Investigation of spatiotemporal variations in soil moisture in Søråsfeltet, Ås, applying different measurement techniques

Abstract

Soil moisture is crucial in biological, hydrological, and meteorological processes. However, it is highly variable in both time and space, and understand ing these variations is key to understand and accurately represent the related land-atmosphere interactions in numerical weather predictions (NWP). Despite increasing research on the topic, the spatiotemporal changes in soil moisture are not yet fully understood, and further research is needed for adequate representation of soil moisture in NWP. As a part of the Norwegian Meteorological Institute’s project Hydrometeorology to Operations (H2O), this thesis aims to broaden the understanding of spatiotemporal variations in soil moisture content and compare the performances of different measurement techniques. To this end, measurements of volumetric soil moisture content obtained from five different techniques applied in Sør˚asfeltet, Ås, have been analysed. The techniques include a manual soil sampling method, three different electromagnetic sensor types (ThetaProbe ML2, SoilVUE10, GroPoint Profile), and a cosmic-ray soil moisture probe. Time series from the manual soil sampling method and ThetaProbe ML2 were obtained during a manual measurement campaign from June to September 2021. The other sensors are permanently installed in the field, and from these, measurements from June to December 2021 have been used. In addition, a soil specific calibration process of the GroPoint Profile sensors has been conducted and evaluated. The different measurement techniques agreed upon the general soil moisture trends. Down to 60 cm depth, soil moisture closely followed the precipitation patterns, and precipitation was found to be a significant (P < 0.05) predictor of soil moisture down to 30 cm depth. At 75 cm depth, soil moisture increased only during larger precipitation events. At 100 cm depth, soil moisture only changed by 1.5 % throughout the research period. On specific dates, the measurements from the ThetaProbe ML2 varied up to 20 % across Sør˚asfeltet. The spatial variability could be due to differences in soil texture or elevation across the field, but influence from buildings was also detected. However, a large variation is to be expected as over 100 ThetaProbe measurements were taken on each date. From these measurements, soil moisture variability seemed to increase with increasing soil moisture. Displaying the standard deviation of the cosmic-ray measurements as a function of the mean yielded an upward convex shape peaking around 41 %. The SoilVUE10 sensor consistently measured lower values than the other sensors but exceeded them during heavy rainfall. The soil moisture peaks during rainfall are likely due to holes near the sensor, making the sensor come in direct contact with water and thus over-representing the soil moisture in the area. The unrealistically low values cannot be as easily explained but might partially be attributed to the sensor’s location. Therefore, the sensor should be moved before any conclusion on the sensor’s performance is drawn. Also, more physically sound quality controls of the permanently installed sensors are recommended to ensure reliable measurements. The soil-specific calibration process provided for the GroPoint Profile sensors was deemed too inaccurate to give satisfactory results. In general, it is not suitable for soils with high clay content. Instead, performing the calibration on-site is recommended.

Fuktinnhaldet i jord er avgjerande i biologiske, hydrologiske og meteorologiske prosessar. Men, dette fuktinnhaldet er høgst variabelt i både tid og rom, og det å forstå desse endringane er essensielt for å kunne forstå og nøyaktig representere dei tilhøyrande land-atmosfære-samhandlingane i numeriske værmodellar. Trass i aukande forsking på feltet, er endringane i jordfukt i tid og rom endå ikkje heilt forståtte, og meir forsking er naudsynt for å opppnå betre representasjon av jordfukt i numeriske værmodellar. Som ein del av Meteorologisk Institutt sitt prosjekt Hydrometeorology to Operations (H2O), har denne oppgåva som mål å utvide forståinga av endringar i fuktinnhald i jord i både tid og rom og å samanlikne ytinga til ulike målemetodar. For å oppnå dette har jordfuktmålingar frå fem ulike målemetodar anvendte på Søråsfeltet, Ås, blitt analyserte. Målemetodane inneber ein manuell målemetode, tre ulike elektromagnetiske sensorar (ThetaProbe ML2, SoilVUE10, GroPoint Profile) og ein ”cosmic-ray”-sensor. Tidsseriar frå den manuelle målemetoden og ThetaProbe ML2 vart samla inn under ein manuell målekampanje frå juni til september 2021. Dei andre sensorane er installerte i feltet, og målingar frå juni til desember 2021 er henta frå desse. I tillegg har ein jordspesifikk kalibreringsmetode for GroPoint Profile-sensorane blitt utført og evaluert.