis (numbers of grids?).
5. Line 488: please check the number ‘69%’, I noticed that this number is 61% in Figure 8.

Comments for hess-2021-394 “The importance of vegetation to understand terrestrial water storage variations” by Trautmann et al.

The authors compare a model experiment with vegetation parameters varying in space and time to a baseline experiment in which all parameters are calibrated as static, globally uniform values to investigate the effect of vegetation-dependent parameterizations of key hydrological processes on TWS partitioning. I think this work is valuable and of interest to the greater community, and the manuscript is generally well written. The paper can be accepted after addressing my following concerns.

Major Concerns:

1. This paper mainly focus on investigating the impact of “space and time varying vegetation parameters” on defining infiltration, root water uptake and transpiration processes and decomposing the TWS. It sounds strange to use the words such as “importance of vegetation”, “including vegetation”, “including vegetation characteristics”, “including vegetation data” and “contribution of vegetation”. These words cannot reflect well the research objective, and I think the impact of vegetation/vegetation characteristics/ vegetation data is also embedded within the baseline experiment using the static and globally uniform parameter values.
2. The authors calibrate and validate the model performance almost for the same period (01/2002-12/2014 vs. 03/2000-12/2014). From the view of traditional calibration and validation procedure, it will be better to use part of the observations (e.g. period of 2002-2008) to calibrate the model, and then using the remaining observations (e.g. period of 2008-2014) to validate the model.
3. For the regional analysis of model performance, the authors derive 5 hydroclimatic regions by performing a cluster analysis, but it sounds strange to treat almost the whole China and Europe as the same group, i.e., the moderate mid latitudes (Temperate). This is contrast to the common sense. The authors are thus suggested to use a better classification such as the Köppen–Geiger climate classification.

Minor Comments:

1. The title of Section 2 is wrong and the structure of this section can be improved. E.g., “1.1 Methods” should be “2 Methods”, “1.2 Overview” should be “2.1 Overview”. In addition, it’s suggested to modifying “2.2.1” to “2.3”.
2. It’s suggested to modifying all the tables to the standard table format.
3. Line 410: “Fig. 2” should be “Fig. 3”
4. The number in Fig.3 is difficult to read.