the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Nov 2024
A listening experiment exploring the relationship between noise annoyance and sound quality metrics for airborne energy systems
Abstract. The present study investigates the relationship between sound quality metrics (SQMs) and noise annoyance for airborne wind energy systems (AWESs) reported in a listening experiment. A convenience sample of 75 participants rated their annoyance on the International Commission on Biological Effects of Noise (ICBEN) scale in response to nine recordings from in-field measurements of two different fixed-wing and one soft-wing ground-generation AWES. All recordings were normalized to have the same A-weighted equivalent sound pressure level. The acoustical analyses showed that the fixed-wing kites presented a more tonal and narrowband sound signature than the soft-wing kite. Linear-mixed effects models indicated that sharpness was the only SQM predicting participants’ annoyance ratings and that the fixed-wing kites were rated as more annoying than the soft-wing kite. In addition, the effect of some SQMs on annoyance depended on participant characteristics, with loudness having a weaker impact on annoyance for participants familiar with AWESs and tonality having a weaker effect on annoyance for older participants. However, these moderation effects could be random due to the non-probability sampling used.
- Preprint
(1315 KB) - Metadata XML
-
Supplement
(657 KB) - BibTeX
- EndNote
Status: open (until 21 Dec 2024)
-
RC1: 'Comment on wes-2024-125', Anonymous Referee #1, 12 Dec 2024
reply
General comments
The present study is interesting and innovative as it gives insight into the properties of a new and developing technology. It applies knowledge about a more established form of wind energy, namely that sound emissions are a central issue for residents, to a context where conflictual cases can yet be prevented. Analyzing the psychoacoustic metrics of different form factors is important in order to understand how design choices affect the experience.
The article is very well written, methodology and results are presented clearly. Where it falls short of its potential yet is in the discussion of what the study means for AWE. Both in the introduction as well as when discussing the results there is little reference to other studies with similar approaches. As there is not a lot of research on AWE specifically, a comparison to the psychoacoustic metrics of other, to some degree similar technologies would help a reader gauge the significance of the results. As it is, a reader without specific psychoacoustic background knowledge would have a difficult time judging whether the results are surprising or expected. Furthermore, it could be clarified what the results imply for the design of AWESs. Do they suggest improvements to be made, for example?
Specific comments
- Section 2.1: How comparable are the three recordings per AWES to each other? Are the recordings selected to cover varying situations or are they supposed to be consistent? The spectrograms in section 3.1 reveal remarkable differences for kite C, whereas the recordings of the other two kites seem fairly homogenous.
- 84: Please add some explanation why only recordings of the reel-out phase were chosen. Are there differing acoustic properties of the phases? Is the reel-out phase more relevant? Are they similar enough that the results of one should translate to the other?
- 256-257: Please elaborate on why the circular flight pattern results in a bump above 1 kHz. How does this explanation relate to the one from l.274-275, stating the bump is a result of rigid materials?
- Section 3.2: Is there a criterium applied to judge which differences are meaningful? Regarding the loudness of kite C, for example, the text says that C3 exhibits higher values than C1. Looking at Figure 6 this is true regarding the maximum values, though not with respect to the distribution or median values. If the statement is based on the 5th percentile what is the threshold of a meaningful difference?
- 329: Please add some words of guidance to Figure 8, help the reader interpret how strong the effect is.
- It was not the aim of this study to compare annoyance levels of AWESs to other emission sources. But can you provide some point of comparison to understand whether the annoyance levels are particularly high or low? At least, it might be worth giving a verbal orientation on how to interpret the annoyance values within the scale (l.314-315).
- You state that the moderation effects might be tied to the sample. There are more specific explanations possible. Familiarity with AWE at a technical university with ties to AWE development might imply a more positive attitude, hence affecting annoyance ratings. On the other hand, the sample is particularly young, raising questions about the effect of age on tonality. Given the age distribution, what age differences result in the effect of age on tonality?
Technical corrections
- 14 & 75: Mostly, in this paper the word “acoustic” is used. In these two instances it is instead “acoustical”. I do not think this is to indicate any semantic difference. Perhaps you might want to keep it consistent.
- 25-27: “during natural disasters“ sounds like the kites are in the air while there is a heavy storm. I assume, you mean something like “following / in the aftermath of natural disasters”?
- 27: I suggest “lower mass” instead of “mass savings”. Also, it should be stated what AWESs are compared to.
- 33: This is the first time the abbreviation AWES is used. It is introduced a few lines later on l.41.
- 55: I suggest “enhance” instead of “enhances” as it refers to multiple aspects.
- 155-156: I suggest “in the analyses this variable was not considered as a confounding factor”. Please clarify that this refers only to experiences with AWE sounds as familiarity is considered in later analyses.
- Figure 3: There is no indication what the color grading represents.
- 230-231 “Examining the spectrograms...”: It is suggested that this refers to spectrograms of other recordings than the ones displayed here. If so this should be clarified.
- 251-252: Is the order of SPL below 2 kHz correct? Judging from Figure 4 the levels of C1 look to be higher than those of C2.
- Description of Figure 6: Although 5th percentile seems to be a standard denotation, next to the explanation of the 25th and 75th percentiles I find it quite irritating as both naming conventions count from different directions. Indicating that the highest 5% are highlighted might avoid confusion.
- 305: There seems to be a mismatch between text and table regarding the %HA.
- Figure 7: In some boxes the central horizontal line is hard to see. I suggest the lines and frames of the boxes be black rather than match the color.
- 369-370: There is a contradiction to the results chapter. Here it says that the fixed-wing kites show higher loudness, whereas in the results the soft-wing kite had the highest loudness values.
Citation: https://doi.org/10.5194/wes-2024-125-RC1 -
RC2: 'Reply on RC1', Anonymous Referee #1, 12 Dec 2024
reply
To specifiy what I was getting at in specific comment 7: The question is if you see any merit to / evidence for these hypotheses.
Citation: https://doi.org/10.5194/wes-2024-125-RC2
Data sets
Respondents' noise annoyance ratings for airborne wind energy systems measured in a laboratory listening experiment Helena Schmidt, Renatto Yupa Villanueva, Daniele Ragni, Roberto Merino-Martinez, Piet van Gool, and Roland Schmehl https://data.4tu.nl/private_datasets/11wCOgnga-GfjzL4oufzdd1UPRPP2YVFVnjRaKcHJhE
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 108 | 25 | 7 | 140 | 30 | 3 | 3 |
- HTML: 108
- PDF: 25
- XML: 7
- Total: 140
- Supplement: 30
- BibTeX: 3
- EndNote: 3
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1