Calibration of a catchment scale cosmic-ray probe network: A comparison of three parameterization methods

- We applied three parameterization methods for cosmic-ray soil moisture probes.
- The methods were evaluated with independent measurements.
- All three methods estimated areal average soil moisture with good accuracy.
- Method specific calibration parameters correlated strongly with aboveground biomass.

SummaryThe objective of this work was to assess the accuracy of soil water content determination from neutron flux measured by cosmic-ray probes under humid climate conditions. Ten cosmic-ray probes were set up in the Rur catchment located in western Germany, and calibrated by gravimetric soil sampling campaigns. Aboveground biomass was estimated at the sites to investigate the role of vegetation cover on the neutron flux and the calibration procedure. Three parameterization methods were used to generate site-specific neutron flux - soil water content calibration curves: (i) the N0-method, (ii) the hydrogen molar fraction method (hmf-method), and (iii) the COSMIC-method. At five locations, calibration measurements were repeated to evaluate site-specific calibration parameters obtained in two different sampling campaigns. At two locations, soil water content determined by cosmic-ray probes was evaluated with horizontally and vertically weighted soil water content measurements of two distributed in situ soil water content sensor networks. All three methods were successfully calibrated to determine field scale soil water content continuously at the ten sites. The hmf-method and the COSMIC-method had more similar calibration curves than the N0-method. The three methods performed similarly well in the validation and errors were within the uncertainty of neutron flux measurements despite observed differences in the calibration curves and variable model complexity. In addition, we found that the obtained calibration parameters NCOSMIC, N0 and NS showed a strong correlation with aboveground biomass.