A novel cosmic-ray neutron detector for soil moisture estimation over large areas

Water scarcity and droughts problems in several parts of the world highlight the necessity of new solutions for better management of water resources. The prerequisite is reliable soil moisture data, measured over large-scales and in real-time. Due to this crucial role, many devices have been developed to measure soil moisture at different spatial and temporal scales. Available technologies range from point-scale invasive approaches as for instances Time Domain Reflectometry (TDR) probes to remote sensing approaches, like satellite remote methods. Nevertheless, practical problems arise when using these techniques: point-scale probe are invasive and their estimate difficult to scale up to field level, on the other hand, remote sensing exhibit unsuitable temporal resolution and, most importantly, they are sensitive only to a thin part of the soil and land surface. In the last decade, to overcome operational challenges of the aforementioned techniques, a proximal geophysical method has been developed, in order to fill the gap between point scale and remote sensing approaches: the Cosmic-ray Neutron Sensing (CRNS). CRNS is a valid and robust alternative, offering many advantages: it is contactless, allows quantification of soil moisture averaged over large areas with only one probe, and is not invasive for agricultural field operations. The significant advantages of the CRNS are its large horizontal footprint (up to tens of hectares) and the penetration depth of tens of centimeters, enough to reach typical roots depth. State-of-the-art probes used in CRNS are based on Helium-3 proportional counter tubes. Helium-3 is a nuclide produced almost entirely in artificial contexts as sub-product of the tritium decay, the current storage is depleting, and the price is high and rising, as it comes mainly from the production or dismantling of the nuclear weapons of the past decades. In the thesis, a new solution was studied. This new probe is based on a composite detector made of commercial scintillation detectors: EJ-276 and EJ-420(6) both manufactured by Eljen Technology (USA). In this kind of detectors, particles are identified and discriminated according to the generated signals, with an algorithm based on Pulse Shape Discrimination (PSD) which exploits the different processes activated by different particles interacting in the scintillator. PSD parameters were optimized with the aim of ensuring optimal discrimination capabilities. The readout is made of a flat panel photomultiplier H8500 from Hamamatsu, a fast digitizer DT5725 from CAEN, a low-cost, low-power, embedded computer Beaglebone black from Beagleboard and a low cost High Voltage power supply A7505 from CAEN. The digitizer is interfaced with the embedded computer, which hosts the acquisition and the analysis software. The online analysis software was developed as a distributed system where each task is handled by an independent server always running. Finally, the probe was installed in a recent orchard with walnut trees for four months. Results showed that the reconstructed soil moisture is well correlated with precipitations, and it shows the limit of standard measurement, that can be influenced by the heterogeneity of the soil and the irrigation distribution.