Determining soil moisture with an integrated monitoring and modelling approach in a humid area in NE Scotland

Soil moisture is a fundamental state variable for understanding hydrological processes in the vadose zone and the storage-discharge relationship of a catchment. However, soil moisture is highly variable in space, and that complicates the estimation of soil moisture values that could be representative of the whole catchment system. The Cosmic Ray Sensor (CRS) is a new promising technique that detects soil moisture over wide areas, covering up to 30 hectares, and it’s currently subject of research. In this thesis, the data detected by a CRS, installed at the intersection point of three fields, were combined with soil moisture data determined from samples and used to calibrate a Hydrus-1D model. Two of the three fields are cropped and the other one is a pasture; not only the land use, but also the soil type of the two cropped fields is different from the one of the pasture. The objectives of this thesis are: - to characterize the soil hydraulic properties of the three fields; - to explore the time variability of soil moisture for the two most dominant of these units by using a combination of field data and hydrological modelling; - to estimate large scale time variability of soil moisture by using values measured by CRS and modelling and to compare it with the results obtained for each unit. To address the first objective, the water retention curves of 44 samples collected from the three fields were determined in the laboratory. The results showed a high variability of the values of the hydraulic parameters both between fields and within each field. After that, two different Hydrus-1D models were calibrated, one representing the pasture and the other one representing the cropped fields. The calibrations aim to find the sets of hydraulic parameters, within suitable ranges, which lead to simulated soil moisture trends that best match the observed data. The observed data used were soil moisture values obtained by the samples from the fields. The ranges of the hydraulic parameters were set according to the results of the water retention curve analysis. Finally, another Hydrus-1D calibration was carried out by using the data detected by the CRS and the results were then compared to a weighted average of the two trends that were previously simulated (the weight was the portion of each field within the footprint). The results showed that the simulated CRS data compared well with the area-weighted average.