Characterizing Hot-Water Consumption at Household and End-Use Levels Based on Smart-Meter Data
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsComments for authors:
This paper analyses residential hot water consumption at the end use level based on a new data set. The data set extends beyond previous similar studies (5-minute resolution, 5 households with different characteristics, almost 400 days) and proposes a multi-step analysis to quantify per capita hot-water consumption, calculate hot water ratios and assess daily profiles. The paper is interesting for potential readers of this journal, although I have several comments that might help to improve the contribution.
General comments
Expectations. As the authors admit in the Conclusions section, there are limitations associated with the presented data set (small sample size, temporal resolution, monitoring limited to cold season, inability to separate hot vs cold water use for most showers, etc.). This does not necessarily detract interest from the paper, because lessons learned from these data / analyses could be useful to improve/expand the household sample. However, I think that the authors should be upfront about this from the beginning. In my opinion, the abstract generates expectations that are too high for this data set. For example, the abstract reads “recorded over nearly 400 days” and one may think that the 5 households were monitored during the same period (Table 1 and supplementary data show that this is not the case). Similarly, it says “over 40 end uses”, but in reality, there are six different types of end-uses (DW, KS, WM, S, BT, F). Also, the abstract refers to “minute-scale resolution”, but the resolution is 5 minutes, which is relatively coarse compared to typical disaggregation studies and so conditions the type of analysis that can be carried out. I suggest reviewing the abstract / Introduction to be more consistent with the data set.
Motivation / Application. The main motivation of the paper is to analyse (in order to better understand) end-use water consumption with special emphasis on residential hot water. In my opinion, data analysis is a first necessary step to then model a phenomenon or even forecast it. I understand that this paper focuses on the analysis part, but still, I think that some comment regarding the usefulness of these findings with view to their practical application (either around modelling, plumbing system design or energy saving strategies, to name a few) would be nice. There are several previous related works along these lines in journals of the field.
Materials and methods. Figure 1 presents the study layout, which involves data pre-processing (conversion and filtering) and a multi-step analysis (which is the main novelty of the paper). But what does “filtering” mean exactly? According to the description in page 5 (lines 175-179), it looks like all households were supposed to measure over 5 months (roughly 150 days), but due to transmission/communication issues, only 17-133 days (depending on the household) are considered to be properly monitored. I am guessing that this means focusing on days where data are available every 5 minutes for all fixtures within each household, and this implies a considerable reduction in the available time series (especially for H3, H4 and H5, <50% days). Then, in page 6 (lines 194-199) it says that these time series were “filtered” to exclude days without consumption. It looks like the supplementary material includes the “filtered” data (352 days in total). If “filter” is understood as simply selecting subsets of data, then each selection (first the monitored period, then periods with consumption) is part of the filtering process. In my opinion, the word “filtering” implies that there has been a greater level of manipulation to select the data, but this is really up to the authors. Also, have the authors considered the possibility of working through each time series to take advantage of a larger amount of data? Even though all measurements are limited to winter, averaging values per household (as done in Analysis III) with only 5 households (of different types) and different moments within a 5-month window (where seasonal tendencies might be captured) is far from ideal. Especially if weekdays/weekends are differentiated. Completing the data has its own risks, but I think that this discussing this possibility in the paper would be interesting.
Minor issues
Page 4, line 149-151. In these households only cold water is used in the dishwasher and washing machine. This is uncommon in some other places as mentioned by the authors later on (page 11, lines 402-406). This fact should be mentioned in page 4 as well.
Page 6, Eq 2. The subindices of Vtot refer to the household (h) and the fact that it is a daily volume (d). Since the notation for the rest of equations only uses as subindices the end-use (i or j, see following comment), the household (h) or the hour (u), it may be cleaner to avoid using “d” as a subindex. It could be a superindex (or simply be written as DVtot). If I am not mistaken, Vtot_{h,d} is not used again along the manuscript.
Page 6, Eq 3 and beyond. It looks like end-uses are referred to with index i for Eq (1) but with index j for Eq (3) and beyond. I think that the difference is that i refers to each end-use point (for example, several bathroom taps) and j refers to each type of end-use (BT), but I think a comment about this after Eqs. 3/4 would be nice for potential readers.
Page 7, line 242. The term “unknown” is sometimes foggy, so I suggest clarifying at this point that in this data set it corresponds to showers for 4 households. Looking at table 2, maybe unknown could be substituted by mixed (hot + cold)?
Page 7, line 255. It would be nice to indicate that HWR is the abbreviation for Hot-Water Ratio.
Eq. (8). Since the denominator includes an extra summation compared to the numerator, it would be nice to put the extra one (addition over hours) before the other two (over households and timesteps) to easy comparison with the numerator. The same applies to Eq. (9).
Table 3. Since DW and WM do not use hot water, maybe they could be removed for the right part of the table (hot water). Including the abbreviations as a table note (apart from at the end of the document) would be nice (the same applies to the rest of tables).
Author Response
Comment #1: This paper analyses residential hot water consumption at the end use level based on a new data set. The data set extends beyond previous similar studies (5-minute resolution, 5 households with different characteristics, almost 400 days) and proposes a multi-step analysis to quantify per capita hot-water consumption, calculate hot water ratios and assess daily profiles. The paper is interesting for potential readers of this journal, although I have several comments that might help to improve the contribution.
Response to Comment #1: The Authors sincerely thank Reviewer #1 for the constructive comments, and appreciate reviewer’s recognition of the value of the study. All suggestions kindly made by Reviewer #1 have been carefully considered, detailed responses to which are provided below. Please also note that all edits are highlighted in yellow in the revised manuscript.
Comment #2. Expectations. As the authors admit in the Conclusions section, there are limitations associated with the presented data set (small sample size, temporal resolution, monitoring limited to cold season, inability to separate hot vs cold water use for most showers, etc.). This does not necessarily detract interest from the paper, because lessons learned from these data / analyses could be useful to improve/expand the household sample. However, I think that the authors should be upfront about this from the beginning. In my opinion, the abstract generates expectations that are too high for this data set. For example, the abstract reads “recorded over nearly 400 days” and one may think that the 5 households were monitored during the same period (Table 1 and supplementary data show that this is not the case). Similarly, it says “over 40 end uses”, but in reality, there are six different types of end-uses (DW, KS, WM, S, BT, F). Also, the abstract refers to “minute-scale resolution”, but the resolution is 5 minutes, which is relatively coarse compared to typical disaggregation studies and so conditions the type of analysis that can be carried out. I suggest reviewing the abstract / Introduction to be more consistent with the data set.
Response to Comment #2: We acknowledge that the Abstract and Introduction may have created unintended expectations due to the use of some ambiguous phrasing. In response to this, both sections have been revised to clearly emphasize the following points:
(1) although a total of 40 domestic fixtures were monitored, these fall into six main end-use categories (i.e., kitchen sink, dishwasher, washing machine, shower, bathroom taps, toilet flusher); (2) monitoring was conducted at a 5-minute resolution; and (3) the availability of valid data varied across households, ranging from approximately two weeks (17 days) to over four months (133 days). All revisions made in response to comment #2 are highlighted in yellow in the revised Abstract and Introduction.
Comment #3. Motivation / Application. The main motivation of the paper is to analyse (in order to better understand) end-use water consumption with special emphasis on residential hot water. In my opinion, data analysis is a first necessary step to then model a phenomenon or even forecast it. I understand that this paper focuses on the analysis part, but still, I think that some comment regarding the usefulness of these findings with view to their practical application (either around modelling, plumbing system design or energy saving strategies, to name a few) would be nice. There are several previous related works along these lines in journals of the field.
Response to Comment #3: We agree with the Reviewer #1 that data analysis is a fundamental first step toward the final objectives, e.g. modeling and forecasting of water demand. Having access to such data, or understanding the related characteristics, could lays the foundations for a wide range of applications, not only in modeling and forecasting hot-water consumption, but also in residential energy management and in the definition of strategies for water and energy conservation. In light of the reviewer’s suggestion, the Introduction has been expanded to further reinforce the motivation beyond the research by including a brief discussion about the main potential applications resulting from the availability of residential hot-water consumption databases (lines 61–68 in the revised manuscript).
“Nevertheless, several examples demonstrating the value of having detailed information on the magnitude and profiles of residential hot-water consumption—obtained through water-consumption monitoring and data analysis—are available in the scientific literature. These examples highlight the critical role such data play in supporting a variety of applications, including hot-water demand modeling [20,22–25], the optimal design and sizing of domestic plumbing systems or their components [18,26,27], and the formulation of effective strategies for water and energy conservation in residential contexts [28–30].”
In greater detail, the following newly added references, which complement the work, have been included in the References section of the revised manuscript.
- Binks, A.N.; Kenway, S.J.; Lant, P.A.; Head, B.W. Understanding Australian household water‑related energy use and identifying physical and human characteristics of major end uses. Clean. Prod. 2016, 135, 892–906. https://doi.org/10.1016/j.jclepro.2016.06.091.
- Blokker, M.; Agudelo-Vera, C.; Moerman, A.; van Thienen, P.; Pieterse-Quirijns, I. Review of Applications for SIMDEUM, a Stochastic Drinking Water Demand Model with a Small Temporal and Spatial Scale. Drinking Water Eng. Sci. 2017, 10, 1–12. https://doi.org/10.5194/dwes-10-1-2017.
- Canale, L.; Cholewa, T.; Ficco, G.; Siuta‑Olcha, A.; Di Pietra, B.; Kołodziej, P.; Dell’Isola, M. The role of individual metering in reducing domestic hot water consumption in residential buildings: A long‑term evaluation. Build. Eng. 2023, 73, 106734. https://doi.org/10.1016/j.jobe.2023.106734
- Chen, L.; Rosenquist, G.; Walker, I.; Meyers, B.; Wyatt, R. Calculating Average Hot Water Mixes of Residential Plumbing Fittings: Using the ANSI 301-2019 Hot Water Draw Model and National Residential Data to Estimate Hot Water Use in Showerheads and Lavatory Faucets; Lawrence Berkeley National Laboratory: Berkeley, CA, USA, 2020; Report No. LBNL-2001352
- Verhaert, I.; Bleys, B.; Binnemans, S.; Janssen, E. A Methodology to Design Domestic Hot Water Production Systems Based on Tap Patterns. Proceedings of the 12th REHVA World Congress CLIMA 2016, Aalborg, Denmark, 22–25 May 2016; vol. 3, pp. 1–10.
Comment #4. Materials and methods. Figure 1 presents the study layout, which involves data pre-processing (conversion and filtering) and a multi-step analysis (which is the main novelty of the paper). But what does “filtering” mean exactly? According to the description in page 5 (lines 175-179), it looks like all households were supposed to measure over 5 months (roughly 150 days), but due to transmission/communication issues, only 17-133 days (depending on the household) are considered to be properly monitored. I am guessing that this means focusing on days where data are available every 5 minutes for all fixtures within each household, and this implies a considerable reduction in the available time series (especially for H3, H4 and H5, <50% days). Then, in page 6 (lines 194-199) it says that these time series were “filtered” to exclude days without consumption. It looks like the supplementary material includes the “filtered” data (352 days in total). If “filter” is understood as simply selecting subsets of data, then each selection (first the monitored period, then periods with consumption) is part of the filtering process. In my opinion, the word “filtering” implies that there has been a greater level of manipulation to select the data, but this is really up to the authors. Also, have the authors considered the possibility of working through each time series to take advantage of a larger amount of data? Even though all measurements are limited to winter, averaging values per household (as done in Analysis III) with only 5 households (of different types) and different moments within a 5-month window (where seasonal tendencies might be captured) is far from ideal. Especially if weekdays/weekends are differentiated. Completing the data has its own risks, but I think that this discussing this possibility in the paper would be interesting.
Response to Comment #4: We acknowledge that the term “filtering” could be misleading, as it may suggest that the data were further processed or altered (e.g., using algorithms such as low-pass filters) to address specific issues in the dataset. However, no such processing or manipulation was performed in this study. Instead, the dataset was reduced solely by excluding days affected by data logging or transmission failures. To avoid ambiguity, the term “filtering” has been replaced with “cleaning,” which more accurately describes the procedure applied to reduce the risk of data misinterpretation. Specifically, the locution “data cleaning” includes two steps: (1) exclusion of days when data from one or more fixtures were missing due to logging or transmission errors; and (2) exclusion of days during which householders were absent, resulting in no water consumption. These steps have been clearly reported in the revised manuscript (lines 208–213), along with the corresponding results (lines 228–236).
“Subsequently, the obtained end-use water consumption dataset was cleaned (i.e., reduced) to exclude: (1) the days during which hot- or cold-water consumption data related to at least one domestic fixture were not available due to issues in data logging, wireless transmission, or Wi-Fi communication; and (2) the days during which the monitored households were not occupied, thus potentially leading to a misestimation of the hot- and cold-water consumption parameter values.”
“Overall, as indicated in Table 1, the above data cleaning process led to a considerable reduction in the number of days with available data. On the one hand, due to issues in wireless transmission or Wi-Fi communication, data availability ranges from a minimum of 17 days (H3) to a maximum of 133 days (H2), with a total of 394 days of available data (i.e., an average of about 79 days per household). On the other hand, householders’ absence resulted in a further reduction in the number of days exploitable in the analysis, which decreased from 394 to 352. Therefore, at the end of the data pre-processing phase, an average of about 70 days with water consumption per household are available, ranging from a minimum of 17 (H3) to a maximum of 111 (H2).”
We also acknowledge that analyzing individual fixture time series could retain a larger dataset, and we thank Reviewer #1 for this kind suggestion. To better point out the reason lying behind our choice, we clarified in the revised manuscript that the decision to exclude all days lacking data from one or more domestic fixtures—despite the resulting reduction in dataset size—was intentionally made to ensure a transparent and consistent characterization of domestic water consumption. In fact, retaining only periods with full data (i.e., with information of water consumption available across all fixtures) allows avoiding scenarios in which, for example, water consumption is recorded from a fixture (e.g., bathroom tap) while the status of other fixtures (e.g., toilet flushers) is unknown. Although the data cleaning process resulted in a smaller dataset, the remaining data are still well-suited to the analyses presented in this manuscript. These analyses require data collected simultaneously from all domestic fixtures within a given household in order to: (1) quantify the aggregate daily per capita hot-water consumption; and (2) compare hot-water consumption and related variables (i.e., hot-water ratio) across different end-use categories within the same household. This point is discussed at lines 213–218 of the revised manuscript.
“[…] days with missing data from any domestic fixture were excluded from the analysis. Although this approach may reduce the dataset size, it prevents the inclusion of partial domestic water-consumption records that could bias aggregated profiles at the household scale or compromise the reliability of comparisons of hot-water consumption data across different end-use categories within a given household.”
Comment #5. Page 4, line 149-151. In these households only cold water is used in the dishwasher and washing machine. This is uncommon in some other places as mentioned by the authors later on (page 11, lines 402-406). This fact should be mentioned in page 4 as well.
Response to Comment #5: We thank Reviewer #1 for this insightful suggestion. In response, the manuscript has been revised to explicitly note that the use of electronic appliances (DW, WM) supplied only by the cold-water line is typical in European countries, while being less common in other contexts (lines 163–169).
“Water-consumption monitoring was carried out by installing smart meters at each individual end-use point. Specifically, a smart meter was installed on the single cold-water line supplying each toilet flusher (F) and electronic appliances (DW and WM), the latter typically being connected only to the cold-water line in European countries, unlike in other contexts [49]. Conversely, two separate meters were installed on the cold- and the hot-water line supplying the other end-use categories (KS, S, and BT).”
Comment #6. Page 6, Eq 2. The subindices of Vtot refer to the household (h) and the fact that it is a daily volume (d). Since the notation for the rest of equations only uses as subindices the end-use (i or j, see following comment), the household (h) or the hour (u), it may be cleaner to avoid using “d” as a subindex. It could be a superindex (or simply be written as DVtot). If I am not mistaken, Vtot_{h,d} is not used again along the manuscript.
Response to Comment #6: Thank you for this suggestion. In response to the comment, the daily overall consumption of the -th household has been renamed as , and all references to this variable have been updated accordingly throughout the manuscript.
Comment #7. Page 6, Eq 3 and beyond. It looks like end-uses are referred to with index i for Eq (1) but with index j for Eq (3) and beyond. I think that the difference is that i refers to each end-use point (for example, several bathroom taps) and j refers to each type of end-use (BT), but I think a comment about this after Eqs. 3/4 would be nice for potential readers.
Response to Comment #7: In light of the above comment, a clarification was added before Equation (1) (lines 195–199 of the revised manuscript), explicitly stating that index ? refers to the smart meter associated with a specific domestic fixture and water line (hot or cold).
“The cumulative volumetric data V_{i,h} recorded over time by each smart meter i (i.e., in relation to a given domestic fixture and a given water line) of each monitored household h (i=1,...,n_{h}, being n_{h} the total number of smart meters installed in the h-th household considered, and h=1,...,H) were preliminarily converted into water consumption values Q_{i,h}(t) in accordance with Equation (1).”
In addition, to avoid any ambiguities, when introducing Equation (3) in the revised manuscript (lines 252–259), it was further clarified that index ? refers to the end-use categories considered in this study—namely KS, DW, WM, S, BT, and F—and can accordingly range from 1 to 6.
“In the present study, daily per capita hot-water consumption was first evaluated analyzed at both the end-use and aggregate levels, as shown in Equations (3) and (4), respectively […] where Q_{j,h}^tot is the hot-water consumption (L) of the j-th end-use category in the h-th monitored household of the sample, e.g., kitchen sink of household H3 (j=1,...,J, with J=6 for all households considered in this study); […]”
Comment #8. Page 7, line 242. The term “unknown” is sometimes foggy, so I suggest clarifying at this point that in this data set it corresponds to showers for 4 households. Looking at table 2, maybe unknown could be substituted by mixed (hot + cold)?
Response to Comment #8: Based on the above suggestion, when introducing the term “unknown water,” it was clearly specified that it refers to cases where both the cold- and the hot-water line are measured by a single meter, as for the showers in households H1–H4 (lines 265–270 of the revised manuscript).
“In addition to hot-water consumption, the following two metrics were also evaluated at both aggregate and end-use levels, for comparative purposes: […] (ii) daily per capita consumption related to water of unknown type […] corresponding to domestic fixtures with a single smart meter monitoring both cold- and hot-water lines, such as showers in households H1–H4.”
Moreover, for the sake of clarity, while retaining the term “unknown” in the graphical elements, the captions of Table 2 and Figures 2–5 were revised to explicitly state that this label refers to a mix of hot and cold water monitored by a single smart meter installed on the corresponding fixture.
Comment #9. Page 7, line 255. It would be nice to indicate that HWR is the abbreviation for Hot-Water Ratio.
Response to Comment #9: We acknowledge that the acronym HWR was not previously defined in the manuscript. Accordingly, in the revised version, it is introduced immediately after the definition of the hot-water ratio (line 277).
Comment #10. Eq. (8). Since the denominator includes an extra summation compared to the numerator, it would be nice to put the extra one (addition over hours) before the other two (over households and timesteps) to easy comparison with the numerator. The same applies to Eq. (9).
Response to Comment #10: Based on the above suggestion, the denominators of Equations (8) and (9) have been revised by moving the additional summation before the two summations that also appear in the numerator. The updated equations can be found at lines 314–315 of the revised manuscript.
Comment #11. Table 3. Since DW and WM do not use hot water, maybe they could be removed for the right part of the table (hot water). Including the abbreviations as a table note (apart from at the end of the document) would be nice (the same applies to the rest of tables).
Response to Comment #11: Based on Reviewer #1’s suggestion, the columns corresponding to end-use categories that do not include hot-water consumption (i.e., DW, WM, and F) have been removed from the right part of Table 3 which includes daily per capita hot-water consumption. Additionally, explanatory notes linking each end-use category to its corresponding abbreviation have been added as footnotes below each graphical element where these abbreviations appear, i.e., Table 1, Table 3, Table 4, Figure 2.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe manuscript explains the analysis about hot and cold water consumption using smart meter data. The manuscript is well written and describes the important pieces of scientific work. Here are some suggestions for further improvement.
What are the criteria for selection of households? Most residents are aged above 40 years except two people (one 30 years and one child). Why were people with 10-40 not considered? Line 335-136: Explain these factors have impact on water consumption rato…….
Line 146: More than 40 residential end uses of water>>>> not clear
Lines 162-163: Provide the layout of the installed smart meters at the end use level.
Line 318: “72.8 of which related to cold water, 27.8 to hot water, and 6.1 to unknown water” what are unit of 72.8, 27.8, 6.1????
Line 178: Was data collection period 394 days? Or data collection period is parallel for each household (H1-H5)
Figure 2a Has only column and repetition, whereas same information is also available Table 2. Similar in Figure 2b and Table 3???
Author Response
Comment #1. The manuscript explains the analysis about hot and cold water consumption using smart meter data. The manuscript is well written and describes the important pieces of scientific work. Here are some suggestions for further improvement.
Response to Comment #1: The Authors wish to thank Reviewer #2 for the positive comments and the constructive feedback on the manuscript. Our responses to the reviewer’s suggestions are provided in detail below, following each specific comment. Please also note that all edits are highlighted in yellow in the revised manuscript.
Comment #2. What are the criteria for selection of households? Most residents are aged above 40 years except two people (one 30 years and one child). Why were people with 10-40 not considered? Line 135-136: Explain these factors have impact on water consumption ratio.
Response to Comment #2: We acknowledge that the inhabitants of the five households included in this study may not constitute a representative population sample, as some demographic profiles (e.g., individuals aged 10 to 30 years) are not included. This limitation is due to the fact that only five households volunteered to participate in the study. Prior to data collection, an analysis was conducted to assess people’s willingness to take part, but only five households gave their consent due to the intrusiveness of the smart-metering systems (including one or two meters per domestic fixture). These considerations have been added to the revised manuscript, following the description of the household characteristics (lines 151–155):
“It is worth noting that the composition of the household sample resulted from preliminary analyses involving a broader group of households to assess their willingness to participate in the study. Ultimately, due to concerns about the intrusiveness of the monitoring system—which required the installation of at least one smart meter at each domestic fixture—only the five aforementioned households consented to participate.”
Furthermore, the above limitation regarding the representativeness of the household sample has also been stated in the Conclusions section, when discussing study limitations (lines 615–618 of the revised manuscript):
“Despite the remarkable outcomes of this study, some limitations have to be highlighted. First, the sample size is relatively small, comprising only five households located within the same geographic area. Moreover, the household composition may not be fully representative of the general population, as certain demographic groups (individuals aged 10 to 30 years) are not included […]”
Comment #3. Line 146: More than 40 residential end uses of water>>>> not clear
Response to Comment #3: We acknowledge that the locution “more than 40 residential end uses of water” may be misleading, as it could be interpreted that over 40 different types of possible residential water uses were investigated. In response to this comment—and in line with Reviewer #1’s suggestions—we have revised the terminology to improve clarity. Specifically, the term “end use” has been replaced with “fixture” when referring to individual domestic devices, whereas it was maintained when referring to the six categories of indoor water consumption analyzed in this study (i.e., DW, KS, WM, S, BT, F). Accordingly, the original sentence was rephrased as follows (lines 159–162 of the revised manuscript):
“Overall, as reported in Table 1, the monitored household sample includes more than 40 domestic fixtures, ranging from a minimum of 4 fixtures per end-use category (in the case of DW) up to a maximum of 15 (in the case of BT).”
Comment #4. Lines 162-163: Provide the layout of the installed smart meters at the end use level.
Response to Comment #4: We acknowledge that the original manuscript lacked details regarding the layout of the installed smart meters, specifically in terms of their location and quantity. In response to the above suggestion, the Materials section—within the End-use data collection subsection—has been expanded to include a more detailed description of both the location and the number of smart meters installed (lines 163–169 and 179–184 of the revised manuscript).
“Water-consumption monitoring was carried out in each household by installing smart meters on the supply line linking each domestic fixture to the domestic plumbing system. Specifically, a smart meter was installed on the single cold-water line supplying each toilet flusher (F) and electronic appliance (DW and WM), the latter typically being connected only to the cold-water line in European countries, unlike in other contexts [49]. Conversely, two separate meters were installed upstream the cold- and the hot-water supply point of each fixture of the other end-use categories (KS, S, and BT).”
“From an operational standpoint, the smart monitoring system consisted of mechanical meters—where varied across households depending on the total number of fixture supply points—each paired with an optical reader and a radio transmitter utilizing the Wireless M-Bus protocol. The system was configured to record cumulative water volumes (with 1-L accuracy) at a specified temporal resolution and to transmit the logged data daily to a digital platform via the household’s Wi-Fi connection.”
Comment #5. Line 318: “72.8 of which related to cold water, 27.8 to hot water, and 6.1 to unknown water” what are unit of 72.8, 27.8, 6.1????
Response to Comment #5: Based on the above comment, the sentence was rephrased to clarify that the unit of measurement is liters per person per day (L/person/day), and to explicitly state that 72.8 L/person/day, 27.8 L/person/day, and 6.1 L/person/day correspond to the cold-water, hot-water, and unknown-water shares, respectively, of the total daily per capita water consumption of 106.7 L/person/day. Specifically, the sentence was revised as follows (lines 347–350 of the revised manuscript):
“An average of 106.7 L/person/day are consumed in households H1–H5. Specifically, 72.8 L/person/day are related to cold-water consumption, 27.8 L/person/day to hot-water consumption, and 6.1 L/person/day to unknown-water consumption (i.e., undifferentiated shower consumption). ”
Comment #6. Line 178: Was data collection period 394 days? Or data collection period is parallel for each household (H1-H5).
Response to Comment #6: In light of the concern raised by Reviewer #2—and considering a similar comment made by Reviewer #1—the aforementioned part was revised to clarify that the overall data collection period covered approximately five months. However, the periods with available (and non-zero) water consumption data varied across households, ranging from a minimum of approximately two weeks (household H3) to a maximum of nearly four months (household H2) (lines 228–236 of the revised manuscript).
“Overall, as indicated in Table 1, the above data cleaning process led to a consider-able reduction in the number of days with available data. On the one hand, due to issues in wireless transmission or Wi-Fi communication, data availability ranges from a minimum of 17 days (H3) to a maximum of 133 days (H2), with a total of 394 days of available data (i.e., an average of about 79 days per household). On the other hand, householders’ absence resulted in a further reduction in the number of days with available data, which decreased from 394 to 352. Therefore, at the end of the data pre-processing phase, an average of about 70 days with water consumption per household are available, ranging from a minimum of 17 (H3) to a maximum of 111 (H2).”
Comment #7. Figure 2a Has only column and repetition, whereas same information is also available Table 2. Similar in Figure 2b and Table 3???
Response to Comment #7: We acknowledge that Figure 2 of the original manuscript did not provide additional or new information, as panels (a) and (b) repeated the values reported in Table 2 and Table 3, respectively. In light of Reviewer #2’s observation, Figure 2 has been removed from the revised manuscript.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have answered all my concerns. I look forward to seeing this research article in print.
