Dear Authors,

I am very interested in the preprint entitled “Global satellite gravity data products for prompt detection of short-term Mass Change (MC)”. This work presents, for the first time in a systematic manner, a global dataset of Line-of-sight Gravity Differences (LGD) derived directly from GRACE-FO Level-1B observations, together with a detailed description of the generation methodology, data structure, and potential applications. I believe this study is of great scientific significance and practical value.

First, as an along-track instantaneous gravity observable, LGD breaks through the temporal resolution limitation of conventional GRACE/GRACE-FO monthly solutions (L2 or Mascon products). It captures surface mass variations at intervals of 5–6 days or even shorter, filling a long-standing gap in sub-monthly hydrological monitoring. More importantly, it paves the way for transitioning satellite gravimetry from pure scientific research towards operational applications, such as near-real-time early warning of flash droughts and rapid floods. The case studies presented (e.g., flash drought in the southeastern United States, sub-monthly changes in the Amazon and South Sudan) clearly demonstrate the unique advantage of the LGD dataset in resolving the onset, peak, and termination phases of hydrological extremes.

Second, since the transfer function method was introduced by Ghobadi-Far et al. (2018), LGD has gained broad interest in fields such as seismology, hydrology, and drought studies. However, the computation of LGD involves complex procedures including dynamic orbit integration, accelerometer calibration, and spectral transformation, which creates a high technical barrier for many potential users. The authors have done an outstanding job in processing the data with great care, making the entire workflow publicly available as a ready-to-use dataset. By downloading the published data, I found the data quality to be very high, with very few outliers, reflecting the authors’ rigorous data processing. Furthermore, the authors provide two additional functional LGD time series (L2-synthetic LGD and climatology-based LGD), which involve substantial extra work and offer users a rich set of product choices.

The paper also provides a clear and detailed description of the entire processing chain from L1B to LGD (reference orbit integration, range-rate residuals, range-acceleration residuals, transfer function, and band partitioning). The open-source code for forward-modeling LGD from L2 monthly solutions is also shared, ensuring good reproducibility.

While I strongly support this work, I would like to offer one minor technical suggestion. In Figure 5c, the LRI-LGD curves show noticeable high-frequency spikes (e.g., near the equator and between 30°N–0°N). Based on my understanding, these spikes are mainly caused by imperfect handling of phase jumps induced by attitude-control thruster firings in the JPL LRI1B RL04 product. If alternative LRI1B products from other institutions—such as HUST V01 (Huazhong University of Science and Technology), SYSU S10 (Sun Yat-sen University), or AEI V54 (Albert Einstein Institute)—are used, these thruster-induced artifacts can be further suppressed, thereby improving the signal-to-noise ratio of LRI-LGD in the frequency band around 20 mHz, albeit to a limited extent. The authors may consider evaluating the impact of different LRI1B products in a future update.

In addition, the current dataset only covers the GRACE-FO period (June 2018 – December 2024). Many long-term hydrological studies require a longer time series. I hope that the authors will process and release the LGD results for the GRACE mission (e.g., 2002–2017) as soon as possible. I believe this is not only my personal wish but also a common expectation among many potential users.

Overall, this dataset represents a major advance in satellite gravimetry, providing a valuable new tool for sub-monthly hydrological monitoring and Earth system science. Thank you again for the authors’ excellent work.

Sincerely,

HongLi