Results from a newly established long-term cosmogenic neutron observatory at kilometer scale with focus on soil water dynamics and distribution

Cosmic-ray neutron sensing (CRNS) has shown its capability for estimating soil water content by providing spatially integrated measurements at an intermediate scale between invasive in-situ and satellite remote sensing observations. This constitutes a major advantage over point-scale sensors, which are often sparsely installed and are affected by small-scale heterogeneity, leading to uncertain absolute values. CRNS thus serves as an important link between local and larger scales and is increasingly used as a reference for remote sensing products and hydrological or land-surface models and other applications related to soil water balance. However, to fully close the scale gap observations are needed that reach the km scale.

Within the DFG-research Cosmic Sense and the European project SoMMet (21GRD08), a multiscale soil moisture monitoring was implemented by establishing a cluster of CRNS integrated with a range of complementary in-situ observations. This Potsdam Soil Moisture Observatory (PoSMO) was established in 2023 and features an accumulated CRNS footprint size of close to one km² in total, constituting the largest long-term observation of epithermal cosmic-ray neutrons so far as well as the highest accumulated count rate of stationary CRNS worldwide. It comprises 16 stationary CRNS sensors located at an agricultural research site in the northeast of Germany, with some of the CRNS stations operated since end of 2019. They provide estimates of root-zone soil moisture at daily resolution, that is soil water content within the first decimeters of soil, but also co-located point-scale soil moisture measurements from shallow depth in 5 cm down to one meter. Intensive manual sampling campaigns of soil water content, bulk density, organic matter, and soil texture complement the dataset and enable robust CRNS calibration.

We discuss the PoSMO field set up, challenges associated with its design and the long-term monitoring operation. And we present the results of two years of harmonized soil water content time series from the different sensor types, including the CRNS cluster, shallow soil water content measurements, and soil water content profile data. Beyond the large area covered, CRNS and point sensors deliver also spatially resolved observations that will be shown as interpolated time-series of soil moisture maps for the inner part of the cluster. A sparser installation at the periphery and more singular sensors in the vicinity provide potential to even derive a soil moisture estimate for an area of up to 3.4 km². Also, the potential benefit of accompanying physical measurements of the neutron spectrum (via Bonner spheres), muon measurements with a scintillator-based CRNS or roving CRNS may be discussed as well as the link to the Brandenburg state CRNS network.