Rainfall interception by buildings for urban hydrology modeling

  • W. van der Linden

Student thesis: Master


One of the results of elimate change is an increased amount of precipitation and a rise in rainfall intensity. This can lead to an overlaad on the sewer system or to the ground water level if no sewer system is present, and even to flooding. During rain events part of the rainwater can be intercepted by building surfaces (rainfall interception) . This amount of water is temporarily stared on these surfaces and eventually evaporates into the atmosphere. Since this intercepted water does nat reach the ground, it also does nat contribute to the laad on the sewer system. Especially porous materials on building roots but also in facades can intercept large quantities of water. Rainfall interception by buildings therefore becomes more and more important in hydrological modeling, particularly in urban areas. In this research rainfall interception by building facades is studied. In order to do this measurements as well as simulations are performed. Full-scale measurements are executed on a test facade at Eindhoven University of Technology, the Netherlands. The moisture response (mass change) to wind-driven rain of tour material samples is registered. Furthermore micro-meteorological data are gathered. The ditterenee in rainfall interception by two porous building materials is demonstrated. For the brick with high porosity the rainfall interception yields 100%, while the low porous brick accommodates for a minimum of 76.6% during the measurement period. Due to local saturation at the exterior boundary runaft occurred for the latter type of brick. Besides the tuli-scale experiments an extensive set of Iabaratory measurements is performed. Various moisture-related material properties are determined for the two porous building materials. For the numerical part of this study the finite element code HAMFEM is used. Toga in insight into the moisture response of porous building materials to wind-driven rain, one-dimensional heat and moisture transfer simulations are performed using differential balance equations for mass and energy. The micro-meteorological data gathered at the test facade are used as atmospheric boundary conditions and the results of the Iabaratory experiments are used to specity the material properties. The simulation results give the mass change of the materials in time and this is compared to the mass change obtained trom the tuli-scale measurements to verify the correct implementation of the model and the capability to reproduce the measurement results. lt is shown that for the brick with high porosity the model accurately prediets the rainfall interception during one rain event. Rainfall interception foranother rain event is described less precisely however. For the other type of brick the model fails to predict runoff because in the simulations the exterior boundary does nat reach the capillary moisture content, while in reality local saturation does occur because the wind-driven rain intensity on the facade exceeds the absorption rate of the brick. The largest deviations between experimental and numerical results are found in the evaporation process, where the model overestimates the mass decrease due to evaporation. This is to a large extent due to the uncertainty of the heat and moisture transfer coefficients. As a next step meteorological data of several elimate groups according to the Köppen elimate classification are used to specify the boundary conditions. With this the influence of different climates on rainfall interception is studied. lt is shown that for four of the six elimate groups used for the analysis no runoff occurred for the brick with high porosity for the period considered . This means that this type of brick accommodates for 100% rainfall interception in these climates. For the remaining two elimate groups the brick with high porosity yields a minimum RI of only 1.9% and 6% in periods of severe rain, increasing to 100% for moderate rain events. Comparison of the rainfall interception by the two bricks results in a rainfall interception of at least 40% by the brick with low porosity based on the rainfall interception by the other type of brick. lt is demonstrated however that the model nat always succeeds in accurately predicting the runoff. Therefore no hard conclusions can be drawn on the performance regarding rainfall interception by the two porous building materia Is in other climates.
Date of Award30 Jun 2010
Original languageEnglish
SupervisorBert Blocken (Supervisor 1), Jan Diepens (Supervisor 2), M.H. de Wit (Supervisor 2), Roel Hendrickx (External coach) & Staf Roels (External coach)

Cite this