Coupling wildfire spread simulations and connectivity analysis for hazard assessment: a case study in Serra da Cabreira, Portugal

Sá, Ana C. L.; Aparicio, Bruno; Benali, Akli; Bruni, Chiara; Salis, Michele; Silva, Fábio; Marta-Almeida, Martinho; Pereira, Susana; Rocha, Alfredo; Pereira, José

doi:https://doi.org/10.5194/nhess-22-3917-2022

Articles | Volume 22, issue 12

https://doi.org/10.5194/nhess-22-3917-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/nhess-22-3917-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 22, issue 12

Research article

|

09 Dec 2022

Research article |

| 09 Dec 2022

Coupling wildfire spread simulations and connectivity analysis for hazard assessment: a case study in Serra da Cabreira, Portugal

Ana C. L. Sá, Bruno Aparicio, Akli Benali, Chiara Bruni, Michele Salis, Fábio Silva, Martinho Marta-Almeida, Susana Pereira, Alfredo Rocha, and José Pereira

Download

Final revised paper (published on 09 Dec 2022)
Preprint (discussion started on 02 May 2022)

Interactive discussion

Status: closed

RC1:
'Comment on nhess-2022-107', Anonymous Referee #1, 21 Jun 2022

Assessment and recommendation

The manuscript presents a model-based study of the influence of vegetation connectivity on the intensity and spread of wildfires in Portugal. Overall, it is an interesting study that fits within the scope of NHESS. The text is well written, the methods are described in detail, and conclusions are supported from the presented results. I suggest to accept the manuscript for publication, subject to the following minor remarks.

Minor comments

1. Many sentences in the manuscript are very long, which makes them hard to follow. Please, review the entire manuscript text and try to break long sentences into smaller pieces.

2. L132-134: According to the authors, WRF output was provided at 3h intervals and the weather variables used were averaged over the 12-20h time window. How can this affect the credibility of their analysis, particularly with respect to extreme fire weather? Are the authors certain that the 3h WRF output and subsequently the averaging allows for pinpointing extreme fire weather? A comment on that, also within the manuscript, would be highly appreciated.

Technical remarks

L22: improve --> improving

L28: increasing in suppression efforts --> increased suppression efforts

L34: rathern than

L91: remove where (?)

L130: Weather Research and Forecasting model

L465: Table C4 --> Table C1

L466: run --> ran

L511: This value or These values

L409: on the landscape

L418: future instead of next future

Citation: https://doi.org/10.5194/nhess-2022-107-RC1
- CC1:
  'Reply on RC1', Ana C. L. Sá, 22 Jun 2022
  
  We would like to thank the reviewer's comments and the time spent for improving this study. Regarding the two minor comments:
  1. We will revise the document and avoid long sentences to make the manuscript easier to follow.
  2.
  We think that the reviewer's questions are relevant, and we need to clarify them. First, because of the lack of information for individual fire events regarding the active fire spread periods and on the corresponding driving weather conditions, we calibrated the fire model using the historical fire regime instead of individual fire-based calibration. Considering this data limitation, we run simulations with constant weather for spreading durations lower than 24h. Therefore, based on expert knowledge, in the study area the main active fire spread frequently occurs during the 12-20h time window, when higher temperature and wind speed, and lower humidity conditions prevail. Second, the calibrated fire model based on this daily subset of the weather data was evaluated for the 9 largest fires (1000 to 4000 ha), using the extreme (95^th percentile) weather condition calculated from that weather subset. The quality assessment was satisfactory (please see Appendix E, line 510) and thus we do not suspect that we are failing to estimate the fire spread under extreme weather conditions. Lastly, even if we fail to capture those extreme weather conditions because of the averaging, our calibration approach compensates that by tunning the fire spread durations to reproduce the historical fire regime. We will include the next sentence in the body text (Line 138):
  “This time window choice is because fire simulations are run for spread durations smaller than 24h with constant weather conditions, and thus it was necessary to exclude milder weather conditions that typically occur during the evening and morning periods. The eventual averaging effects of the extreme weather conditions is compensated by tunning the duration of fire spread (Sect. 2.4.4).”
  All the technical remarks were corrected and will be shown in the final revised version of this study.
  Best regards
  ana sá
  
  Citation: https://doi.org/10.5194/nhess-2022-107-CC1
  - AC1: 'Reply on RC2', Ana C. L. Sá, 04 Oct 2022
    
    We would like to thank the reviewer´s suggestion that is relevant to compare fireline intensity (as measured by the CFL estimate) with the forest fire loss (FFL) dataset. Here we include a pdf file with a bried analysis, that may be included as Supplement in the final version of the manuscript.
    Best regards,
    Ana Sá
    
    Citation: https://doi.org/10.5194/nhess-2022-107-AC1
- AC1: 'Reply on RC2', Ana C. L. Sá, 04 Oct 2022
  
  We would like to thank the reviewer´s suggestion that is relevant to compare fireline intensity (as measured by the CFL estimate) with the forest fire loss (FFL) dataset. Here we include a pdf file with a bried analysis, that may be included as Supplement in the final version of the manuscript.
  Best regards,
  Ana Sá
  
  Citation: https://doi.org/10.5194/nhess-2022-107-AC1
RC2:
'Comment on nhess-2022-107', Anonymous Referee #2, 01 Oct 2022

This is a well develop research methodology and should be consider for publication in NHESS. To further improve:

My only suggestion would be to include a small analysis of validation (accuracy assessment) of your predicted fire size or intensity/ hazard against forest loss due to fire dataset for Serra da Cabreira in 2018 (https://glad.umd.edu/dataset/Fire_GFL/).

Citation: https://doi.org/10.5194/nhess-2022-107-RC2
- AC1: 'Reply on RC2', Ana C. L. Sá, 04 Oct 2022
  
  We would like to thank the reviewer´s suggestion that is relevant to compare fireline intensity (as measured by the CFL estimate) with the forest fire loss (FFL) dataset. Here we include a pdf file with a bried analysis, that may be included as Supplement in the final version of the manuscript.
  Best regards,
  Ana Sá
  
  Citation: https://doi.org/10.5194/nhess-2022-107-AC1

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to minor revisions (review by editor) (06 Nov 2022) by Vassiliki Kotroni

AR by Ana C. L. Sá on behalf of the Authors (11 Nov 2022) Author's response Author's tracked changes Manuscript

ED: Publish as is (15 Nov 2022) by Vassiliki Kotroni

AR by Ana C. L. Sá on behalf of the Authors (17 Nov 2022)

Short summary

Assessing landscape wildfire connectivity supported by wildfire spread simulations can improve fire hazard assessment and fuel management plans. Weather severity determines the degree of fuel patch connectivity and thus the potential to spread large and intense wildfires. Mapping highly connected patches in the landscape highlights patch candidates for prior fuel treatments, which ultimately will contribute to creating fire-resilient Mediterranean landscapes.