Review of 'Normal background levels of air and surface mould reserve in UK residential building stock'

Dear Ulf,

Many thanks for reviewing our paper and apologies for our delayed response.

Please see below our responses to your individual comments:

• The authors have chosen to use a single method for determination of mould, the quantification of the N-acetylhexoaminidase (NAHA) activity supported by microscopy of the filter used for the NAHA activity determination. The limitation is that the use of NAHA activity determination is not a worldwide standard. Furthermore, the results are impossible to compare to results obtained from more widespread methods being agar based contact plates or air-samplers of the impact type using Petri dishes. Still the present manuscript is an important step forward.

Many thanks for pointing out the importance of our work! We are indeed convinced that this study, from a methodological point of view, can provide a much-needed reference point for establishing benchmarks indicative of normal background levels of indoor mould, using multiple methods. We agree with you that NAHA based method might not be as widespread, however it is a standard method recognised by many organisations as we outline within the paper. Furthermore we have shown a mid-level of correlation between these readings and culture based methods in a previous study (Aktaş, Y.D., Ioannou, I., Altamirano, H., Reeslev, M., May, N., D’Ayala, D., Canales, M. (2018) Surface and passive/active air mould sampling: A testing exercise in a North London housing estate, Science of the Total Environment 643, pp. 1631-1643, doi: 10.1016/j.scitotenv.2018.06.311). While the reasons of this limited correlation is beyond the remit of this review, the fact that culture based methods do have the crucial downside to be able to capture only the culturable microbial load is a factor, i.e. while it is a more widespread method, it has certainly other disadvantages - like all mould quantification methods for that matter.

In any case we have edited the title of our manuscript to highlight that this is a preliminary work, which requires further elaboration.

• General: When citations are (author, year) the list of references should be alphabetical and not in order of appearance as now.
• Page 2: The citations ‘Statens Byggeforskningsinstitut 2003 a&b’ should rather be (Valbjørn, 2003; Koch and Nielsen, 2003), respectively. The references are:
  • Valbjørn O. 2003. Undersøgelse og vurdering af fugt og skimmelsvampe i bygninger (By og Byg Anvisning 204). Hørsholm: Statens Bygge­forskningsinstitut. Localised 3rd of September 2019 at http://anvisninger.dk/204
  • Koch AP, Nielsen PA. 2003. Renovering af byg­ninger med skimmelsvampevækst (By og Byg Anvisning 205). Hørsholm: Statens Bygge­forskningsinstitut. Localised 3rd of September 2019 at http://anvisninger.dk/205

Thank you for pointing this out. We have now fixed the citations, including the two reports who were initially reported as authored by Statens Byggeforskningsinstitut.

• Page 2, Section 2, first line: How were the 140 non-water damaged homes classified as such? Measurement of relative humidity? Humidity in construction parts?

The non-water damaged homes were classified as such as they did not have any indication of water induced damage, including but not limited to damp patches, spalling or discolouration which may be linked to moisture ingress. This has now been clarified in the paper.

• Page 3, Study site: Any reason for the distribution between type and age of buildings?

There is no any particular reason for the type-age distribution, as the buildings set we tested were made available to us by volunteers. This point has now been clarified.

• Page 4, last paragraph, line 2: bad reference to a figure(?).

The broken reference to Figure 1 has been fixed.

• Page 4, last paragraph, line 4: “17 properties in Denmark”? Not mentioned in Materials and Methods! New data, or already published data? If published, a citation to publication is needed.

The testing of 17 properties in Copenhagen have been carried out previously by Mycometer and is the basis of the Danish Building Research Institute guidelines. These have not been previously published, but the data have been provided for our use in this paper by Dr Morten Reeslev. These points have been made clearer now.

• Page 5, paragraph marked (4): it was not the Danish Building Research Institute (= Statens Byggeforskningsinstitut) that brought forward the limits 25 RFU and 450 RFU, respectively. It was the private company Mycometer A/S, as clearly presented in 2003 publication from Statens Byggeforskningsinstitut. You need to find the background information for the use of these RFU levels, as it is not clear why you choose to use 25 and 450. How is the distribution of all your RFU readings? Mean and median values?
• Page 6, first line below figure 2: “…1700 RFU are most likely indicative…”. It is a very fluffy argument, why not choose 1500, or 2000? How is the distribution of all your RFU readings? Mean and median values?

Our text does not suggest that the benchmarks were brought forward by the Danish Building Research Institute itself, but it was currently used by them. We have changed the wording slightly so that this is clearer.

There is no specific reason for our suggested benchmarks to be, say, 25 but not 20 and 1700 but not 1500 RFU, apart from the fact that we chose to work with a 98% threshold. One can choose to cut the distribution at 99%, surely, which could make a slightly different cut-off point. Benchmarking, as we report here, is not a rigid concept and is aimed to provide the much needed guidance and insight about the test results.

• Page 7, Section 4 (Discussion): is not really a discussion – and Section is “Results and Discussion” – Section 4 is rather some concluding remarks. Rephrase and integrate into Section 3 or 5.1.

We have now integrated Section 4 and 5 upon your suggestion.

• Page 8: It is recommended to show some statistical evidence for the RFU benchmarks presented.

We have added further statistical description for our data and developed a figure (Figure 1) to summarise these.