L-DONAX, a growth model of the invasive weed species, Arundo donax L. David Thornby a, * , David Spencer a , Jim Hanan b , Anna Sher a,c,d a USDA-ARS Exotic and Invasive Weeds Research Unit, Robbins Hall, 1 Shields Avenue, University of California, Davis, CA 95616, USA b Advanced Computational Modelling Centre, ARC Centre for Complex Systems, University of Queensland, St Lucia, QLD 4072, Australia c Department of Biological Sciences, University of Denver, 2190 E. Illif Avenue, Denver, CO 80208, USA d Department of Research, Herbaria and Records, Denver Botanic Gardens, 909 York St., Denver, CO 80206, USA Received 14 September 2005; received in revised form 14 June 2007; accepted 22 June 2007 Available online 12 July 2007 Abstract Arundo donax L. is a perennial reed and is an invasive weed of riparian systems in North America. A structural model (L-DONAX) of the species was constructed using L-system modelling in order to assist in understanding and demonstrating the complexities of the plant’s development and structure. The model produces a realistic number of plant components from a single rhizome segment over the course of the first year of growth, using empirical relationships derived from outdoor experiments. Biomass production is also simulated, through the use of relationships found between aerial plant portion sizes and masses. L-DONAX demonstrates that control of A. donax clumps is likely to require more than annual biomass removal, due to the bulk of biomass being present underground, and the ability of remaining rhizome or stem segments to produce large clumps quickly. The model extrapolates to years of growth beyond the first, but is found to require some re-parameterisation to improve accuracy. Published by Elsevier B.V. Keywords: Arundo donax; L-DONAX; Giant reed; L-system; Structural model; Invasive species 1. Introduction Giant reed (Arundo donax L.) is a successful invasive, non- native weed of riparian zones in California, as well as other areas in North America. A. donax is a clonal plant that expands by rhizome expansion and, in California, propagates solely through shooting of separated rhizome pieces or stem pieces (Decruyenaere and Holt, 2001; DiTomaso and Healey, 2003). A. donax is difficult to control for several reasons. Firstly, it produces large amounts of biomass rapidly, both above and below ground (Sharma et al., 1998). Secondly, it propagates very readily vegetatively, by regrowth from rhizome and stem pieces (Decruyenaere and Holt, 2001), which are easily dispersed by river currents. Finally, because a large proportion of its biomass is sequestered underground in the rhizome, removal of whole clumps is difficult. In order to more successfully manage infestations of A. donax in a given site, information is required regarding its potential invasiveness and its responses to environmental conditions and stimuli, including temperature, light, soil moisture, nutrition, and the activities of other organisms in the plant’s environment. In particular, information is needed on the relationships between environ- mental conditions and the plant’s rate of growth, including the rate of new shoot production (from rhizome segments), the rate of expansion of the new shoots, the production of branches, and the longevity of plant parts. The plant’s light environment is a key factor in the ecophysiology of the plant. Recent work in structural modelling using computational methods can help in the analysis of the effects of light and other factors on plant structures, and vice versa. Structural models have been produced for a number of significant crop plants—for example, cotton (Hanan and Hearn, 2003; Thornby et al., 2003; Hanan et al., 2005) and maize (Fournier and Andrieu, 1999). However, morphogenetic models of plants that have less immediate economic value (such as weeds) have been less frequently produced. Empirical www.elsevier.com/locate/aquabot Aquatic Botany 87 (2007) 275–284 * Corresponding author. Present address: Queensland Department of Primary Industries and Fisheries, P.O. Box 2282, Toowoomba, QLD 4350, Australia. Tel.: +61 7 46398811; fax: +61 7 46398800. E-mail address: david.thornby@dpi.qld.gov.au (D. Thornby). 0304-3770/$ – see front matter. Published by Elsevier B.V. doi:10.1016/j.aquabot.2007.06.012