Review of 'In-situ measurements of wall moisture in a historic building in response to the installation of an impermeable floor'

Overall assessment:

Most queries raised from the previous manuscript have been addressed, well justified and discussed providing appropriate references. These have been carefully considered and the content of the article has been accordingly modified and improved. The clarity of the manuscript has also been improved.

However, please consider that there are still some important queries to address that could further improve the discussion and outcomes of this case study.

1-Regarding to the following response made by the authors: ‘Response – The conclusion of the paper is that the capillary forces are not causing moisture flow because the capillaries/pores are not connected. Hence any properties to describe flow through porous materials that is based on a continuum will not be applicable. So, properties such as the open porosity, pore size distribution, tortuosity, capillarity coefficient will not be helpful for understanding the flow. They were also not measured, and it would be difficult to know how to measure them to represent the whole wall and the composite nature of the various layers. For these reasons, we chose in-situ monitoring to gain information about the wall as a system, rather than measuring the individual elements/materials forming the wall, which we could not do.’

Building materials such as the described in this case study (limestone, sandstone, lime, cement, should have certain degree of open porosity, capillary pores and connected porosity. It would be very rare having not connectivity at all since these are sedimentary type materials which pores and connectivity may vary but is not usually low, as it could be for volcanic or other igneous rocks with vitreous matrix with closed porosity and low or not connected pores.

Therefore, instead of saying that the capillaries/pores are not connected...which is a strong statement considering that porosity has not been measured, it is suggested to correct it. This has been better explained by the authors in the conclusions’ section when talking about why the in-situ measurements of wall moisture contradicted predictions based on a theoretical model of capillary rise for an idealised wall:

i.e. ‘This is because (better to say ‘might be’) the heterogeneous fabric of the rubble-fill wall (better to say ‘seems to’ contain) contained a discontinuous pore network and therefore restricted capillary flow and capillary rise within the wall.

It is also recommended to suggest this possibility as one of the causes that contradicts the prediction by the model rather than state that this is the cause for this.

2- Regarding the comment on porosity measurements in the same ‘Response’ (paragraph above):

There could be some possibilities to infer the porosity and related properties such as water absorption by capillarity or ‘moisture flow’. Although not specifically to measure porosity or capillary pores on site, indirect correlations of the wall (as a system, not for the individual layers) could be made from onsite measurements of non-destructive and portable techniques, such as ultrasonic velocity waves (Vp), which speed is related to porosity and density of materials. Vp waves change when materials are wet or dry, so, measurements taken in different weather conditions/ seasons and several measurements per point could be taken to map the obtained values measured on selected areas of the wall. In this way heterogeneities could be plotted for having a better understanding of the physical properties and hygric behaviours of the building materials.

Other non-destructive onsite measurements to support this could be air permeability, sponge, or Karsten pipe tests.

3- Regarding former Figure 2:

Despite the picture of former Figure 2 was blurry and there is no replacement photo, it would be better to show it instead of being removed.

A better description of building materials and vapour-proof barrier have been provided by the authors showing a new picture of the wall with the stratigraphy of the building materials, the core and the different vertical layers (render and coating).

Despite being blurry, former figure 2 was providing a lot of information since this was showing the internal face of the wall, dowel and moisture probe access holes, the state of conservation of the wall (which showed signs of rising damp…) and the polyethylene sheet that was installed on the floor.

4- Regarding the following Response made by the authors: ‘Response-The local climate was measured with a weather station. The results are typical of the seasonal temperate climate in Wales and the UK. It would be possible to compare these values to long-term averages (LTAs) but the results would not contribute to the results or conclusions of the case study. Fill has been changed to made ground, to align with the termed used by the British Geological Survey. A citation has been added. In terms of the geology, we could include general information about open porosity and capillarity rates, but from my experience these will not be enough to reliably model the flow through these materials. Also, the bedrock geology was not encountered during excavation, so the moisture supply and flow will be from the near-surface topsoil layers (and made ground). Not only do we not have that information, it would not give us any further insight into the results than was provided by the measurements (e.g. see the approach used in the Smethurst citation).

Please, consider to comment or discuss the influence of the presence of organic topsoil layers and 'made ground'.. mixed with clays and the thickness of these layers in the ground underneath the building compared to other soil types. These should condition the results of this case study considering that clays have low permeability, low infiltration rates, displaying poor drainage and impeding water flow compared to sands or other soil types in which the outcomes would have been different.

This should be mentioned or discuss somehow in the results and discussion section since it is also related to soil moisture deficit (SMD) and soil moisture content and the water transfer to building materials.

5- Regarding the following Response made by the authors: ‘Response – We could include information on the long-term average (LTA) values for this site but that is not the focus of the paper. Nor is the paper about climate change or future implications. The weather data are there to test the Hall & Hoff model and the influence of sealing the floor. Other implications would not be well-served by this case study. However, to allow comparison with long term (+10 years) weather data, citations for two sites in southern England have been added for readers interested in this comparison.’

Please, just to clarify that the query raised about making a better discussion of results and the influence of climate and the weather during the monitoring period was not about discussing climate change or future implications. This was about discussing the results from the model and the influence of sealing the floor under the environmental conditions registered by the weather station during the monitoring period, e.g. the influence or correlations between the environmental data recorded (i.e. solar radiation (kWm-2), air temperature (⁰C), humidity (%rainfall (mm/tip), wind speed (ms-1), wind direction (⁰) and the results of wetting or water content of the wall and the implications on the water absorption and evaporating drying processes of the ground and walls in this case study.

6-Regarding the following question and the response by the authors:

What can be the causes of discrepancy between model and measurements? Is there something that the model is missing? ‘Response – In this instance the model is testing whether moisture movement is due to capillary flow of water from the ground. The results suggest that this capillary flow is not occurring. This is due to the heterogeneous composition of the wall, with pores that are not connected. If the material porosity was fully defined, more complex models of liquid and vapour water flow would also simulate capillary flow (if the pores were assumed to be continuous). But the in-situ wall moisture data and inspection of the wall composition shows that this is unlikely to be the case.’

A better discussion and conclusion are still needed when talking about contradictions from model and measurements outcomes. It should be good to mention that more monitoring studies should be needed to have other outcomes and more representative conclusions based on monitoring: (i) different building materials (eg. bricks, stone ashlars, etc versus this case study of a complex composite based on mortar and stone rubble-filled core bedded in cement render, etc) and (ii) the ground soils underneath the building (eg. sands, gravels, etc. versus clay soils), as well as (iii) other climate conditions.

Moisture content of building and soil materials depends on their physical properties (eg. different permeabilities, pore network (including heterogeneities, fissures and cracks) and capillary coefficients or water absorption rates through capillary pores. This case study shows a nice example of a three-year monitoring measurements under specific environmental, geologic and building materials conditions and physical properties encountered onsite. It shouldn’t be used to generalize the behaviour of moisture levels in external walls of all historic buildings in response to seasonal potential evaporation rates and the influence by the installation of impermeable ground bearing slabs in all cases since the outcomes may change depending on the aforementioned factors.

Additionally, it should be discussed that the model was designed to be applied to predict rising damp and moisture movement within a porous masonry wall without finishes. Therefore, the values for input parameters for the model such as sorptivity that were used by the authors should be reconsidered since according to them:

‘The sorptivity (S) and moisture content of the wetted part of the wall (θw) were not measured, but were assumed to be 1.0 mm.min-1/2 and 0.2 respectively, as used by Hall & Hoff (2007).’

Should these values being assumed to have lower levels than those used by Hall & Hoff (2007) and more according to the system with the described characteristics which also include finishes such as render and coating? Please, consider this together with the comments previously made on the porosity, capillarity, permeability, of both building materials and the ground composed by clay soils mixtures, and discuss if these could also have had an influence in the discrepancy between model and measurements.

7-Figure 8 shows that the intervention (installation of polyethylene sheet on the floor) was done in mid-September 2019, in the middle of a period of high Soil Moisture Deficit (mm) (according to figure 8) and a very low soil moisture content (according to figure 7).

Could this have influenced the results obtained from the measurements of moisture content in the timber dowels (rubble wall) which does not show changes before and after installation?

It would be good to mention that the outcomes of this research are specific to this case study with this type of underground and building materials. Therefore, it shouldn’t be used to generalise the behaviour of moisture levels in other climates, geological settlements and different types of building materials.

8- Regarding the following questions and corresponding response: ‘where is the water from the wall coming from’? and (according to Figure 9) Why is there more moisture content at the base of the wall compared to higher height? Response - We do not have a clear answer to this question as we did not design the experiment to explore wall moisture distribution, but instead to explore changes in wall moisture in response to the intervention. In our experience, the moisture content of walls in historic buildings is usually greater at the base than further up (unless there is a roofing, guttering or plumbing defect). At Court House there may have been some water uptake at the base of the wall from the adjacent soil during wet weather, but because of the composite construction and lack of capillary pathways its ability to rise was limited. Alternatively, the moisture distribution in the wall may reflect moisture equilibrium with the room by a process of diffusion rather than capillary rise (which was shown to be not occurring).

Please, provide a better evidence, justification, correlation and discussion of results on the statement about the lack of capillary pores and limited capillary rise due to the composite construction, and the presence of water in the walls considering monitored data from the weather station (eg, rain fall), picture of the indoor wall in former Figure 2 (eg. showing signs of decay by rising damp), sorptivity properties of building materials (eg. Gummerson, Hall & Hoff, 1980), soil moisture content and deficit (Figs 7 and 8) at the time of the intervention, etc.