The article discusses the negative impacts of prolonged soil moisture deficit, such as reduced vegetation growth, nutrient and water cycling, and increased fire and insect outbreaks. The paper notes that early warning through monitoring and forecasting can help mitigate these effects. The study presents a new version of Global BROOK90, which integrates seasonal meteorological forecasts to predict soil moisture drought on a local scale. The study finds that the hindcasted soil moisture fits well with the reference model runs only within the first month of lead time and that the probabilistic forecast is reasonable only for short-to-medium range lead times. Additionally, within the drought period, the ECMWF hindcast forcing results show overestimation of the soil moisture for most of the catchment, indicating an earlier end of the drought period. 

I would like to commend the authors for their interesting work and analysis. The manuscript in its current state needs to be re-written. I am including a few points below:

• Expand any new abbeviation once for example, SPI in line 55 is not expanded earlier in the tex, (I am guessing it is standardized precipitation index)
• line 146: "As the framework aims to reach/cater-to local scale and professional and non-professional users"
• line 168: "test case"
• line 156: "above mentioned"  and "yield 1.5 minutes" 
• Also, I have usually seen medium-range forecast associated to 3 to10 days forecast  but the study goes out to months of forecast.  

The above are a few of the examples I noticed in the manuscript and for this I encourge the authors to revise the language/terminologies in the manuscript.

This study uses a new version of Global BROOK90 model to generate and evaluate soil moisture forecasts at a local scale. It focuses on a selected number of small European catchments over a short analysis period to examine the skill of the soil moisture forecasts. The methods of this study are similar to past studies that have used ECMWF or other seasonal climate forecast systems to drive a hydrologic model and generate and evaluate skill of hydrologic forecasts. The main difference is that the authors claim that the Global BROOK90 framework may be computationally inexpensive and relies on open-access input datasets, presumably making it applicable anywhere in the world.

Other than that, the results of this study lack robustness and novelty as it is focused on a small number of catchments in Europe and over a short period of analysis. It is well established that the skill of soil moisture forecasts is higher over the first 1-3 months (often partly due to the skill coming from the initial hydrologic state) and decreases as the lead time increases. Therefore, the results of this study do not add any novelty beyond what is already well known.

If the study had focused on a longer period of analysis, evaluated multiple basins across the globe (which would have helped highlight the value of this framework for global applications), and evaluated the skill relative to relevant benchmarks, the study could have been more appropriate for publication. A suitable benchmark in this case, given the focus on local soil moisture forecasts, could have been statistically interpolated soil moisture forecasts from global models.

The question remains whether the Global BROOK90 framework  adds to soil moisture forecast skill at the local scale beyond what could be attained by simply statistically interpolating soil moisture forecasts from a global soil moisture forecast system.