Spatial, Temporal and Electrical Characteristics of Lightning in Reported Lightning-Initiated Wildfire Events
, J. Brent Wachter 3, Christopher R. Hain 1 and Jordan R. Bell 4Round 1
Reviewer 1 Report
Lines 59-61. These two sentences need to be re-worded for a more descriptive opening for the paragraph.
Lines 95-96 & 104-105. Both of these paragraph's opening remarks say the same thing
Line 135. Change "250m of from the fire start location" to "250m from the fire start location".
Section starting on line 117: Somewhere it should be clarified whether the study used "stroke" data as opposed to "flash" data from the NLDN. I'm assuming you're getting stroke data from NLDN since this is the default. It would also be appropriate to define the difference between the two..i.e. multiple strokes make an individual flash. If this is the case, you should replace all instances of the word "flash" with "stroke". If the NLDN data is flash, a discussion of how NLDN classifies strokes into flashes would be appropriate.
Line 170: "These test" with "These tests".
Line 170-175 This is the first time that the peak current component of the stroke has been discussed. Are you using the value of the peak current as an ignition indicator or just the polarity?
Figure 2 & 3: A legend for the colored bars is needed.
Figure 6: The blue diamonds and triangles are hard to tell apart....maybe color them differently??
Author Response
The authors would like to thank the reviewer for their time in reviewing this manuscript. The reviewer’s comments have help fine tune the manuscript and we appreciate the attention to detail that was provided by the reviewer.
1) The authors have reworked these two sentences and combined them into one statement. Additionally, the authors have eliminated the words “This study” from the following line to mitigate confusion to the reader and not make them think the present study by the authors was responsible for the flash density metrics. The current lines between 59 and 63 read:
“Ground flash density is a quantity used in lightning efficiency metrics used by the United States Forest Service (USFS) to assess lightning-initiated wildfire potential [22]. The flash densities used in the USFS lightning ignition efficiency product were derived from the amount of electrical current and duration of continuing current needed to ignite fuels that are commonly found in nature.”
2) Yes those lines do state the same thing, with one caveat: the duration of the period of study. The vision is to first generally introduce the wildfire database, and then provide context for the area and time period chosen in the second paragraph.The authors have removed the 4 km2 disclaimer from the opening sentence on line 95. These lines now read:
Lines 99-100: “Reported lightning-initiated wildfires were extracted from the USFS wildfire database [26] which characterizes wildfire events from multiple federal and local agencies from 1992 to 2015. “
108-109 “All lightning-initiated wildfires which grew to at least 4 km2 between 2012 and 2015 were used for analysis.”
3) The authors have removed the word “of” from line 137. Flash data was preferred to stroke data for this analysis because the flash data provide multiplicity information that can be directly compared to previous studies. Furthermore, the flash data allowed us to determine the percentage of flashes that contained a single stroke or multiple strokes. Ultimately the stroke data could be used for this analysis. We did a quick analysis of the stroke information and the same number of fires were identified using the fire radius method as the sample using the flash information. The authors have added in a subsection 4.4 to explain how strokes are combined into flashes within the Vaisala dataset, and to illustrate that the stroke level data produces similar results to that of the flash data, with very minor offset differences between the two datasets. This new paragraph can be found between lines 419 and 433 of the revised manuscript. The text reads:
“In the present study, NLDN flash data were preferred over the NLDN stroke data because of the ability to assess flash multiplicity and compare with previous work [e.g., 10-12]. However, another dataset that is available to the community is NLDN stroke level data, which are the locations of individual ground connections that make up a flash. The stroke-level data are clustered to produce the NLDN flash dataset using a spatial criteria of 10 km and a maximum temporal criteria of 1s [43]. Temporal intervals between strokes cannot exceed 500 ms, and the maximum number of strokes per flash is 15 [43]. Multiple NLDN strokes can make up an individual NLDN flash, and the location, timing, and peak amplitude of the first stroke is recorded in the NLDN flash dataset. The number of strokes per flash is reported as multiplicity in the flash dataset.
The same set of lightning-initiated wildfires was examined using the stroke level data and the fire radius method. The reanalysis indicates that the same number of fires were identified, with very similar spatial offsets from the fire start location. Therefore, NLDN stroke level data can be used for analysis, which is an important consideration for any future real-time lightning initiation wildfire product given some operational users of lightning data only receive the stroke level data.”
4) This has been corrected as suggested.
5) The authors have included the words peak current on line 173 to lead into the discussion about the statistical tests in the lines that follow. The authors wanted to show that there was not statistical independence between fire starters and non-fire starters using polarity and multiplicity.
6) The authors assume that the reviewer is asking for text on the frequency bar graphs that states which search method was used. The authors have added the words “10 km fixed method” and “Fire radius method” to the panels in Figures 2 and 3. If this is not what the reviewer was seeking, let us know so we can adjust the figures as necessary.
7) The diamonds, triangles, and plus signs that indicate IC, -CG, and +CG flash information have been increased in size, and color coded for each flash type (IC, purple, -CG, blue, +CG red). These designations are included in the figure caption for Figure 6.
Reviewer 2 Report
The manuscript "Spatial, temporal, and electrical characteristics of lightning in reported lightning-initiated wildfire events" by Schultz et al. examines the association between lightning flashes and the occurrence of lightning-initiated wildfires. The authors compare two methods, the fixed and fire radius method, and conclude that the fire radius approach is more promising to locate the time and location of lightning-initiated fire starts.
It is well written, well organized and thoroughly covers the subject. Figures and Tables are easy to understand.
Minor comment:
Introduction: page 2 lines 60-61. Sentence starting with "Lightning efficiency metrics used for ..." is missing a verb.
Author Response
Thank you for your time reviewing the article and your kind words. One of the recommendations from reviewer 1 was to reconstruct the first two sentences of the paragraph you've correctly identified as missing a word. These lines can be found between 59 and 63 of the revised manuscript.
