Field monitoring and numerical analysis of the influence of trees on soil moisture and ground movement in an urban environment

摘要

A great attribute of almost modern, liveable city is the presence of trees. Homeowners enjoy gardens with shrubs and trees which can improve real estate value, shade, and energy efficiency. Street trees are provided by local government authorities to improve the landscape, enhance the environment and to increase land values. Trees, however, can also cause problems if they become too large for the streetscape. They obstruct light, lose branches in storms and uplift pavements. More importantly, trees can cause damage to lightly loaded structures such as pavement and residential buildings. Trees use soil moisture for transpiration, the loss of water from the leaves generate a negative potential or suction on the leaves, which provides the pulling power to drive water from the soil to the leaves through the root and xylem. During the dry season, when evaporation rates are high, the tree roots must extend more deeply for finding water. This can cause greater ground movements in the dry season than would be expected without the presence of trees. Trees near to a residential building, especially Australian native species can extract large quantities of moisture from soils and lead to localized settlement. If the shrinkage settlement is significant, the buildings may deflect significantly which can result in structural damage. In Australia, distortions of pavements and dwellings caused by trees are widely reported, particularly in areas of expansive soil. Current engineering guidelines given in Australian Standard AS2870 (2011) are not based on adequate field research and measurement. Consequently, attempts to design footings to resist the additional ground movement due to trees are often flawed owing to poor understanding of the water demands of various tree species, wilting points and potential root development. This study is part of a long-term and comprehensive research programme aimed at minimising the risk of planting trees on clay soils in an urban environment. It involves field monitoring, laboratory testing, numerical simulation and case studies. In order to collect high-quality field data and develop an improved understanding of the physical processes that drive tree root-expansive soil interaction, a field site was established in 2011 in Glenroy, a northern suburb of Melbourne. This site was selected because it is a highly reactive site with basaltic clay in an urban environment. A 2.4 m high Eucalyptus ficifiolia and a 2 m high Meyer Lemon tree were planted on the site. Eucalyptus ficifiolia was chosen for this study because it is widely used as a street tree while the Meyer Lemon tree is one of the most popular citrus trees for home gardens. The instrumentation installed at the site includes: (a) automatic weather station; (b) HRM sap flow meters; (c) neutron moisture probe (soil moisture contents); (d) surface and sub-surface movement probes; (e) soil moisture sensors. Daily transpiration of the trees, in-situ soil moisture variations, ground movements, relative humidity, solar radiation, wind direction and velocity and rainfall have been closely monitored since May 2011. Field monitoring over a period of 5.5 years reveals that the presence of the tree resulted in an increase in the depth of soil moisture (suction) variation and a larger shrinking settlement of the ground. The changes in soil moisture and the associated soil movements were recorded mainly in the upper 2.0 m of the soil profile as the tree is still young. The experience gained to date from the field monitoring has shown that a sap flow meter is a reliable tool for measuring transpiration rate and water uptake by tree. The measured transpiration rate of the tree increases as the tree grows. The transpiration rate recorded in dry seasons is significantly higher than in wet seasons. A comprehensive laboratory investigation was also performed to complement the field data, which include (a) soil shrink-swell tests, (b) permeability tests, (c) soil suction measurements, (d) triaxial tests, (e) shear box tests and (f) suction-controlled oedometer tests (soil water characteristic curve, SWCC). The outcome of the field and laboratory investigation led to the publication of a comprehensive data set that not only benefits practitioners but also researchers as it can be used to evaluate the numerical models. As well as the field and laboratory work, a numerical model for tree root-soil interaction analysis has been developed to study the influence of tree root drying on the behaviour of unsaturated soil. A three-dimensional finite element approach based on ABAQUS has been employed to back-analysis the field experiment at the Glenroy site. The predicted results calculated using the soil, tree and atmospheric parameters obtained from the laboratory tests and field measurement, compared favourably with the field measurements. In addition to the field monitoring and numerical modelling, a case study of a cracked residential house in a southwest suburb of Melbourne was carried out. During the field investigation, daily transpiration and water uptake of a large street tree, located about 7 m from the northwest corner of the property, was monitored using two sap flow meters. The soil suction, shrink-swell indices and SWCC were measured. A numerical analysis was also conducted to estimate the soil suction distribution and soil movement under influence of large Eucalypt tree. The case study revealed the major cause of house distortion could be attributed to tree root drying, which resulted in non-uniform soil moisture conditions and significant footing settlement at the northwest corner. It has clearly shown that trees, growing in close proximity to a house could cause more severe damage to the buildings than the expected moisture changes due to seasonal influences and re-distribution of soil moisture arising from construction on the site. This research project has been unique in that it has incorporated and integrated a range of disciplines such as plant science, tree physiology, geotechnical and structural engineering. It includes the field monitoring, laboratory tests, numerical modelling and case study. The outcomes of this study have added valuable information to the body of knowledge in geotechnical and environmental engineering, therefore contributing to a higher level of understanding of the behaviour of unsaturated soil, daily transpiration and water uptake of trees, physical processes of the root-soil interaction and the impact of tree on residential footings.