Speaker
Description
Automated monitoring of root-zone soil moisture is essential for understanding tree water use and early drought stress. This study deployed an automated ERT time-lapse system along a transect with 100 electrodes at 0.5 m spacing, acquiring measurements every 8 hours in the ARBOfun Großpösna. A preprocessing workflow ensured data reliability, including removal of rainfall- and snow-affected measurements, outlier filtering, and temperature normalization. ERT data were integrated with meteorological observations and ten soil moisture sensors at 15 cm and 50 cm depths. Only sensors with strong correlation to ERT signals were retained (sensors 2, 6, and 7; R² > 0.32). Using paired observations (n = 420), univariate and multivariate regression models were evaluated. Multivariate models incorporating soil temperature consistently outperformed univariate models. At 15 cm, the multivariate model achieved R² = 0.77 (RMSE = 3.32%), compared to R² = 0.57 for univariate. At 50 cm, the multivariate model reached R² = 0.86 (RMSE = 2.86%), outperforming the univariate model (R² = 0.60).
Calibrated models were applied to the full ERT time series to derive continuous soil moisture profiles. Results show automated ERT can estimate root-zone soil moisture with ~80% accuracy, enabling detection of seasonal drying patterns and early tree water stress indicators.
| Status Group | Postdoctoral Researcher |
|---|---|
| FOR TALKS: Poster Presentation Option | No, I prefer to present only as a talk. |