Signal response of bare and moderated cosmic-ray neutron sensors to varying soil and biomass conditions

Rasche, Daniel; Brogi, Cosimo; Köhli, Markus; McJannet, David; Weimar, Jannis; Schrön, Martin; Blume, Theresa; Güntner, Andreas

doi:10.5194/gi-15-291-2026

Articles | Volume 15, issue 2

https://doi.org/10.5194/gi-15-291-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gi-15-291-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 15, issue 2

Research article

|

28 Oct 2026

Research article |

| 28 Oct 2026

Signal response of bare and moderated cosmic-ray neutron sensors to varying soil and biomass conditions

Daniel Rasche, Cosimo Brogi, Markus Köhli, David McJannet, Jannis Weimar, Martin Schrön, Theresa Blume, and Andreas Güntner

Download

Final revised paper (published on 28 Oct 2026)
Preprint (discussion started on 31 Mar 2026)

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2026-1495', Anonymous Referee #1, 08 Apr 2026
The authors present a timely and useful modeling study examining how bare and moderated cosmic-ray neutron detectors respond to changing soil moisture, biomass, soil chemistry, bulk density, and above-surface water conditions. A major strength of the manuscript is the move from idealized thermal/epithermal energy-window interpretations toward detector-response-function-based analysis, which makes the results more relevant to real CRNS instruments in the field. The paper also makes an important conceptual contribution by showing that bare detector signals are not equivalent to “pure thermal neutron” behavior, but instead reflect a mixed response that still contains the legacy of the epithermal stage of neutron transport. Overall, I found the manuscript strong, well motivated, and likely to be of interest to the CRNS community. I believe it is suitable for publication after the following revisions are addressed.
Major comments
Practical workflow / guidance for CRNS users.

As a long-time CRNS user, I would strongly appreciate a short practical workflow figure or summary box that helps readers decide when a simpler transfer-function framework is sufficient and when the more complete formulations are needed. At present, the manuscript does a very good job showing where complexity arises, but it offers less guidance on how practitioners should respond to that complexity in routine use. This would be especially helpful for newer users, for whom Eq. 6 may appear quite complex relative to the traditional appeal of CRNS as a robust and operationally simple soil-moisture method. The manuscript already demonstrates that biomass, water layers, bulk density, and detector configuration can all modify the soil-moisture response, and that the modified UTS (Eq. 6) can improve the representation of biomass effects, although its broader validation remains outside the scope of the study.
I would encourage the authors to add a brief “recommended use cases” workflow along the following lines:
Use a simpler transfer function (e.g., Desilets-style / N0-type approach) when the site is relatively simple: low to modest biomass, limited seasonal biomass change, no persistent snow or ponded water, fairly typical mineral soils, and no expectation of strong heterogeneity or unusual absorber chemistry. In these settings, simplicity, interpretability, and operational robustness are often more valuable than adding poorly constrained correction terms.

Use the UTS framework when the site is dry, when improved performance at low soil moisture is important, or when bulk density / atmospheric effects need to be handled more consistently in a physically based framework.

Use the full Eq. 6 or similarly expanded approaches with caution and only when justified by site conditions—for example, where above-ground biomass is substantial and varying, where independent biomass information is available, and where the additional model complexity is supported by the measurement objective.

Advise caution regardless of transfer function choice in settings with tall or changing vegetation, shallow water or snow layers, strong soil-chemistry contrasts, unusual bulk density structure, or configurations where bare and moderated detector behavior diverges from standard assumptions. These are precisely the cases where the manuscript shows the response can become non-monotonic or more site-specific.

I think even a simple schematic like this would make the paper much more helpful to practitioners. It would also help balance an important tension in the paper: on one hand, the manuscript correctly highlights important limitations and complexities; on the other hand, most users still need practical guidance on when those complexities materially affect soil-moisture estimation and when simpler transfer functions remain adequate.
Suggested CRNS workflow for transfer-function choice
Start with the study objective: routine field-scale soil moisture monitoring, or process-rich interpretation in a complex environment?

Check for major complicating factors:

strong seasonal biomass change

tall vegetation interacting with detector height

snow, ponding water, or shallow surface water

unusual soil chemistry / absorbers

strong bulk density variation
If none or only minor factors are present, a simple Desilets/N0-type calibration may be sufficient.
If the site is dry or accuracy at low water content matters, favor UTS.
If biomass is substantial and variable, and biomass information is available, consider Eq. 6 or another biomass-aware framework.
If several complicating factors co-occur, emphasize uncertainty and site-specific calibration rather than defaulting to a more complex equation automatically.

Minor Comments

Abstract, p. 1, line 2

“researchers have hypothesize” → “researchers have hypothesized”.
Abstract, p. 1, lines 14–15

“the simulation results shed further light on empirical findings made in previous studies, they set a baseline…”

Consider revising to avoid the comma splice, for example:

“the simulation results shed further light on empirical findings made in previous studies, set a baseline…”

or split into two sentences.
Introduction, p. 3, line 86

“the simulations results” → “the simulation results”.
Methods, p. 4, line 112

“However, In this study” → “However, in this study”.
Methods, p. 5, lines 131–132

“A detailed description of the different homogeneous setups in given in the following sections.”

→ “A detailed description of the different homogeneous setups is given in the following sections.”
Methods, p. 5, line 132

“Fig. Fig. 1a-c” → “Fig. 1a–c”.

Duplicate “Fig.” should be removed.
Methods, p. 7, lines 148–149

“The use of a homogeneous material layers…”

→ “The use of homogeneous material layers…”
Methods / Table 2 caption, p. 7

“different amount of above-ground biomass” → “different amounts of above-ground biomass”.
Methods, p. 8, line 155

“increments .” → remove extra space before the period.
Figure 1 caption, p. 10

“with an vertically expanding soil layer” → “with a vertically expanding soil layer”.
Results, p. 11, lines 208–211

“In the case of a thermal neutrons…” → “In the case of thermal neutrons…”

Also, “continuos decrease” → “continuous decrease.”
Results, p. 13, lines 243–245

“and thus, exhibits a lower signal-to-noise ratio making it less favourable…”

This sentence reads awkwardly. Consider:

“and thus exhibits a lower signal-to-noise ratio, making it less favourable…”
Results, p. 14, line 249

“previous studies(e.g.,” → add missing space: “previous studies (e.g.,”
General style throughout pp. 1–15

The manuscript alternates a bit between “bare detector,” “bare neutron detector,” “thermal neutrons observed with bare detectors,” and “bare detector signal.” I would recommend tightening terminology so that “thermal neutrons” and “bare detector signals” are not used interchangeably, especially since one of the paper’s conceptual points is that they are not strictly the same thing.

Citation: https://doi.org/10.5194/egusphere-2026-1495-RC1

AC1: 'Reply on RC1', Daniel Rasche, 11 Jun 2026

Please find the author responses to the comments of both reviewers in the supplement.

Citation: https://doi.org/10.5194/egusphere-2026-1495-AC1

RC2: 'Comment on egusphere-2026-1495', Anonymous Referee #2, 01 May 2026

The manuscript presents a simulation-based assessment of bare and moderated CRNS detector responses under varying environmental conditions. The topic is technically relevant to the CRNS community, and the use of detector response functions is potentially useful. However, in its current form, the manuscript reads primarily as a technical sensitivity study, with limited demonstration of broader methodological or practical impact. I also have concerns regarding the justification of some modelling choices, the treatment of uncertainty, and the accessibility of the manuscript to a wider GI readership. I therefore do not recommend this manuscript for publication. However, if the Editor considers the topic suitable for GI, the authors could be invited to submit a substantially revised manuscript.

Specific comments:

L103–105: It is not clear to me why soil moisture and absolute humidity need to be fixed here. Please expand the explanation. Is the detector not sensitive to those variables?

L106–108: My understanding is that, due to lower costs, the BF3 tubes from Hydroinnova, such as the CRS1000/B, tend to be more popular for deployment, especially at larger volume or in developing countries with financial constraints. Why not base the results on the CRS1000/B design? Would the CRS1000 design not further limit the potential reach of this study?

L112–113: How would that choice affect our understanding of actual co-located setups in situ? It would seem more appropriate to model what is effectively “seen” in the field.

L144–145: Why is the detector simulated in URANOS like a detector, while plants are simulated as a “gas layer”? Have the authors tested the impacts of this simplification on the overall results? They earlier mention the potential for biomass estimation using a bare detector, so it seems appropriate to mimic the actual site conditions as accurately as possible in order to evaluate such potential.

Section 2.2.3: How did the authors consider the fact that snow density is different from liquid water density? It appears that the proposed scenarios are based on liquid water layers only.

Section 3.2.2: The authors have not clearly explained why this transfer function is needed, only that it was lacking for bare detectors. What is the intended usage of this equation? Are users expected to switch from moderated to bare detectors?

Results section: In general, I found several statements to lack a follow-up implications sentence to help the general reader understand the contribution of this paper. The manuscript appears to have been written primarily for CRNS detector specialists. Even hydrologists, agricultural scientists, and many non-specialist CRNS users may struggle to follow this manuscript.

Figure 7, right-column panels: It seems that the region where there is large variability in the signal, between approximately 1 and 10 BWE, is also where fewer simulated cases were carried out in the study. Would it improve the results to simulate this range with more points?

L398–413: There is no discussion about how the plant “gas layer” simplification could have driven these differences against empirical estimates. Could the authors discuss this?

L433–435: How could uncertainty in the non-linear equation hinder this improvement? The non-linear equation adds significant complexity compared with the linear approach developed by Baatz. It is not clear to me that this added complexity significantly changes the response to BWE sufficiently to justify abandoning the linear approach. This needs to be further investigated. Are the data points in Figure 8b significantly different from those in Figures 8c or 8d?

L480–482: Knowing that the neutron counts observed by the sensor follow a Poisson distribution, with mean = N and standard deviation = sqrt(N), and generally assuming that a considerable number of neutrons will have interacted with the soil moisture “source”, this would reduce the overall “leftover count/signal” to be attributed to other processes. Lower effective counts mean higher overall uncertainty. Yet here, and in similar papers trying to assess the nature of CRNS measurements of other quantities simultaneously, the manuscript does not adequately account for the uncertainty propagated to these other “observed” processes if the soil moisture signal is removed from the overall signal. The authors should address this.

Conclusions: This section is heavily written for readers with high expertise in CRNS and perhaps particle transport physics. It seems to target a very specialised and niche audience. If I am simply a typical CRNS user, it is unclear what I gain from this study. For example: should I replace my CRS1000/B with a CRS1000? What happens if I only have the moderated tube? Which conditions are more or less recommended for deploying both tubes from a CRS1000? How do I simultaneously utilise the signal from both moderated and bare tubes if their footprints are different?

Citation: https://doi.org/10.5194/egusphere-2026-1495-RC2

AC2: 'Reply on RC2', Daniel Rasche, 11 Jun 2026

Please find the author responses to the comments of both reviewers in the supplement.

Citation: https://doi.org/10.5194/egusphere-2026-1495-AC2

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

AR by Daniel Rasche on behalf of the Authors (07 Jul 2026) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (11 Jul 2026) by Mehrez Zribi

RR by Anonymous Reviewer #1 (23 Jul 2026)

ED: Publish as is (04 Aug 2026) by Mehrez Zribi

AR by Daniel Rasche on behalf of the Authors (19 Aug 2026) Author's response Manuscript

Short summary

Cosmic-Ray Neutron Sensing is a passive technique to estimate soil moisture, plant traits and snow at the field scale by using instruments sensitive to different neutron energy ranges. For devices sensitive to lower energies, a generalised assessment of the signal response under varying environmental conditions is currently missing. Our simulation-based study highlights the differing signal response of different instruments to environmental variables and provides a baseline for further research.

Signal response of bare and moderated cosmic-ray neutron sensors to varying soil and biomass conditions

Download

Interactive discussion

Peer review completion

Suggestions for revision or reasons for rejection