Preliminary Metrological Characterization of Low-Cost MEMS Inclinometer for Tree Stability Assessment: From Laboratory to Field

Overview

The title of manuscript is not correct in my opinion. The term ‘Static Tree Stability Monitoring’ is invalid and it is not used in literature in this field. As a term it incorrect as it should state ‘static pulling tests for tree stability assessment’ of something on those lines. The monitoring indicates that something is measured during longer period of time, and that is not the nature of this measurement as it is done mostly once and in rare occasions (such as trees that had been severed by construction works) it can lead to several measurements with larger time gap among them.

In my opinion there are some research gaps in inconsistencies that hamper the flow of thoughts on understanding for readers therefore major revisions of manuscript are suggested in order to improve the quality of this paper. It is unclear to me why a measurement sensor such as inclinometer was used at height of 1,3 and 2,7 m, when usually in pulling test measurement protocols this sensor is used to monitor the anchorage of root system during loading?

I see the application of this sensors in measurement of trees during pulling test to construct a bending curve from inclinometer data which may indicate the stability issues in stems (large strains on certain parts of tree) which is interesting application of this technique that is currently missing due to high price of reference precision sensor used in commercial products.

• We sincerely thank the Reviewer for the rigorous and technical critique, which has significantly contributed to improving both the clarity and methodological precision of the manuscript.

• Terminology and “Monitoring”. We fully accept the Reviewer’s observation. The term “monitoring” was indeed inappropriate to describe a single-event static pulling test. Accordingly, the title has been revised to “Preliminary Metrological Characterization of a Low-Cost MEMS Inclinometer for Tree Stability Assessment: From Laboratory to Field”. Throughout the manuscript, “monitoring” has been replaced with “assessment” or “testing” when referring to the experimental activities, and “static basal tilt” has been revised to “static inclination analysis”.

• Sensor placement (1.3 m and 2.7 m) and reference instrument. We clarify that the choice of installing the sensors at 1.3 m and 2.7 m was motivated by a metrological objective. While we acknowledge that standard diagnostic protocols (e.g., SIM) prescribe sensor placement at the root collar to isolate basal rotation, the primary aim of this study was the characterization of the MEMS prototype against a professional reference instrument. The larger inclination amplitudes observed at these heights facilitate a clearer comparison between the MEMS sensor and the reference inclinometer than the very small rotations typically recorded at the root collar. As requested in Comment 6, the reference instrument has now been explicitly specified in Section 2.3 as a Dynatim™ biaxial inclinometer (RINNTECH, Heidelberg, Germany), with a nominal resolution of 0.001°.

• Research gaps and bending curves. We particularly appreciate the Reviewer’s suggestion regarding bending curve analysis. We agree that this represents a promising perspective for low-cost sensing approaches. The deployment of multiple sensors along the stem explored here through measurements at two heights suggests the potential for future analyses of stem deformation patterns that are currently difficult to implement with professional equipment due to cost and logistical constraints. This perspective has been explicitly acknowledged in the Conclusion as a potential direction for future research.

Comment 1:

L46-47 second part of this sentence needs further explanation as it is not correct. This tests were never intended to be used for long term monitoring per se, so it is not appropriate to highlight this as disadvantage of this method.

We thank the Reviewer for this clarification. We have revised the sentence to avoid characterizing the pulling test as a long-term monitoring tool, which was indeed an inappropriate comparison. The revised text now correctly defines pulling tests as 'point-in-time stability assessments' and clarifies that the limitation for large-scale application is due to high instrumentation costs and logistical complexity. Furthermore, we have expanded the introduction to distinguish between these static assessments and the continuous monitoring of tree movements under wind loading, which represents a separate and complementary assessment approach

Comment 2:

L50 ‘static basal tilt’ is not correct term as this does not represent the static pulling test or static load test. The static nature of this test and something that different it from dynamic measurement is known pulling direction and continues measurement of steadily applied pulling force (which produces moment on tree) under appropriate threshold which ensures that trees are not damaged after the test. This paragraph needs further refinement as terms are not represented in correct way. This is repeated several times through the manuscript which indicates that is systematic error in description. Furthermore, the large emphasis is given to the use of this approach in continuous monitoring of trees which fails to point to actual use that they have as part of tree stability process and on large scale the tree risk assessment.

Comment 3:

L55 ‘Static and dynamic monitoring’ it is not correct to use this phrase

We thank the Reviewer for the terminology clarification. We agree that the term “monitoring” can be misleading when referring to static pulling tests, which are typically point-in-time assessments or repeated checks rather than continuous monitoring. In the revised manuscript, we therefore removed “monitoring” when referring to pulling-test applications (e.g., in the study aims we replaced “static inclination analysis monitoring” with “quasi-static inclination measurements”), and we also revised the keywords accordingly. The term “monitoring” is now reserved exclusively for continuous measurements under natural wind loading, which we explicitly describe as a separate and complementary approach.

Comment 4:

L77 too much and too different references to be included in this sentence

We agree with the Reviewer. The term “structural monitoring” was inappropriate in this context and overstated the scope of the cited literature. The sentence has been revised to more precisely refer to the growing importance of tree stability assessment in urban forestry contexts, in line with the referenced studies.

Comment 5:

L93 this should not be the separate aim of this study as it goes beyond what is measured

We agree with the Reviewer. The objective referring to long-term deployment was too broad with respect to the experimental scope of the study. The aims have been revised to focus on the metrological characterization and field agreement of the MEMS inclinometer, while considerations on scalability and long-term use are now explicitly framed as implications and perspectives discussed on the basis of the experimental results.

Comment 6:

L130-131 the model and manufacture of this commercial inclinometer should be stated for comparison purpose. From image it seems that this is sensor that is part of Rinntech measurement system Dynatim

The model and manufacturer of the reference inclinometer have now been explicitly specified in Section~2.3 as a Dynatim™ biaxial inclinometer (RINNTECH, Heidelberg, Germany).

Comment 7:

L135 ‘and strain’?

We thank the Reviewer for the comment. No strain measurements were performed in this study, as the objective was limited to the metrological characterization of inclination measurements. The reference instrument was therefore used exclusively as a high-precision inclinometer, and the text has been clarified accordingly.

Comment 8:

L157-158 The issue with this peak force is that it does not state did you use calculate it based on angle of pulling cable in which you need to calculate it by multiplying with cosines of rope angle during peak force to know the applied force

We thank the Reviewer for this valuable comment on force calculation. We fully acknowledge that standard pulling test protocols require correcting the measured cable force for the pulling angle (F_lateral = F_measured × cos(θ)) to accurately compute the applied moment at the base, ensuring tree safety and result comparability. In this study, the cable angle θ (~15–20° from terrain slope) was trigonometrically estimated during setup and remained stable across pulls, with peak forces maintained below 10% DBH-equivalent thresholds. However, as both MEMS and reference sensors were co-located and experienced identical moments, trigonometric decomposition was not required for the metrological comparison of inclinations. Force served solely to stabilize plateaus. To enhance clarity, we have added a note in Section 2.3: 'Cable angle effects on lateral force were accounted for in protocol design but omitted from analysis, as sensor agreement was evaluated under shared loading conditions

Comment 9:

averaging window of 20 sec how was it selected? you describe it in protocol on processing the field data measurements, but did you did and statistic computation to back it up? As it includes around 520 measurements in this timeframe based on 26hz sampling interval it seems as long time frame to stabilize the readings from sensor? What is your explanation for this? it seems unpractical to have such 20 s window gaps to use this reading in field (for tree stability measurement using pulling test procedure)

Comment 10:

In my view it is unnecessary to have both table 3 and figure 5 and 6 for results of laboratory testing of applicability of sensors. You have determined the best average moving window and additional figures can be part of supplementary material. The issue is that this needs to be explained (which is related to previous comment about 20 s window). As it stands the position of figure 5 and 6 is wrong as they represent the field measurements results that are presented up to line 240 in manuscript.

We thank the Reviewer for these important and closely related remarks.

The 20 s averaging window used for field data was not selected arbitrarily, but was informed by the laboratory calibration, where the effect of different averaging windows (1–120 s) on noise reduction and mean inclination was systematically evaluated under fully static conditions. These laboratory results showed a progressive stabilisation of the MEMS signal with increasing window length, without altering the mean inclination.

In the field, data were acquired during 3-min pull–hold plateaus, and a single centred 20 s moving average (≈520 samples at 26 Hz) was adopted as an operational compromise to stabilise plateau estimates while preserving the step-like transitions between successive loading phases.

In line with this clarification, and to improve conciseness, we reduced redundancy in the presentation of laboratory and field results by grouping related figures into composite panels and moving non-essential material to a more compact layout, as suggested by the Reviewer.

Comment 10:

Figure 7 and 8 are too big and use too much white space. Results are similar and comparable so there is no need to use two pages for this.

We thank the Reviewer for these important and closely related remarks.

The 20 s averaging window used for field data was not selected arbitrarily, but was informed by the laboratory calibration, where the effect of different averaging windows (1–120 s) on noise reduction and mean inclination was systematically evaluated under fully static conditions. These laboratory results showed a progressive stabilisation of the MEMS signal with increasing window length, without altering the mean inclination.

In the field, data were acquired during 3-min pull–hold plateaus, and a single centred 20 s moving average (≈520 samples at 26 Hz) was adopted as an operational compromise to stabilise plateau estimates while preserving the step-like transitions between successive loading phases. No additional window-length sensitivity analysis was performed on field data, as the objective was instrument-to-instrument agreement under a common processing pipeline rather than filter optimisation.

In line with this clarification, and to improve conciseness, we reduced redundancy in the presentation of laboratory and field results by grouping related figures into composite panels and moving non-essential material to a more compact layout, as suggested by the Reviewer.

Comment 11:

What is difference between field test A and B? it is used on same tree and figures 7 -10 show that you have reference sensor to compare the results to? it does not show what table 1 is displaying that there was difference as with and without reference high precision sensor (A and B)

We thank the Reviewer for highlighting this ambiguity. Test~A and Test~B were both performed on the same tree and with the same experimental setup, and in both cases the low-cost MEMS inclinometer was co-located with the reference high-precision inclinometer to enable direct instrument-to-instrument comparison. The two tests do not represent configurations with and without the reference sensor; rather, they correspond to pulling sessions conducted under different (progressively increasing) load levels. Their purpose was to evaluate the stability and robustness of the MEMS–reference agreement across loading conditions, rather than strict test repeatability. We have revised Table~1 and the corresponding text in the Methods and Results sections to clarify this point and to remove any possible misunderstanding.

Comment 12:

L281-287 this behavior that you explain in this paragraph is called hysteresis and it is known to influence pulling test results in certain conditions (cyclic loading or as in your case prolonged period under load)

We thank the Reviewer for this clarification. We agree that the time-dependent inclination response observed during the constant-load phases can be interpreted, from a biomechanical perspective, as a hysteretic or relaxation behaviour of the stem–root–soil system under sustained loading. In the present study, however, this effect was not analysed in terms of force–displacement loops or energy dissipation, but was considered only phenomenologically, as it was synchronously detected by both the MEMS and the reference inclinometer. 

Comment 13:

L374 It is commendable that you point to some of disadvantages of use of this sensors, but you fail to point the amount of data that is collected during measurement (through data logger). Processing, and previous storing, of this data could be demanding and can influence the scalability of inclination sensors.

We thank the Reviewer for this insightful comment. The sampling frequency of the MEMS sensor (26 Hz) is explicitly stated in the Methods, and in the present study data storage and processing did not represent a limitation, given the scope and scale of the experiments and the fact that the data‐logging server was configured to handle the expected data flow. In addition, in our intended operational use the inclinometer is not designed to record continuously at 26 Hz, but rather to acquire data only during limited measurement windows (e.g. once or twice per day, for durations comparable to those identified as optimal in the laboratory tests). This acquisition strategy substantially reduces the overall data volume and mitigates potential constraints on storage and processing, improving the scalability of inclination sensors for larger monitoring networks.

Comment 14:

L479 View project??

We thank the Reviewer for pointing this out. The text “View project” was an editorial artefact from the reference manager and has been removed.

Comment 15:

The literature can be further expanded as discussion part is not really abundant with other studies on this topic. I would move away stacking of references in brackets during citation (as seen in discussion)

We thank the Reviewer for this constructive remark. We agree that, in some parts of the Discussion, references were grouped at the end of sentences, which may reduce the clarity of how the present results relate to previous studies.

In the revised manuscript, we have therefore reworked selected paragraphs of the Discussion to better integrate the literature into the argumentative flow, explicitly linking individual studies to specific aspects of our findings (e.g., MEMS agreement under quasi-static loading, use of low-cost sensors in tree biomechanics, and comparison with reference instrumentation).

Overview

The presented research is interesting as it addresses the problem with maintaining urban forest without risk of human health and well-being (risk of falling trees). Currently, there are some methods to evaluate tree stability, but they are not easy to apply in some cases and are related to significant costs. So, an innovative approach has been tested and validated in both laboratory and field experiments, allowing for a more easy to perform long-term surveys on urban trees, at a lower price and with less effort.

The topic corresponds well to the Forests journal aim and scope, and the study presents some new insights in the field, overcoming the limitations found in previous studies, reported in the scientific literature. The methodological part is well explained, allowing for reproducing the experiments. All instruments used are mentioned and technical data were presented. Figures, photos and tables are of a high quality and give an added value of the manuscript. Conclusions are robust and reflect the main results of the study. The authors can add more interpretations here aiming to highlight their significance. 

All references are adequate to the studied area.

Your work is interesting and has a significant practical implications, as it addresses the problem of ensuring the safety of population in relation to the sustainability of urban trees.

The manuscript follows the Instructions for authors, all sections are presented, the methodology is well explained, the figures and photos are of a high quality, English is fine.

We thank the Reviewer for the positive feedback. Following the suggestion to highlight the significance of our results, we have expanded the Conclusions section. We added a discussion on how this low-cost technology can facilitate large-scale, urban tree assessment, moving beyond site-specific assessments to improve public safety management.

Comment 1:

1) authorship: The authors' affiliations should be numbered chronologically, so Francesca Giannetti shoul be marked as 2,3,4 and so on.

We thank the reviewer, author affiliations and their corresponding numbering have been rearranged chronologically throughout the manuscript.

Comment 2:

2) Line 128: Please, add the name of the authors before ...[32].

We thank the reviewer, the names of the authors have been added before the citation [32] at Line 128, as requested.

Comment 3:

3) Line 310: The subsection title could be removed

We thank the reviewer for this comment, we deleted the subsection title

Comment 4:

4) Line 411: Authors names should be abbreviated (initials only)

The author names have been abbreviated to initials in the Author Contributions section. For the two authors sharing the same initials (F.G.), the full name has been added in parentheses to ensure correct attribution.

Overview

I highly appreciate the scientific quality of the presented research. Its practical value is also interesting. The use of cheaper measuring devices compared to those currently in use is of great importance in view of the large number of measurements required recently due to social pressure and climate change.

My main question concerns the research methodology. As a standard practice, during tensile tests, the inclinometer is placed as low as possible on the tree, between the root runs. The maximum inclinometer readings should not exceed 0.250. As shown by the presented research, for small angles of deflection, the MEMS sensor gives greater inaccuracies. Suspending the sensors at a height of 1.3 m and 2.7 m resulted in the observed angles of deflection being obviously greater (and therefore the accuracy of the sensor greater), because not only the movement of the root ball but also the deflection of the trunk was recorded. Therefore, the presented results clearly demonstrate the consistency of the results obtained in the tested methods, but they do not fully comply with the currently applied SIM method standards for determining the risk of tree fall. This should be clearly stated in the paper.

We thank the reviewer for this important methodological comment. We fully agree that, according to standard diagnostic protocols for static traction tests (e.g., SIM), inclinometers are typically installed at the base of the collar to isolate basal rotation and compare it with pre-established safety thresholds (e.g., 0.25°).

However, as the reviewer rightly points out, our experimental setup was intentionally designed for metrological characterization rather than for standard biomechanical stability assessment. The choice of heights of 1.30 m and 2.70 m was dictated by two main objectives:

Sensor characterization: higher positions provide greater deflection angles, which allows us to better characterize the performance of the MEMS sensor.

Practicality: these heights are more representative of potential long-term urban assessment, where protecting equipment from vandalism or accidental impacts is a priority.

As requested, we have explicitly clarified this point in the revised manuscript (Section 2.3, lines [row: 153 - 157]), stating that, although this configuration is ideal for comparing instruments, it deviates from the strict SIM protocol for anchor assessment. We believe that this clarification better defines the scope of our work.

Comment 1:

I have no fundamental comments regarding the English language. Precise technical terminology has been used (laboratory calibration, setup geometry, averaging windows, relative errors, etc.), the sentences are well balanced, logical and clear, and the passive voice, typical of academic style, has been used correctly.

We thanks the Reviewer for the positive feedback regarding the language and the technical terminology used in the manuscript

Overview

The title of manuscript is not correct in my opinion. The term ‘Static Tree Stability Monitoring’ is invalid and it is not used in literature in this field. As a term it incorrect as it should state ‘static pulling tests for tree stability assessment’ of something on those lines. The monitoring indicates that something is measured during longer period of time, and that is not the nature of this measurement as it is done mostly once and in rare occasions (such as trees that had been severed by construction works) it can lead to several measurements with larger time gap among them.

In my opinion there are some research gaps in inconsistencies that hamper the flow of thoughts on understanding for readers therefore major revisions of manuscript are suggested in order to improve the quality of this paper. It is unclear to me why a measurement sensor such as inclinometer was used at height of 1,3 and 2,7 m, when usually in pulling test measurement protocols this sensor is used to monitor the anchorage of root system during loading?

I see the application of this sensors in measurement of trees during pulling test to construct a bending curve from inclinometer data which may indicate the stability issues in stems (large strains on certain parts of tree) which is interesting application of this technique that is currently missing due to high price of reference precision sensor used in commercial products.

• We sincerely thank the Reviewer for the rigorous and technical critique, which has significantly contributed to improving both the clarity and methodological precision of the manuscript.

• Terminology and “Monitoring”. We fully accept the Reviewer’s observation. The term “monitoring” was indeed inappropriate to describe a single-event static pulling test. Accordingly, the title has been revised to “Preliminary Metrological Characterization of a Low-Cost MEMS Inclinometer for Tree Stability Assessment: From Laboratory to Field”. Throughout the manuscript, “monitoring” has been replaced with “assessment” or “testing” when referring to the experimental activities, and “static basal tilt” has been revised to “static inclination analysis”.

• Sensor placement (1.3 m and 2.7 m) and reference instrument. We clarify that the choice of installing the sensors at 1.3 m and 2.7 m was motivated by a metrological objective. While we acknowledge that standard diagnostic protocols (e.g., SIM) prescribe sensor placement at the root collar to isolate basal rotation, the primary aim of this study was the characterization of the MEMS prototype against a professional reference instrument. The larger inclination amplitudes observed at these heights facilitate a clearer comparison between the MEMS sensor and the reference inclinometer than the very small rotations typically recorded at the root collar. As requested in Comment 6, the reference instrument has now been explicitly specified in Section 2.3 as a Dynatim™ biaxial inclinometer (RINNTECH, Heidelberg, Germany), with a nominal resolution of 0.001°.

• Research gaps and bending curves. We particularly appreciate the Reviewer’s suggestion regarding bending curve analysis. We agree that this represents a promising perspective for low-cost sensing approaches. The deployment of multiple sensors along the stem explored here through measurements at two heights suggests the potential for future analyses of stem deformation patterns that are currently difficult to implement with professional equipment due to cost and logistical constraints. This perspective has been explicitly acknowledged in the Conclusion as a potential direction for future research.

Comment 1:

L46-47 second part of this sentence needs further explanation as it is not correct. This tests were never intended to be used for long term monitoring per se, so it is not appropriate to highlight this as disadvantage of this method.

We thank the Reviewer for this clarification. We have revised the sentence to avoid characterizing the pulling test as a long-term monitoring tool, which was indeed an inappropriate comparison. The revised text now correctly defines pulling tests as 'point-in-time stability assessments' and clarifies that the limitation for large-scale application is due to high instrumentation costs and logistical complexity. Furthermore, we have expanded the introduction to distinguish between these static assessments and the continuous monitoring of tree movements under wind loading, which represents a separate and complementary assessment approach

Comment 2:

L50 ‘static basal tilt’ is not correct term as this does not represent the static pulling test or static load test. The static nature of this test and something that different it from dynamic measurement is known pulling direction and continues measurement of steadily applied pulling force (which produces moment on tree) under appropriate threshold which ensures that trees are not damaged after the test. This paragraph needs further refinement as terms are not represented in correct way. This is repeated several times through the manuscript which indicates that is systematic error in description. Furthermore, the large emphasis is given to the use of this approach in continuous monitoring of trees which fails to point to actual use that they have as part of tree stability process and on large scale the tree risk assessment.

Comment 3:

L55 ‘Static and dynamic monitoring’ it is not correct to use this phrase

We thank the Reviewer for the terminology clarification. We agree that the term “monitoring” can be misleading when referring to static pulling tests, which are typically point-in-time assessments or repeated checks rather than continuous monitoring. In the revised manuscript, we therefore removed “monitoring” when referring to pulling-test applications (e.g., in the study aims we replaced “static inclination analysis monitoring” with “quasi-static inclination measurements”), and we also revised the keywords accordingly. The term “monitoring” is now reserved exclusively for continuous measurements under natural wind loading, which we explicitly describe as a separate and complementary approach.

Comment 4:

L77 too much and too different references to be included in this sentence

We agree with the Reviewer. The term “structural monitoring” was inappropriate in this context and overstated the scope of the cited literature. The sentence has been revised to more precisely refer to the growing importance of tree stability assessment in urban forestry contexts, in line with the referenced studies.

Comment 5:

L93 this should not be the separate aim of this study as it goes beyond what is measured

We agree with the Reviewer. The objective referring to long-term deployment was too broad with respect to the experimental scope of the study. The aims have been revised to focus on the metrological characterization and field agreement of the MEMS inclinometer, while considerations on scalability and long-term use are now explicitly framed as implications and perspectives discussed on the basis of the experimental results.

Comment 6:

L130-131 the model and manufacture of this commercial inclinometer should be stated for comparison purpose. From image it seems that this is sensor that is part of Rinntech measurement system Dynatim

The model and manufacturer of the reference inclinometer have now been explicitly specified in Section~2.3 as a Dynatim™ biaxial inclinometer (RINNTECH, Heidelberg, Germany).

Comment 7:

L135 ‘and strain’?

We thank the Reviewer for the comment. No strain measurements were performed in this study, as the objective was limited to the metrological characterization of inclination measurements. The reference instrument was therefore used exclusively as a high-precision inclinometer, and the text has been clarified accordingly.

Comment 8:

L157-158 The issue with this peak force is that it does not state did you use calculate it based on angle of pulling cable in which you need to calculate it by multiplying with cosines of rope angle during peak force to know the applied force

We thank the Reviewer for this valuable comment on force calculation. We fully acknowledge that standard pulling test protocols require correcting the measured cable force for the pulling angle (F_lateral = F_measured × cos(θ)) to accurately compute the applied moment at the base, ensuring tree safety and result comparability. In this study, the cable angle θ (~15–20° from terrain slope) was trigonometrically estimated during setup and remained stable across pulls, with peak forces maintained below 10% DBH-equivalent thresholds. However, as both MEMS and reference sensors were co-located and experienced identical moments, trigonometric decomposition was not required for the metrological comparison of inclinations. Force served solely to stabilize plateaus. To enhance clarity, we have added a note in Section 2.3: 'Cable angle effects on lateral force were accounted for in protocol design but omitted from analysis, as sensor agreement was evaluated under shared loading conditions

Comment 9:

averaging window of 20 sec how was it selected? you describe it in protocol on processing the field data measurements, but did you did and statistic computation to back it up? As it includes around 520 measurements in this timeframe based on 26hz sampling interval it seems as long time frame to stabilize the readings from sensor? What is your explanation for this? it seems unpractical to have such 20 s window gaps to use this reading in field (for tree stability measurement using pulling test procedure)

Comment 10:

In my view it is unnecessary to have both table 3 and figure 5 and 6 for results of laboratory testing of applicability of sensors. You have determined the best average moving window and additional figures can be part of supplementary material. The issue is that this needs to be explained (which is related to previous comment about 20 s window). As it stands the position of figure 5 and 6 is wrong as they represent the field measurements results that are presented up to line 240 in manuscript.

We thank the Reviewer for these important and closely related remarks.

The 20 s averaging window used for field data was not selected arbitrarily, but was informed by the laboratory calibration, where the effect of different averaging windows (1–120 s) on noise reduction and mean inclination was systematically evaluated under fully static conditions. These laboratory results showed a progressive stabilisation of the MEMS signal with increasing window length, without altering the mean inclination.

In the field, data were acquired during 3-min pull–hold plateaus, and a single centred 20 s moving average (≈520 samples at 26 Hz) was adopted as an operational compromise to stabilise plateau estimates while preserving the step-like transitions between successive loading phases.

In line with this clarification, and to improve conciseness, we reduced redundancy in the presentation of laboratory and field results by grouping related figures into composite panels and moving non-essential material to a more compact layout, as suggested by the Reviewer.

Comment 10:

Figure 7 and 8 are too big and use too much white space. Results are similar and comparable so there is no need to use two pages for this.

We thank the Reviewer for these important and closely related remarks.

The 20 s averaging window used for field data was not selected arbitrarily, but was informed by the laboratory calibration, where the effect of different averaging windows (1–120 s) on noise reduction and mean inclination was systematically evaluated under fully static conditions. These laboratory results showed a progressive stabilisation of the MEMS signal with increasing window length, without altering the mean inclination.

In the field, data were acquired during 3-min pull–hold plateaus, and a single centred 20 s moving average (≈520 samples at 26 Hz) was adopted as an operational compromise to stabilise plateau estimates while preserving the step-like transitions between successive loading phases. No additional window-length sensitivity analysis was performed on field data, as the objective was instrument-to-instrument agreement under a common processing pipeline rather than filter optimisation.

In line with this clarification, and to improve conciseness, we reduced redundancy in the presentation of laboratory and field results by grouping related figures into composite panels and moving non-essential material to a more compact layout, as suggested by the Reviewer.

Comment 11:

What is difference between field test A and B? it is used on same tree and figures 7 -10 show that you have reference sensor to compare the results to? it does not show what table 1 is displaying that there was difference as with and without reference high precision sensor (A and B)

We thank the Reviewer for highlighting this ambiguity. Test~A and Test~B were both performed on the same tree and with the same experimental setup, and in both cases the low-cost MEMS inclinometer was co-located with the reference high-precision inclinometer to enable direct instrument-to-instrument comparison. The two tests do not represent configurations with and without the reference sensor; rather, they correspond to pulling sessions conducted under different (progressively increasing) load levels. Their purpose was to evaluate the stability and robustness of the MEMS–reference agreement across loading conditions, rather than strict test repeatability. We have revised Table~1 and the corresponding text in the Methods and Results sections to clarify this point and to remove any possible misunderstanding.

Comment 12:

L281-287 this behavior that you explain in this paragraph is called hysteresis and it is known to influence pulling test results in certain conditions (cyclic loading or as in your case prolonged period under load)

We thank the Reviewer for this clarification. We agree that the time-dependent inclination response observed during the constant-load phases can be interpreted, from a biomechanical perspective, as a hysteretic or relaxation behaviour of the stem–root–soil system under sustained loading. In the present study, however, this effect was not analysed in terms of force–displacement loops or energy dissipation, but was considered only phenomenologically, as it was synchronously detected by both the MEMS and the reference inclinometer. 

Comment 13:

L374 It is commendable that you point to some of disadvantages of use of this sensors, but you fail to point the amount of data that is collected during measurement (through data logger). Processing, and previous storing, of this data could be demanding and can influence the scalability of inclination sensors.

We thank the Reviewer for this insightful comment. The sampling frequency of the MEMS sensor (26 Hz) is explicitly stated in the Methods, and in the present study data storage and processing did not represent a limitation, given the scope and scale of the experiments and the fact that the data‐logging server was configured to handle the expected data flow. In addition, in our intended operational use the inclinometer is not designed to record continuously at 26 Hz, but rather to acquire data only during limited measurement windows (e.g. once or twice per day, for durations comparable to those identified as optimal in the laboratory tests). This acquisition strategy substantially reduces the overall data volume and mitigates potential constraints on storage and processing, improving the scalability of inclination sensors for larger monitoring networks.

Comment 14:

L479 View project??

We thank the Reviewer for pointing this out. The text “View project” was an editorial artefact from the reference manager and has been removed.

Comment 15:

The literature can be further expanded as discussion part is not really abundant with other studies on this topic. I would move away stacking of references in brackets during citation (as seen in discussion)

We thank the Reviewer for this constructive remark. We agree that, in some parts of the Discussion, references were grouped at the end of sentences, which may reduce the clarity of how the present results relate to previous studies.

In the revised manuscript, we have therefore reworked selected paragraphs of the Discussion to better integrate the literature into the argumentative flow, explicitly linking individual studies to specific aspects of our findings (e.g., MEMS agreement under quasi-static loading, use of low-cost sensors in tree biomechanics, and comparison with reference instrumentation).

Overview

The presented research is interesting as it addresses the problem with maintaining urban forest without risk of human health and well-being (risk of falling trees). Currently, there are some methods to evaluate tree stability, but they are not easy to apply in some cases and are related to significant costs. So, an innovative approach has been tested and validated in both laboratory and field experiments, allowing for a more easy to perform long-term surveys on urban trees, at a lower price and with less effort.

The topic corresponds well to the Forests journal aim and scope, and the study presents some new insights in the field, overcoming the limitations found in previous studies, reported in the scientific literature. The methodological part is well explained, allowing for reproducing the experiments. All instruments used are mentioned and technical data were presented. Figures, photos and tables are of a high quality and give an added value of the manuscript. Conclusions are robust and reflect the main results of the study. The authors can add more interpretations here aiming to highlight their significance. 

All references are adequate to the studied area.

Your work is interesting and has a significant practical implications, as it addresses the problem of ensuring the safety of population in relation to the sustainability of urban trees.

The manuscript follows the Instructions for authors, all sections are presented, the methodology is well explained, the figures and photos are of a high quality, English is fine.

We thank the Reviewer for the positive feedback. Following the suggestion to highlight the significance of our results, we have expanded the Conclusions section. We added a discussion on how this low-cost technology can facilitate large-scale, urban tree assessment, moving beyond site-specific assessments to improve public safety management.

Comment 1:

1) authorship: The authors' affiliations should be numbered chronologically, so Francesca Giannetti shoul be marked as 2,3,4 and so on.

We thank the reviewer, author affiliations and their corresponding numbering have been rearranged chronologically throughout the manuscript.

Comment 2:

2) Line 128: Please, add the name of the authors before ...[32].

We thank the reviewer, the names of the authors have been added before the citation [32] at Line 128, as requested.

Comment 3:

3) Line 310: The subsection title could be removed

We thank the reviewer for this comment, we deleted the subsection title

Comment 4:

4) Line 411: Authors names should be abbreviated (initials only)

The author names have been abbreviated to initials in the Author Contributions section. For the two authors sharing the same initials (F.G.), the full name has been added in parentheses to ensure correct attribution.

Overview

I highly appreciate the scientific quality of the presented research. Its practical value is also interesting. The use of cheaper measuring devices compared to those currently in use is of great importance in view of the large number of measurements required recently due to social pressure and climate change.

My main question concerns the research methodology. As a standard practice, during tensile tests, the inclinometer is placed as low as possible on the tree, between the root runs. The maximum inclinometer readings should not exceed 0.250. As shown by the presented research, for small angles of deflection, the MEMS sensor gives greater inaccuracies. Suspending the sensors at a height of 1.3 m and 2.7 m resulted in the observed angles of deflection being obviously greater (and therefore the accuracy of the sensor greater), because not only the movement of the root ball but also the deflection of the trunk was recorded. Therefore, the presented results clearly demonstrate the consistency of the results obtained in the tested methods, but they do not fully comply with the currently applied SIM method standards for determining the risk of tree fall. This should be clearly stated in the paper.

We thank the reviewer for this important methodological comment. We fully agree that, according to standard diagnostic protocols for static traction tests (e.g., SIM), inclinometers are typically installed at the base of the collar to isolate basal rotation and compare it with pre-established safety thresholds (e.g., 0.25°).

However, as the reviewer rightly points out, our experimental setup was intentionally designed for metrological characterization rather than for standard biomechanical stability assessment. The choice of heights of 1.30 m and 2.70 m was dictated by two main objectives:

Sensor characterization: higher positions provide greater deflection angles, which allows us to better characterize the performance of the MEMS sensor.

Practicality: these heights are more representative of potential long-term urban assessment, where protecting equipment from vandalism or accidental impacts is a priority.

As requested, we have explicitly clarified this point in the revised manuscript (Section 2.3, lines [row: 153 - 157]), stating that, although this configuration is ideal for comparing instruments, it deviates from the strict SIM protocol for anchor assessment. We believe that this clarification better defines the scope of our work.

Comment 1:

I have no fundamental comments regarding the English language. Precise technical terminology has been used (laboratory calibration, setup geometry, averaging windows, relative errors, etc.), the sentences are well balanced, logical and clear, and the passive voice, typical of academic style, has been used correctly.

We thanks the Reviewer for the positive feedback regarding the language and the technical terminology used in the manuscript

Overview

The title of manuscript is not correct in my opinion. The term ‘Static Tree Stability Monitoring’ is invalid and it is not used in literature in this field. As a term it incorrect as it should state ‘static pulling tests for tree stability assessment’ of something on those lines. The monitoring indicates that something is measured during longer period of time, and that is not the nature of this measurement as it is done mostly once and in rare occasions (such as trees that had been severed by construction works) it can lead to several measurements with larger time gap among them.

In my opinion there are some research gaps in inconsistencies that hamper the flow of thoughts on understanding for readers therefore major revisions of manuscript are suggested in order to improve the quality of this paper. It is unclear to me why a measurement sensor such as inclinometer was used at height of 1,3 and 2,7 m, when usually in pulling test measurement protocols this sensor is used to monitor the anchorage of root system during loading?

I see the application of this sensors in measurement of trees during pulling test to construct a bending curve from inclinometer data which may indicate the stability issues in stems (large strains on certain parts of tree) which is interesting application of this technique that is currently missing due to high price of reference precision sensor used in commercial products.

• We sincerely thank the Reviewer for the rigorous and technical critique, which has significantly contributed to improving both the clarity and methodological precision of the manuscript.

• Terminology and “Monitoring”. We fully accept the Reviewer’s observation. The term “monitoring” was indeed inappropriate to describe a single-event static pulling test. Accordingly, the title has been revised to “Preliminary Metrological Characterization of a Low-Cost MEMS Inclinometer for Tree Stability Assessment: From Laboratory to Field”. Throughout the manuscript, “monitoring” has been replaced with “assessment” or “testing” when referring to the experimental activities, and “static basal tilt” has been revised to “static inclination analysis”.

• Sensor placement (1.3 m and 2.7 m) and reference instrument. We clarify that the choice of installing the sensors at 1.3 m and 2.7 m was motivated by a metrological objective. While we acknowledge that standard diagnostic protocols (e.g., SIM) prescribe sensor placement at the root collar to isolate basal rotation, the primary aim of this study was the characterization of the MEMS prototype against a professional reference instrument. The larger inclination amplitudes observed at these heights facilitate a clearer comparison between the MEMS sensor and the reference inclinometer than the very small rotations typically recorded at the root collar. As requested in Comment 6, the reference instrument has now been explicitly specified in Section 2.3 as a Dynatim™ biaxial inclinometer (RINNTECH, Heidelberg, Germany), with a nominal resolution of 0.001°.

• Research gaps and bending curves. We particularly appreciate the Reviewer’s suggestion regarding bending curve analysis. We agree that this represents a promising perspective for low-cost sensing approaches. The deployment of multiple sensors along the stem explored here through measurements at two heights suggests the potential for future analyses of stem deformation patterns that are currently difficult to implement with professional equipment due to cost and logistical constraints. This perspective has been explicitly acknowledged in the Conclusion as a potential direction for future research.

Comment 1:

L46-47 second part of this sentence needs further explanation as it is not correct. This tests were never intended to be used for long term monitoring per se, so it is not appropriate to highlight this as disadvantage of this method.

We thank the Reviewer for this clarification. We have revised the sentence to avoid characterizing the pulling test as a long-term monitoring tool, which was indeed an inappropriate comparison. The revised text now correctly defines pulling tests as 'point-in-time stability assessments' and clarifies that the limitation for large-scale application is due to high instrumentation costs and logistical complexity. Furthermore, we have expanded the introduction to distinguish between these static assessments and the continuous monitoring of tree movements under wind loading, which represents a separate and complementary assessment approach

Comment 2:

L50 ‘static basal tilt’ is not correct term as this does not represent the static pulling test or static load test. The static nature of this test and something that different it from dynamic measurement is known pulling direction and continues measurement of steadily applied pulling force (which produces moment on tree) under appropriate threshold which ensures that trees are not damaged after the test. This paragraph needs further refinement as terms are not represented in correct way. This is repeated several times through the manuscript which indicates that is systematic error in description. Furthermore, the large emphasis is given to the use of this approach in continuous monitoring of trees which fails to point to actual use that they have as part of tree stability process and on large scale the tree risk assessment.

Comment 3:

L55 ‘Static and dynamic monitoring’ it is not correct to use this phrase

We thank the Reviewer for the terminology clarification. We agree that the term “monitoring” can be misleading when referring to static pulling tests, which are typically point-in-time assessments or repeated checks rather than continuous monitoring. In the revised manuscript, we therefore removed “monitoring” when referring to pulling-test applications (e.g., in the study aims we replaced “static inclination analysis monitoring” with “quasi-static inclination measurements”), and we also revised the keywords accordingly. The term “monitoring” is now reserved exclusively for continuous measurements under natural wind loading, which we explicitly describe as a separate and complementary approach.

Comment 4:

L77 too much and too different references to be included in this sentence

We agree with the Reviewer. The term “structural monitoring” was inappropriate in this context and overstated the scope of the cited literature. The sentence has been revised to more precisely refer to the growing importance of tree stability assessment in urban forestry contexts, in line with the referenced studies.

Comment 5:

L93 this should not be the separate aim of this study as it goes beyond what is measured

We agree with the Reviewer. The objective referring to long-term deployment was too broad with respect to the experimental scope of the study. The aims have been revised to focus on the metrological characterization and field agreement of the MEMS inclinometer, while considerations on scalability and long-term use are now explicitly framed as implications and perspectives discussed on the basis of the experimental results.

Comment 6:

L130-131 the model and manufacture of this commercial inclinometer should be stated for comparison purpose. From image it seems that this is sensor that is part of Rinntech measurement system Dynatim

The model and manufacturer of the reference inclinometer have now been explicitly specified in Section~2.3 as a Dynatim™ biaxial inclinometer (RINNTECH, Heidelberg, Germany).

Comment 7:

L135 ‘and strain’?

We thank the Reviewer for the comment. No strain measurements were performed in this study, as the objective was limited to the metrological characterization of inclination measurements. The reference instrument was therefore used exclusively as a high-precision inclinometer, and the text has been clarified accordingly.

Comment 8:

L157-158 The issue with this peak force is that it does not state did you use calculate it based on angle of pulling cable in which you need to calculate it by multiplying with cosines of rope angle during peak force to know the applied force

We thank the Reviewer for this valuable comment on force calculation. We fully acknowledge that standard pulling test protocols require correcting the measured cable force for the pulling angle (F_lateral = F_measured × cos(θ)) to accurately compute the applied moment at the base, ensuring tree safety and result comparability. In this study, the cable angle θ (~15–20° from terrain slope) was trigonometrically estimated during setup and remained stable across pulls, with peak forces maintained below 10% DBH-equivalent thresholds. However, as both MEMS and reference sensors were co-located and experienced identical moments, trigonometric decomposition was not required for the metrological comparison of inclinations. Force served solely to stabilize plateaus. To enhance clarity, we have added a note in Section 2.3: 'Cable angle effects on lateral force were accounted for in protocol design but omitted from analysis, as sensor agreement was evaluated under shared loading conditions

Comment 9:

averaging window of 20 sec how was it selected? you describe it in protocol on processing the field data measurements, but did you did and statistic computation to back it up? As it includes around 520 measurements in this timeframe based on 26hz sampling interval it seems as long time frame to stabilize the readings from sensor? What is your explanation for this? it seems unpractical to have such 20 s window gaps to use this reading in field (for tree stability measurement using pulling test procedure)

Comment 10:

In my view it is unnecessary to have both table 3 and figure 5 and 6 for results of laboratory testing of applicability of sensors. You have determined the best average moving window and additional figures can be part of supplementary material. The issue is that this needs to be explained (which is related to previous comment about 20 s window). As it stands the position of figure 5 and 6 is wrong as they represent the field measurements results that are presented up to line 240 in manuscript.

We thank the Reviewer for these important and closely related remarks.

The 20 s averaging window used for field data was not selected arbitrarily, but was informed by the laboratory calibration, where the effect of different averaging windows (1–120 s) on noise reduction and mean inclination was systematically evaluated under fully static conditions. These laboratory results showed a progressive stabilisation of the MEMS signal with increasing window length, without altering the mean inclination.

In the field, data were acquired during 3-min pull–hold plateaus, and a single centred 20 s moving average (≈520 samples at 26 Hz) was adopted as an operational compromise to stabilise plateau estimates while preserving the step-like transitions between successive loading phases.

In line with this clarification, and to improve conciseness, we reduced redundancy in the presentation of laboratory and field results by grouping related figures into composite panels and moving non-essential material to a more compact layout, as suggested by the Reviewer.

Comment 10:

Figure 7 and 8 are too big and use too much white space. Results are similar and comparable so there is no need to use two pages for this.

We thank the Reviewer for these important and closely related remarks.

The 20 s averaging window used for field data was not selected arbitrarily, but was informed by the laboratory calibration, where the effect of different averaging windows (1–120 s) on noise reduction and mean inclination was systematically evaluated under fully static conditions. These laboratory results showed a progressive stabilisation of the MEMS signal with increasing window length, without altering the mean inclination.

In the field, data were acquired during 3-min pull–hold plateaus, and a single centred 20 s moving average (≈520 samples at 26 Hz) was adopted as an operational compromise to stabilise plateau estimates while preserving the step-like transitions between successive loading phases. No additional window-length sensitivity analysis was performed on field data, as the objective was instrument-to-instrument agreement under a common processing pipeline rather than filter optimisation.

In line with this clarification, and to improve conciseness, we reduced redundancy in the presentation of laboratory and field results by grouping related figures into composite panels and moving non-essential material to a more compact layout, as suggested by the Reviewer.

Comment 11:

What is difference between field test A and B? it is used on same tree and figures 7 -10 show that you have reference sensor to compare the results to? it does not show what table 1 is displaying that there was difference as with and without reference high precision sensor (A and B)

We thank the Reviewer for highlighting this ambiguity. Test~A and Test~B were both performed on the same tree and with the same experimental setup, and in both cases the low-cost MEMS inclinometer was co-located with the reference high-precision inclinometer to enable direct instrument-to-instrument comparison. The two tests do not represent configurations with and without the reference sensor; rather, they correspond to pulling sessions conducted under different (progressively increasing) load levels. Their purpose was to evaluate the stability and robustness of the MEMS–reference agreement across loading conditions, rather than strict test repeatability. We have revised Table~1 and the corresponding text in the Methods and Results sections to clarify this point and to remove any possible misunderstanding.

Comment 12:

L281-287 this behavior that you explain in this paragraph is called hysteresis and it is known to influence pulling test results in certain conditions (cyclic loading or as in your case prolonged period under load)

We thank the Reviewer for this clarification. We agree that the time-dependent inclination response observed during the constant-load phases can be interpreted, from a biomechanical perspective, as a hysteretic or relaxation behaviour of the stem–root–soil system under sustained loading. In the present study, however, this effect was not analysed in terms of force–displacement loops or energy dissipation, but was considered only phenomenologically, as it was synchronously detected by both the MEMS and the reference inclinometer. 

Comment 13:

L374 It is commendable that you point to some of disadvantages of use of this sensors, but you fail to point the amount of data that is collected during measurement (through data logger). Processing, and previous storing, of this data could be demanding and can influence the scalability of inclination sensors.

We thank the Reviewer for this insightful comment. The sampling frequency of the MEMS sensor (26 Hz) is explicitly stated in the Methods, and in the present study data storage and processing did not represent a limitation, given the scope and scale of the experiments and the fact that the data‐logging server was configured to handle the expected data flow. In addition, in our intended operational use the inclinometer is not designed to record continuously at 26 Hz, but rather to acquire data only during limited measurement windows (e.g. once or twice per day, for durations comparable to those identified as optimal in the laboratory tests). This acquisition strategy substantially reduces the overall data volume and mitigates potential constraints on storage and processing, improving the scalability of inclination sensors for larger monitoring networks.

Comment 14:

L479 View project??

We thank the Reviewer for pointing this out. The text “View project” was an editorial artefact from the reference manager and has been removed.

Comment 15:

The literature can be further expanded as discussion part is not really abundant with other studies on this topic. I would move away stacking of references in brackets during citation (as seen in discussion)

We thank the Reviewer for this constructive remark. We agree that, in some parts of the Discussion, references were grouped at the end of sentences, which may reduce the clarity of how the present results relate to previous studies.

In the revised manuscript, we have therefore reworked selected paragraphs of the Discussion to better integrate the literature into the argumentative flow, explicitly linking individual studies to specific aspects of our findings (e.g., MEMS agreement under quasi-static loading, use of low-cost sensors in tree biomechanics, and comparison with reference instrumentation).

Overview

The presented research is interesting as it addresses the problem with maintaining urban forest without risk of human health and well-being (risk of falling trees). Currently, there are some methods to evaluate tree stability, but they are not easy to apply in some cases and are related to significant costs. So, an innovative approach has been tested and validated in both laboratory and field experiments, allowing for a more easy to perform long-term surveys on urban trees, at a lower price and with less effort.

The topic corresponds well to the Forests journal aim and scope, and the study presents some new insights in the field, overcoming the limitations found in previous studies, reported in the scientific literature. The methodological part is well explained, allowing for reproducing the experiments. All instruments used are mentioned and technical data were presented. Figures, photos and tables are of a high quality and give an added value of the manuscript. Conclusions are robust and reflect the main results of the study. The authors can add more interpretations here aiming to highlight their significance. 

All references are adequate to the studied area.

Your work is interesting and has a significant practical implications, as it addresses the problem of ensuring the safety of population in relation to the sustainability of urban trees.

The manuscript follows the Instructions for authors, all sections are presented, the methodology is well explained, the figures and photos are of a high quality, English is fine.

We thank the Reviewer for the positive feedback. Following the suggestion to highlight the significance of our results, we have expanded the Conclusions section. We added a discussion on how this low-cost technology can facilitate large-scale, urban tree assessment, moving beyond site-specific assessments to improve public safety management.

Comment 1:

1) authorship: The authors' affiliations should be numbered chronologically, so Francesca Giannetti shoul be marked as 2,3,4 and so on.

We thank the reviewer, author affiliations and their corresponding numbering have been rearranged chronologically throughout the manuscript.

Comment 2:

2) Line 128: Please, add the name of the authors before ...[32].

We thank the reviewer, the names of the authors have been added before the citation [32] at Line 128, as requested.

Comment 3:

3) Line 310: The subsection title could be removed

We thank the reviewer for this comment, we deleted the subsection title

Comment 4:

4) Line 411: Authors names should be abbreviated (initials only)

The author names have been abbreviated to initials in the Author Contributions section. For the two authors sharing the same initials (F.G.), the full name has been added in parentheses to ensure correct attribution.

Overview

I highly appreciate the scientific quality of the presented research. Its practical value is also interesting. The use of cheaper measuring devices compared to those currently in use is of great importance in view of the large number of measurements required recently due to social pressure and climate change.

My main question concerns the research methodology. As a standard practice, during tensile tests, the inclinometer is placed as low as possible on the tree, between the root runs. The maximum inclinometer readings should not exceed 0.250. As shown by the presented research, for small angles of deflection, the MEMS sensor gives greater inaccuracies. Suspending the sensors at a height of 1.3 m and 2.7 m resulted in the observed angles of deflection being obviously greater (and therefore the accuracy of the sensor greater), because not only the movement of the root ball but also the deflection of the trunk was recorded. Therefore, the presented results clearly demonstrate the consistency of the results obtained in the tested methods, but they do not fully comply with the currently applied SIM method standards for determining the risk of tree fall. This should be clearly stated in the paper.

We thank the reviewer for this important methodological comment. We fully agree that, according to standard diagnostic protocols for static traction tests (e.g., SIM), inclinometers are typically installed at the base of the collar to isolate basal rotation and compare it with pre-established safety thresholds (e.g., 0.25°).

However, as the reviewer rightly points out, our experimental setup was intentionally designed for metrological characterization rather than for standard biomechanical stability assessment. The choice of heights of 1.30 m and 2.70 m was dictated by two main objectives:

Sensor characterization: higher positions provide greater deflection angles, which allows us to better characterize the performance of the MEMS sensor.

Practicality: these heights are more representative of potential long-term urban assessment, where protecting equipment from vandalism or accidental impacts is a priority.

As requested, we have explicitly clarified this point in the revised manuscript (Section 2.3, lines [row: 153 - 157]), stating that, although this configuration is ideal for comparing instruments, it deviates from the strict SIM protocol for anchor assessment. We believe that this clarification better defines the scope of our work.

Comment 1:

I have no fundamental comments regarding the English language. Precise technical terminology has been used (laboratory calibration, setup geometry, averaging windows, relative errors, etc.), the sentences are well balanced, logical and clear, and the passive voice, typical of academic style, has been used correctly.

We thanks the Reviewer for the positive feedback regarding the language and the technical terminology used in the manuscript