A 2D Real-Time Flood Forecast Framework Based on a Hybrid Historical and Synthetic Runoff Database
Round 1
Reviewer 1 Report
1. There is not any description regarding to the rainfall conditions(previous records and synthetic conditions respectively). Please show the spatial and temporal distributions of rainfall conditions on your watershed.
2. There is no clear description about spatial resolution of the FD/FV domain and roughness coefficients. It is also needed to describe your upstream and downstream boundary conditions with historic flood and synthetic flood conditions in 2D modeling respectively.
3. Detail explanations are needed regarding to the physical and statistical chararteristics of SD, REBD and HD database for the future application of real time flood forecast. Please show the some criteria for making decision on early flood warning.
4. For the real time application of the suggested model, it is important to show the total processing time(or lead time) from rainfall forecast to warning to the residents.
5. The study area is relatively small. What seem to be the constraints if this processes applied to large river basin in the future study ?
6. Please show some charateristics of your model comparing with data-driven model such as neural network or support vector machine in terms of modeling accuracy and executing time.
7. Some symbols in the figure 7 cannot identified properly.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Based on the work of Bhola in 2018, this manuscript proposes a runoff database that comprises both historical and synthetic runoff serials. The authors claim that the performance of 2D real-time flood forecast framework can be improved by implementing such hybrid database.
The manuscript itself is overall well prepared and easy to follow. However, my major concern is the novelty and contribution of this manuscript. If I understand it correctly, the framework in this manuscript is the same one in Bhoal et, al 2018, except for the REBD. I don’t think REBD can justify the publication in Water. Moreover, the figures look much like those in Bhoal et, al 2018 as well.
Other issues:
Figure 1. I think the middle section is the contribution of previous paper. Figure 5 can be removed. It difficult to get useful information. For real-time flood forecasting of such area, the lower boundary condition is also very important but is usually unknown. How you solve this problem. Do you consider the precipitation within the domain? For floods have high return period, the precipitation itself can be very large in the domain.Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 3 Report
This paper covers an important topic of realtime forecasting of flood extent in river channels. The idea of relying on a large database of past events seems sound and is helpful to speed things up (especially due to the high computational expense of 2D hydraulic models).
The paper is well written and I would recommend its publication after minor corrections/edits. I have one suggestion regarding the offline databases. The use of real event database (REBD) seems appropriate to me. However, in the methods section, I did not see any explanation as to why a synthetic event database is required as well. Maybe this is obvious to hydraulic modellers or was explained in the Bhola 2018 paper. However, to increase the appeal of this paper to wider audience, I would suggest providing a few lines of explanation in the methods section.
Minor point: The X axis of Figure 4 ends up with 0 in the right hand side. This is a bit confusing, please amend the X axis of Figure 4.
Author Response
please, see the attachment
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The original manuscript has been faithfully revised as indicated.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 2 Report
I'm fine with the authors' reponses. I only have following 2 concers:
(1) The authors have to specify the lower boundary they use for simulation/forecasting.
(2) I'm not talking about pluvial flooding, I'm talking about the runoff generated from catchment but eventually routed to river. From my experience, such runoff component can account for considerable proportion in the downstream flood. The authors better clarify this issue.
Author Response
Please see the attachment
Author Response File: Author Response.docx
