Modeling and assessing the impact of tunnel drainage on terrestrial vegetation

Highlights

• An assessment framework is built for the tunnel drainage impact on plants.

• A numerical model is introduced to simulate water circulation around tunnels.

• The tunnel factor can be integrated into the soil–plant–atmosphere continuum.

• The framework can optimize tunnel parameters to reduce ecological influences.

• A complete case study shows the applicability of the framework.

Abstract

The drainage of mountain tunnels can cause groundwater loss and result in shallow groundwater depletion, which jeopardizes terrestrial vegetation. The impact of tunnel drainage on vegetation can be examined from a physiological perspective using monitoring techniques or evaluated as one item in an environmental assessment of the entire tunnel project. Nevertheless, few previous studies have quantitatively assessed the regional impact of tunnel drainage on plants. We established an assessment framework based on a regionally coupled hydrological model that integrated the tunnel factor into the soil–plant–atmosphere continuum (SPAC) to evaluate the vulnerability status of vegetation threatened by tunnel drainage on a regional scale. The framework comprises five components: (1) a one-dimensional (1D) topsoil model for the vertical unsaturated flow through soil and plants; (2) a three-dimensional (3D) groundwater seepage model delineating water movement between the tunnel and groundwater; (3) a one-way coupling scheme for saturated–unsaturated flow; (4) vegetation-dependent atmospheric boundary conditions representing moisture exchanges between the topsoil and atmosphere; and (5) blockwise vulnerability assessments based on the soil matric potential and vegetation wilting point. The proposed framework was applied to an actual mountain tunnel project. The results showed that the framework can help (i) compare and optimize tunnel design/construction parameters to minimize the impact of tunnel drainage on vegetation, (ii) evaluate the regional impact of long-term/transient drainage of a specific tunnel on plants, and (iii) determine the dominant factor controlling vegetation survival in a given region. Furthermore, the regional impacts of groundwater discharge from the Mingtang Tunnel on vegetation were quantitatively investigated using the assessment framework. It was found that the limited drainage solution of the three typical drainage designs could significantly reduce the area of vulnerable vegetation affected by this tunnel. Groundwater discharge from the Mingtang Tunnel that adopted the limited drainage scheme could harm the growth and yield of vegetation in certain areas, but was unlikely to wither plants, even during drought disasters. In addition, vegetation in this region was more influenced by atmospheric conditions than tunnel drainage. The proposed method can provide a novel perspective and practical assessment tool for promoting environmentally friendly tunnel engineering.

Introduction

Tunnels have been extensively used for the development of transportation infrastructure such as highway and railway projects, especially in mountainous areas. Concerns about the negative environmental impacts of tunnels have grown as environmental protection concepts have become increasingly prevalent (van Geldermalsen, 2004, Sweetenham et al., 2017). Because tunnels are often surrounded by groundwater, their drainage may change groundwater flow patterns (Zaidel et al., 2010) and cause groundwater decline (Yoo, 2005, Moon and Fernandez, 2010), further affecting surface hydrological processes (Gargini et al., 2008, Vincenzi et al., 2009, Raposo et al., 2010) and threatening the surrounding eco-environment (Kværner and Snilsberg, 2011, Li et al., 2016). For example, the maximum groundwater inflow into the Zhongliangshan Tunnel of the Xiang-Yu Railway reached 54,100 m³/d, which resulted in the drying up of 48 wells and declines in the water levels of over 100 wells and springs (Zheng et al., 2017). Moreover, the Huayingshan Tunnel of the Guang-Yu Highway discharged approximately 730 × 10⁴ m³ of groundwater annually, accounting for 44% of the annual groundwater recharge in this region, which dried up springs, unbalanced the water circulation, and caused severe soil erosion (Liu et al., 2001).

In projects similar to those described above, vegetation, which is a vital component of the eco-environment, can also be imperiled by tunnel-induced groundwater decline (Lv et al., 2020). Consequently, the ecosystem services (Daily, 1997, Millenium Ecosystem Assessment, 2005) provided by plants, including climate regulation, soil erosion control, habitat provision, and food/material production (Krieger, 2001, García-Nieto et al., 2013), can also be adversely affected by groundwater discharge, which may become a concern for underground space planning (Bobylev, 2018). In more severe cases, the detrimental effects of tunnel drainage on plants can escalate to damage to the surrounding ecosystem (Kværner & Snilsberg, 2008) and local agriculture (Fais and Nino, 2004, Wei and Pan, 2011). Accordingly, modeling and assessing tunnel drainage impacts on vegetation will facilitate the inclusion of ecosystem service considerations in underground space planning and promote more environmentally friendly tunnel engineering.

Previous studies have primarily examined the impact of tunnel drainage on vegetation based on long-term monitoring and on-site tests. By monitoring variations in tree ring width, it was found that tunnel excavation reduced vegetation growth rates for 15 y after construction (Zheng et al., 2017). The isotopic analysis (${\text{Î́}}^{\text{2}}\text{H}$ and ${\text{Î́}}^{\text{18}}\text{O}$) conducted by Liu et al. (2019) indicated that groundwater decline caused by tunnel drainage led to a transformation in plant water absorption patterns. Remote sensing techniques have also been utilized to analyze changes in vegetation coverage before and after tunnel construction (Fais and Nino, 2004, Wei and Pan, 2011). However, the heavy dependence on monitoring or field tests makes these approaches less applicable during the planning and design stages of tunnel projects.

Before excavation, the eco-environmental influences of tunnel projects are often evaluated through environmental impact assessments (EIA), strategic environmental assessments (SEA), life cycle assessments (LCA), and sustainability assessments (SA). For example, Namin et al., 2014, Phillips, 2016 considered tunnel-caused interference with surface/underground water in an EIA. Zhang et al. (2012) adopted the dewatering funnel area and surface runoff as indexes to evaluate the water ecological effect of tunnel engineering. Liu et al. (2015) combined geological, hydrological, and tunnel indicators to assess the negative effects of tunneling on the groundwater environment. Xu et al. (2016) analyzed the impact of groundwater level drawdown caused by underground works on the geological environment based on the view of SEA, which can evaluate the environmental impacts of policies, plans, and programs for underground infrastructure at a strategic level (Bobylev, 2006). Moreover, Huang et al., 2015, Audi et al., 2020 evaluated the environmental impacts (e.g., global climate and human toxicity) of each phase in the whole-life cycle of a tunnel following the principles of LCA. Some SA cases for tunneling or underground space development have also included environmental factors such as greening, vegetation community restoration, biodiversity, and water resources/supply (Qiu et al., 2020, Zargarian et al., 2016).

Although many assessments have considered the eco-environmental influences of tunnel projects, to the best of our knowledge, few have focused on the tunnel-induced impacts on vegetation. Moreover, most previous assessments have understated the hydraulic connections and ecological feedback between underground structures, groundwater, soil, plants, and the atmosphere. Hence, the integrated tunnel-vegetation assessment framework in this study may expand the application scope of environmental assessments (EA) in tunnel engineering and improve the theoretical system of existing EA methods.

To quantify the impact of tunnel drainage on vegetation, Gokdemir et al. (2019) proposed an assessment method that integrated the tunnel factor into the soil–plant–atmosphere continuum (SPAC). SPAC regards the soil, plants, and atmosphere as a dynamic physical system in which water transfers freely between components until a balance is reached (Philip, 1966). However, classical SPAC models have paid little attention to the hydraulic influences of subsurface structures on the environment, making it difficult for them to evaluate the impact of tunnel drainage on vegetation (Gokdemir et al., 2019, Li et al., 2020). To introduce the tunnel factor into the SPAC, Gokdemir et al. (2019) input the maximum stable groundwater level after tunnel construction into a one-dimensional (1D) hydrological model as the bottom boundary condition. By failing to account for the regional groundwater environment, the method only evaluated ecological influences at specific sites where the maximum groundwater drawdown occurred instead of over the entire research region. In addition, the method did not consider groundwater fluctuations during tunneling.

This study followed the tunnel-SPAC integration scheme and extended the original assessment method to a regional framework using a quasi-3D hydrological numerical model. The quasi-3D model coupled the 1D unsaturated flow and 3D groundwater seepage (Abbott et al., 1986, Seo et al., 2007, Mao et al., 2019) based on two developed hydrological models: Hydrus-1D (Šimůnek et al., 2009) and Modflow-NWT (Niswonger et al., 2011). The topsoil model (Hydrus-1D) simulated the vertical unsaturated flow and dynamic root wilting within the vadose zone. The groundwater model (Modflow-NWT) simulated the 3D groundwater seepage affected by tunnel drainage. The coupling between the topsoil and groundwater models was conducted using a one-way method (Zeng et al., 2019), which delivered the groundwater levels to the topsoil model as bottom boundary conditions (BCs) (Hanson et al., 2014, Markstrom et al., 2008, Xu et al., 2012). In addition, we used atmospheric BCs at the top of the quasi-3D model to represent the moisture exchange between the atmosphere and land. The atmospheric BCs were calculated based on the Penman–Monteith equation and varied with the vegetation growth stages. The entire research area was divided into multiple blocks, and in each block, the vertical distribution of the soil matric potential (h) was computed using the 1D topsoil model. The framework evaluated the regional vegetation vulnerability status through a blockwise comparison between h and the wilting point.

The assessment framework was applied to an actual highway tunnel to validate its applicability. The results indicated that this framework had several contributions:

• The framework can help compare and optimize tunnel design/construction parameters (such as drainage design criteria and excavation speeds) to reduce the influence of tunnel drainage on vegetation.

• The framework can evaluate the regional impact of tunnel drainage on vegetation during or after construction. Both normal and extreme atmospheric conditions are included in the assessments.

• The framework can elucidate the interaction between environmental factors and determine which factors control vegetation survival.

The remainder of this paper is organized as follows: Section 2 introduces the tunnel-SPAC integration and the criteria used to evaluate the vulnerability status of vegetation. Section 3 presents the regional assessment framework and focuses on the quasi-3D numerical model. The case study of the assessment framework is presented in Section 4, followed by a summary and conclusions in Section 5.