Impact of urban trees' rainfall interception on soil water dynamics

Vegetation characteristics are among the primary factors that influence soil water storage dynamics. Thus, this study aims to determine the interception capacity of urban trees and how differences in effective rainfall beneath these trees regulate soil water dynamics. To achieve this objective, interception capacity was estimated from measured throughfall. The soil water budget elements (including transpiration, soil evaporation, soil water storage, and deep percolation) were simulated using the HYDRUS-1D model. Model inputs include gross rainfall or effective rainfall (throughfall for under-tree soils), volumetric water content (VWC) of the soil, potential evapotranspiration (PET), leaf area index (LAI), and soil hydraulic parameters.

The study was carried out in a small urban park in the City of Ljubljana, Slovenia, between September 2024 and July 2025. The experimental plot includes various tree species, such as deciduous - birch (Betula pendula) and maple (Acer negundo), and evergreen - black pine (Pinus nigra), white pine (Pinus strobus), and yew (Taxus baccata). A tipping-bucket rain gauge was installed in the open area to measure gross rainfall. Throughfall beneath the birch and pine canopies was measured using a V-shaped steel trough collectors equipped with tipping-bucket flow gauges. Throughfall under other tree types was monitored using the same equipment to record open-area rainfall, but positioned under tree canopies. Each instrument was equipped with an automatic data logger that recorded data every 5 minutes. Additionally, soil VWC was monitored in the open area and beneath tree canopies (e.g., pine and birch) using TEROS 10 sensor probes. The probes were positioned horizontally at three successive depths within the soil profile (i.e., top: 16-20 cm, middle: 51-54 cm, and bottom: 74-76 cm). They were connected to data loggers programmed to record VWC at 5-minute intervals, enabling continuous monitoring of moisture variations across the soil profile. Meteorological variables (wind speed, solar radiation, air humidity, air temperature, rainfall, etc.) required to compute PET were collected from a remote weather monitoring station installed in the open area of the experimental site and recorded at 5-minute intervals. The LAI was measured using an LAI-2200C plant canopy analyzer at least twice per week to capture vegetation dynamics during the study period. The soil hydraulic parameters (saturated and residual VWC, saturated hydraulic conductivity, relative saturation, and shape parameters) under each tree were determined in the laboratory. Simulations were executed at an hourly timestep to capture short-term variations in the various water-balance components.

The calibrated HYDRUS-1D model was subsequently used to simulate soil water balance components across different tree species, using different effective rainfall as model input and employing different soil characteristics. The results show that rainfall interception, which defines effective rainfall beneath tree canopies, differs among trees. Thus, it impacts the various soil water budget parameters and soil water dynamics. The analyses conducted indicate the inter-relationship of rainfall interception processes and soil water dynamics.

Acknowledgment: The work was supported through the Ph.D. grant of the first author, which is financially supported by the Slovenian Research and Innovation Agency (ARIS). This study is also part of ongoing research programme P2-0180.