Spatio-temporal soil moisture retrieval at the catchment-scale usinga dense network of cosmic-ray neutron sensors

Abstract. The method of Cosmic-Ray Neutron Sensing (CRNS) is a powerful technique to retrieve representative estimates of soil water content at a horizontal scale of hectometers (the field scale ) and depths of tens of centimeters ( the root zone ). This study demonstrates the potential of the CRNS technique to obtain spatio-temporal patterns of soil moisture beyond the integrated volume from isolated CRNS footprints. We use data from an observational campaign between May and July 2019 which featured a network of more than 20 neutron detectors with partly overlapping footprints in an area that exhibits pronounced soil moisture gradients within 1 km2. The present study is the first to combine these observations in order to represent the heterogeneity of soil water content at the sub-footprint scale as well as between the CRNS stations. First, we apply a state-of-the-art procedure to correct the observed neutron count rates for static effects such as sensor sensitivity, vegetation biomass, soil organic carbon and lattice water, as well as for the influence of the temporally dynamic factors barometric pressure, air humidity, and incoming flux of neutrons. Based on the homogenised neutron data, we investigate the robustness of a uniform calibration approach using one calibration parameter value across all CRNS stations. Finally, we benchmark two different interpolation techniques in order to obtain space-time representations of soil moisture: first, Ordinary Kriging with a fixed range; second, a heuristic approach that complements the concept of spatial interpolation by the idea of a geophysical inversion ( constrained interpolation ). For the latter, we define a geostatistical model of the spatial soil moisture variation in the study area, and then optimize the parameters of that model so that the error of the forward-simulated neutron count rates is minimized. In order to make the optimization problem computationally feasible, we use a heuristic forward operator that is based on the physics of horizontal sensitivity of the neutron detector. The comparison with independent measurements from a cluster of soil moisture sensors (SoilNet) shows that the constrained interpolation approach outperforms Ordinary Kriging by putting a stronger emphasis on horizontal soil moisture gradients at the hectometer scale. The study demonstrates how a CRNS network can be used to generate consistent interpolated soil moisture patterns that could be used to validate hydrological models or remote sensing products.