General comments

This manuscript investigates the potential of applying Seasonal Climate Outlooks to localized precipitation forecasting by compiling 0.5-month lead time NOAA CPC seasonal precipitation tercile probability forecasts for two study basins. Specifically, these forecasts are incorporated as conditioning information in a local-scale stochastic weather generator to produce precipitation ensembles. Through the evaluation, two non-parametric ensemble generation methods are found to be suitable for different seasonal regimes, and the predictive skill is shown to be associated with large-scale climate signals (i.e., ENSO).

Overall, the manuscript is of good quality, with a clear and easy-to-follow structure, and it addresses a scientifically promising question using an innovative approach. What is more important, the proposed framework and the associated findings are generalizable, and it provides a guidance for further exploiting seasonal precipitation forecasts at the local scale. In my opinion, the manuscript is suitable for publication following minor revisions. My concerns mainly relate to the clarity of the presentation and the way the results are presented, as detailed in the specific comments below.

Specific comments

1. The authors appear to overemphasize hydrological modeling, as references to it appear in the abstract, introduction, and conclusion. While precipitation forecasts are indeed important for hydrological modeling, the methods and results presented in this study do not directly involve any hydrological model. I therefore suggest either removing these references or adding a dedicated discussion section to assess the potential impact of such precipitation forecasts on hydrological modeling, which would help improve the overall logical flow of the manuscript. If such a discussion is added, I also believe it would be highly beneficial to include some additional commentary on the predictability of precipitation.
2. The figures in this manuscript warrant further revision, particularly regarding font size and color differentiation. For example, in Figure 1, the font size for the latitude/longitude labels and basin names is too small; in Figure 2, the text is so small that it is difficult to read; and in Figures 3, 5, and 6, the colors and marker styles lack sufficient contrast, making it hard to distinguish different elements.
3. The manuscript contains an excessive amount of numerical comparisons and descriptions, which would be more clearly presented in a table. This primarily concerns the section from Lines 346–354, as the current format makes it difficult to follow.
4. The description of the observed precipitation data is not sufficiently clear. I suggest that the authors provide additional details in Section 2.2, specifically indicating which stations’ precipitation observations were used to validate the forecasts. Additionally, the relative locations of the stations within each basin should be marked in a figure.
5. It is also recommended that the abstract include a brief description of the observational data used to assess the NOAA precipitation outlooks.

technical corrections

1. Line 245. There is a mistake in the equation.

The manuscripts evaluates NOAA CPC 3-month precipitation outlooks skills at 0.5-month lead for two basin in west-central Florida basins over 25 years. It then proposes two simple non-parametric methods, Proportional Tercile Sampling (PTS) and Dominant Tercile Sampling (DTS), to translate CPC tercile probabilities into basin-scale ensemble rainfall realizations using an existing stochastic rainfall simulation dataset. The resultsshow the strongest categorical and probabilistic skill during late fall and winter, particularly for above-normal precipitation forecast and in El Nino Years. A hybrid recommendation is suggested (DTS during high-skill seasons and PTS otherwise). Overall, the study offers a coherent and potentially transferable approach, but a few edits to terminology and methodology are needed.

Below are the several specific comments

1. Line 65: The basin location mentioned as ‘Southwest Florida’ conflicts with the title. Please use one term consistently.

2. The Equal Chances (EC) category is defined clearly; however, it is unclear how EC cases are handled in the categorical verification framework when building 2×2 tables (e.g., whether they are excluded or pooled into the “non-event” category). This should be clarified in the methodology.

3. FAR is labeled as “False Alarm Rate” in the figure caption, but Equation (10) and the following text define FAR as “False Alarm Ratio.” Consistent terminology for FAR is needed throughout.

4. Equation (15) requires correction.

5. Section 2.4 describes how ensemble rainfall forecasts are generated using tercile probabilities, but it is unclear how the method is applied when the CPC outlook is classified as Equal Chances. The handling of EC cases in the ensemble-generation step should be stated explicitly.

6. The manuscript references the wrong equation: it states that the “above-normal” pool is defined by Equation (6), but Equation (6) defines the below-normal (BN) category. Please correct this citation.

7. Please correct the panel label in the Figure 5 caption: the Alafia River Basin panel should be labeled “(a)” (not “(b)”).

8. In the section on performance of sampling schemes, qualitative comparisons should be complemented (or replaced) with quantitative improvements (e.g., the decrease in RMSE of one sampling method relative to another).

9. The typo in the data availability statement (“sis”) should be corrected.

Climate forecasts generated by global climate models (GCMs) are valuable for hydrological modelling. In this paper, the attention is paid to the NOAA’s seasonal precipitation forecasts. The skill at different lead times is evaluated. In general, the analysis is useful for the West-Central Florida.

There are four comments for further improvements of the paper.

1. Besides the NOAA’s forecasts, there are other sets of forecasts available in the NMME (). The authors are suggested to add more GCM forecasts and conduct a more comprehensive evaluation of GCM forecast skill. In the meantime, the multiple GCM forecasts under investigation can be summarized by a table.

1. In the evaluation of GCM forecasts, the authors are suggested to refer to Slater et al. (2019) that presented a comprehensive of the NMME forecasts across the continental USA.

1. Given the existence of spatiotemporal biases in GCM forecasts, forecast post-processing plays a key part in exploiting the potential raw GCM forecasts. The quantile mapping is probably the simplest yet effective method to use.

1. The Oceanic Niño Index (ONI) is mentioned in the paper. There are statistical forecasts produced from hydroclimatic teleconnections such as ONI. Is it possible to perform some comparisons of GCM forecasts with statistical forecasts?

References:

Kirtman, Ben P., et al. "The North American multimodel ensemble: phase-1 seasonal-to-interannual prediction; phase-2 toward developing intraseasonal prediction." Bulletin of the American Meteorological Society 95.4 (2014): 585-601.

Slater, Louise J., Gabriele Villarini, and Allen A. Bradley. "Evaluation of the skill of North-American Multi-Model Ensemble (NMME) Global Climate Models in predicting average and extreme precipitation and temperature over the continental USA." Climate dynamics 53.12 (2019): 7381-7396.

Zhao, T., Bennett, J. C., Wang, Q. J., Schepen, A., Wood, A. W., Robertson, D. E., & Ramos, M. H. (2017). How suitable is quantile mapping for postprocessing GCM precipitation forecasts?. Journal of Climate, 30(9), 3185-3196.