Contents lists available at ScienceDirect Catena journal homepage: www.elsevier.com/locate/catena A simplified probabilistic analysis of water content and wilting in soil vegetated with non-crop species Ankit Garg a , Budhaditya Hazra b , Hong Zhu c, ⁎ , Yangping Wen d a Department of Civil and Environmental Engineering, Shantou University, Guangdong 515063, China b Department of Civil Engineering, IIT Guwahati, Guwahati, Assam, India c Department of Civil and Environmental Engineering, Hong Kong University of Science & Technology, Hong Kong, SAR, China d Institute of Functional Materials and Agricultural Applied Chemistry/Collaborative Innovation Center of Jiangxi Typical Trees Cultivation and Utilization, Jiangxi Agricultural University, Nanchang 330045, China ARTICLE INFO Keywords: Probabilistic analysis Non-crop species Vegetation wilting Volumetric soil water content ABSTRACT The long-term performance of green infrastructures (e.g., vegetated slopes, green roof, bio-retention systems, etc.) can be affected by the wilting of vegetation (i.e., tendency of evapotranspiration (ET) to be nil) attributed to the decrease in volumetric soil water content. Therefore, it is important to explore an approach to investigate the wilting time of vegetation so that mitigation measures can be taken to ensure the serviceability of green in- frastructure. The technical note aims to 1) conduct probabilistic analysis of volumetric soil water content in slopes vegetated with non-crop tree and grass species; and 2) further explore this analysis method for estimating probability of wilting (tendency of transpiration becoming nil) of the selected vegetation species under natural climate conditions. For this investigation, two non-crop species namely, Ivy tree (Schefflera heptaphylla) and Bermuda grass (Cynodon dactylon), commonly grown in subtropical regions in world were selected. A normal distribution was found as the best fit to the measured volumetric water content under the impact of both species. The estimated mean value was found significantly higher in Cynodon dactylon slope compared with that of Schefflera heptaphylla slope, while the coefficient of variation among the species is similar. The probability of wilting of C. dactylon is found lower than that of Schefflera heptaphylla. In the present study, the significance of using probabilistic approach for characterizing the distribution of volumetric soil water content to assess the reliability of vegetated infrastructure was highlighted. The study is useful for estimating the behaviour of ve- getation for better planning of restoration and rehabilitation measures. This differs from those in agricultural fields with the usually shorter period for monitoring of vegetation growth (mostly within 3 months), and they are usually subjected to controlled manual irrigation. 1. Introduction Ensuring long-term growth of vegetation is critical for the service- ability of green infrastructures (e.g., vegetated slopes, green roof, bio- retention systems, etc.). One of the most critical periods during the long-term maintenance of green infrastructure is the period of drought. During the drought, vegetation is very likely to wilt (i.e., tendency of evapotranspiration (ET) to be nil) since the volumetric soil water con- tent drops under a certain threshold value (i.e., wilting point) (Feddes et al., 1978). Beyond this wilting point, the evapotranspiration (ET) will be inhibited, which may lower the performance of a given green in- frastructure. Hence, the quantification of volumetric soil water content under natural atmospheric variation is highly conducive to many future probabilistic assessment of drought and any design/analysis of green infrastructure. To evaluate volumetric soil water content or soil matric potential under various land covers, many studies have been conducted in the field (Jackson et al., 1982; Michot et al., 2003; Garg and Ng, 2015; Garg et al., 2015a, 2015b, 2015c; Morgan et al., 2018). In most of these studies, deterministic approaches have been employed. This approach may not help to access the critical stage where vegetation is likely to wilt (i.e., tendency of transpiration to be nil). Moreover, both soil matric suction and water content are highly variable due to natural atmospheric variation (i.e., relative humidity, air temperature, wind speed and radiant energy) (Allen et al., 1998), thereby making de- terministic approaches difficult to perform. Previous works concerned with the field monitoring of crops showed significant variability in the values of volumetric soil water content (Michot et al., 2003; Morgan https://doi.org/10.1016/j.catena.2018.12.016 Received 28 January 2018; Received in revised form 27 November 2018; Accepted 8 December 2018 ⁎ Corresponding author. E-mail address: zhuhong@connect.ust.hk (H. Zhu). Catena 175 (2019) 123–131 0341-8162/ © 2018 Elsevier B.V. All rights reserved. T