The manuscript has a clear methodological motivation: standard land DA is often column-wise and does not directly improve the spatial organization of soil-moisture anomalies, which matters for land–atmosphere coupling. The curvelet-based framework is also conceptually well aligned with multiscale spatial structure, and the manuscript explains this reasonably clearly with the reconstruction examples in Figure 1 and the spectral-space assimilation formulation. The paper also includes both global diagnostics and independent station-based checks, which is important. Overall, the study is interesting, potentially useful, and relevant to hydrological modeling and land data assimilation and worth publishing HESS. I will put some minor comments:

1) Please define more explicitly what makes the method “EnKF-like” rather than a standard EnKF. At present, the wording may confuse readers expecting a full ensemble state update in model space.

2) The description of the ensemble perturbation strategy for estimating BB and RR is too brief. Please state the ensemble size, perturbation distributions, and whether any localization or inflation is used.

3) Figures 3–5 are informative, but the captions could better state which fields are instantaneous and which are accumulated or averaged, to avoid ambiguity. 

4) Please report the number of U.S. ISMN stations actually used after screening, as this is less explicit than for the CMA dataset. 

5)  The manuscript should discuss whether the method is expected to remain effective at higher spatial resolution, where fine-scale heterogeneity and representativeness errors become more important.

6) Some language needs polishing for grammar and precision. For example, there are occasional awkward phrases such as “he assimilation experiment” and repeated wording around “spatial structure.”

7) 

I strongly recommend that the authors cite the following relevant study, which appears closely related to the manuscript’s topic of data assimilation for hydrologic and land-surface interactions: "Assimilation of Sentinel-Based Leaf Area Index for Modeling Surface-Ground Water Interactions in Irrigation Districts."