Cosmic-Ray Neutron Sensing

Updated 31 January 2026

Cosmic-Ray Neutron Sensing is a non-invasive technique measuring environmental hydrogen content, particularly soil moisture, via detection of albedo neutrons.
It employs moderated neutron detectors and rigorous Monte Carlo simulations to correlate neutron count rates with spatially-averaged subsurface water content.
CRNS finds applications in hydrology, precision agriculture, and climatology, offering large-scale, continuous monitoring with effective bias corrections.

Cosmic-Ray Neutron Sensing (CRNS) is a non-invasive measurement technique for quantifying environmental hydrogen content, especially soil moisture, by detecting albedo neutrons in the epithermal to fast energy range. CRNS exploits the strong moderation effect of hydrogen on cosmic-ray induced neutrons, making neutron flux above ground an integrative proxy for subsurface water content over hectare-scale footprints. The technique leverages moderated neutron detectors—traditionally ^3He or BF₃ counter tubes, but also scintillator-based and Cherenkov detectors—to register neutron count rates, which are correlated with spatially-averaged soil moisture and other hydrogen pools such as vegetation and snow. A rigorous modeling framework based on Monte Carlo neutron transport enables quantitative interpretation of environmental neutron data, supporting applications spanning hydrology, precision agriculture, climatology, and particle astrophysics (Köhli et al., 2018, Stowell et al., 2021, Sarmiento-Cano et al., 24 Jan 2026).

1. Physical Principles and Environmental Sensitivity

Primary cosmic rays interact with the upper atmosphere, producing cascades of fast neutrons that reach the ground and diffuse into the soil. In the presence of hydrogen—primarily from water molecules—neutrons are efficiently moderated (slowed), altering the energy spectrum and intensity of upwardly scattered "albedo" neutrons detectable above ground. The key sensitivity of CRNS arises because hydrogen is the most effective neutron moderator among terrestrial elements. As a result, the near-surface neutron flux responds strongly to variations in soil moisture, snow cover, and biomass water content over a probing depth of decimeters and a lateral footprint in the range of 100–300 m radius (Köhli et al., 2018, Woolf et al., 2020, Sarmiento-Cano et al., 24 Jan 2026).

The observable—the neutron count rate $N$ —is linked to moisture via an exponential attenuation:

$N(\theta) = N_0 \exp(-\alpha \theta)$

where $N_0$ is the dry-soil count rate, $\theta$ is volumetric water content, and $\alpha$ is an attenuation coefficient encapsulating neutron cross-sections and effective path length in the hydrogenous environment (Woolf et al., 2020, Stowell et al., 2021, Sarmiento-Cano et al., 24 Jan 2026).

2. Detector Architectures and Response Functions

Classic CRNS detectors consist of one or more proportional counter tubes (either ^3He or ^10BF₃ as neutron-converter gases) surrounded by a $\sim$ 2.5 cm thick polyethylene moderator. The incoming epithermal and fast neutrons are slowed to thermal energies within the moderator, after which they are captured by the converter gas, producing charged particles that are registered electronically. Instrument variants deployed include stationary vertical tubes (e.g., Hydroinnova CRS1000, CRS1000/B) and large horizontal tubes for mobile surveys (the "Rover" system), each with distinctive footprint integration properties (Köhli et al., 2018, Schrön et al., 2017).

Detector response is fully characterized by an energy-dependent efficiency $\epsilon(E)$ and angular dependence $g(\theta)$ , combined in a two-dimensional response function:

$R(E,\theta) = \epsilon(E) \cdot g(\theta)$

$\epsilon(E)$ peaks between 1–10 eV (epithermal range), remains above 50% of peak over $0.1~{\rm eV}$ to $100~{\rm keV}$ , and reflects moderation thickness as the dominant factor shaping the energy response.
$g(\theta)$ declines monotonically from unity at normal incidence ( $\theta=0$ ) to zero at grazing, with vertical neutron fluxes being 4–5 times more likely to be detected than oblique ones.

Comparison of efficiency curves to Bonner Sphere reference detectors shows spectral agreement, confirming that polyethylene-moderated CRNS geometries act as partial Bonner Spheres (Köhli et al., 2018).

Detector	Converter Gas	$\epsilon_{\max}$ (%)	Orientation	Pseudo-Efficiency $\epsilon^*$
CRS1000	^3He	15 (side face)	Vert/Side	0.37
CRS1000/B	^10BF₃	6	Vert	0.02
Rover	^3He	25	Horiz/Side	0.85

3. Calibration, Modeling, and Signal Processing

Neutron transport simulations employing codes such as URANOS, GEANT4, and MCNP, informed by ENDF/B cross-section libraries, are integral for quantifying detector behavior under realistic environmental neutron spectra (Köhli et al., 2018, Stowell et al., 2021). Instead of detailed detector modeling in every large-scale simulation, the response function $R(E,\theta)$ allows rapid, accurate folding of flux spectra to expected count rates:

$N = \int_\Omega \int_0^\infty \Phi(E,\theta)\, \epsilon(E)\, g(\theta)\, dE \, d\Omega$

Here, $\Phi(E,\theta)$ is the environmental neutron flux at the detector's height and orientation. Detector geometry "unfolding" is only required for calibration or new design variants. Empirical tuning of $\epsilon(E)$ is frequently employed—field calibration against local gravimetric soil moisture corrects for uncertainties in soil composition or moderator tolerances (Woolf et al., 2020).

Neutron background corrections include:

Barometric pressure compensation, as neutron flux is anti-correlated with ambient pressure,
Adjustments for cosmic-ray intensity variations (solar-magnetospheric modulation), and
Corrections for surrounding material, including surface water bodies and man-made structures (Woolf et al., 2020, Schrön et al., 2017).

4. Alternative Detector Modalities: Scintillation and Cherenkov Approaches

Recent research targets cost and scalability constraints inherent to ^3He and BF₃ tubes:

Scintillator-Based Detectors: Li-ZnS(Ag) and BN-ZnS(Ag) composite scintillators enable smaller, lower-cost neutron detectors. The primary neutron reaction involves ^6Li or ^10B capture, producing $\alpha$ and triton or $^7$ Li, yielding sufficient light output (e.g., $\sim$ 47,000 photons/MeV for Li-ZnS) for reliable detection. Optimized moderator thickness (20–25 mm HDPE) and wavelength-shifting light guides maximize efficiency. Capture efficiency up to 52% of an ideal detector is demonstrated, with $\sim$ 0.34% soil moisture precision hourly over a 200 m footprint (Stowell et al., 2021).
Water Cherenkov Detectors (WCDs): Repurposed WCDs with salt (NaCl) doping leverage enhanced Cl-capture gamma emission for cosmic-ray neutron hydrometry in agriculture. NaCl-doped tanks reach detection efficiencies of 0.3–0.5% for albedo neutrons, with a hectare-scale footprint and minimum detectable changes of 0.4% volumetric water content in 30-minute integrations. Salinity-modified optical and gamma conversion parameters are handled in Geant4-based models (ARTI-MEIGA simulation chain) (Sarmiento-Cano et al., 24 Jan 2026). Gadolinium-loaded WCDs (Gd-WCDs) facilitate high-energy neutron monitoring in extensive air showers, where Gd's high thermal-neutron cross-section ( $\sim$ 50,000 barns) enables multi-microsecond timing discrimination between neutron-initiated and electromagnetic signals, with neutron detection efficiencies of 4–6% for $E_n>200$ MeV (Stowell et al., 2021).

Feature	WCD-CRNS	^3He Counter	Li-ZnS Scintillator
Invasiveness	Non-invasive	Requires burial	Non-invasive
Footprint	$\sim$ 2 ha	$\sim$ 0.05 ha	$\sim$ 200 m radius
Gas supply	None	High-pressure ^3He	None
Sensitivity to low moisture	Enhanced (Cl)	Moderate	High (Li-ZnS), moderate (BN-ZnS)
Cost	$\sim$ 1–2k USD	$\sim$ 10–20k USD	$\sim$ 1k USD
Scalability	High	Moderate	High

5. Mobile Surveys, Bias Corrections, and Spatial Integration

Mobile CRNS—ground vehicles equipped with neutron detectors—enables rapid spatial mapping of soil moisture over large domains. The neutron rover integrates field-scale moisture with a nominal footprint of 100–200 m radius, with the centroid of sensitivity trailing the detector by half the travelled distance per interval. Mobile deployment is sensitive to landscape heterogeneities; especially roadways, which can induce up to 40% bias in neutron count rates due to their typically lower hydrogen content compared to adjacent fields.

Extensive Monte Carlo simulations identified a bias decay length of $\sim$ 10 m from the road edge; analytical and empirical correction functions parameterized by road width, material, and relative moisture are validated against ground truth. Once corrected, mobile surveys achieve agreement with independent TDR (Time-Domain Reflectometry) measurements to within 1 vol% (Schrön et al., 2017).

6. Applications, Operational Aspects, and Emerging Directions

CRNS is deployed in hydrology (spatial soil moisture monitoring, flood forecasting), agriculture (irrigation scheduling, field-scale water resource management), and climate science (snow water equivalent mapping, evapotranspiration studies) (Woolf et al., 2020, Sarmiento-Cano et al., 24 Jan 2026). Its non-invasive, large-footprint nature complements point sensors and remote sensing.

Operational best practices:

Standardization of field calibration (local gravimetric measurements) for site-dependent conversion between count rate and moisture,
Regular barometric and meteorological correction for continuous deployments,
Placement of stationary detectors at $\sim$ 1–2 m above ground, away from local shielded or reflective structures.

Emerging areas include:

Wide-area real-time sensing networks leveraging low-cost scintillator detectors (Stowell et al., 2021),
Gd-WCD array deployments for air-shower neutron tagging (Stowell et al., 2021),
Cross-disciplinary integration with agronomy (precision farming) and watershed-scale modeling (Sarmiento-Cano et al., 24 Jan 2026).

7. Detector Optimization and Comparative Studies

Monte Carlo-based optimization underlines the trade-offs in geometry (planar vs. cylindrical), moderator thickness, and scintillator chemistry. Li-ZnS detectors achieve more than 50% of ideal efficiency at 20–25 mm HDPE moderation. Fluorescent WLS rods of high absorption length (EJ280 plastic) permit photon collection over economical tube lengths. For soil-moisture resolution and cost, well-designed scintillator-based CRNS systems rival or exceed traditional ^3He and BF₃ counters for agricultural-scale deployments (Stowell et al., 2021, Sarmiento-Cano et al., 24 Jan 2026).

In platforms like NEUCAL, scintillator-integrated neutron and moderation layer arrays facilitate electron/hadron discrimination for cosmic-ray calorimetry. Active moderation, prompt-scintillation and delayed-capture discrimination enable improvement in electron/hadron separation by 1–2 orders of magnitude over purely topological methods, critical for future cosmic-ray identification in satellite missions (Bonechi et al., 2010).

References:

(Köhli et al., 2018) Response Functions for Detectors in Cosmic Ray Neutron Sensing
(Schrön et al., 2017) The Cosmic-Ray Neutron Rover - Mobile Surveys of Field Soil Moisture and the Influence of Roads
(Woolf et al., 2020) Measurement of secondary cosmic-ray neutrons near the geomagnetic North Pole
(Stowell et al., 2021) Optimising Scintillating Neutron Detectors for Cosmic Ray Soil Moisture Monitoring
(Sarmiento-Cano et al., 24 Jan 2026) Water Cherenkov Detectors in Precision Agriculture
(Stowell et al., 2021) Gadolinium Loaded Cherenkov Detectors for Neutron Monitoring in High Energy Air Showers
(Bonechi et al., 2010) NEUCAL: a prototype detector for electron/hadron discrimination through neutron measurement