Candidate crops for biofuel production are generally chosen for their rapid growth rate, high resource-use efficiency (regarding water, radiation and nutrients) and broad tolerance of pests, diseases and stressful environments [104-107]. However, the traits that characterize an ideal biofuel crop are also the same traits that characterize many invasive species [93, 104-106].

The potential invasiveness of Miscanthus species has been assessed in various locations using either a weed risk assessment (WRA) or field observations (Table 4.1). WRA is a model adapted from the Australian weed risk assessment system or AWRA developed by Pheloung [108] to a local environment. The WRA uses an additive approach, with set scores ranging from -3 to 5 for each of 49 questions dealing with the species’ growth habit and persistence, distribution, reproductive system and whether it has become a weed elsewhere.

Using the WRA, M. x giganteus identifies as noninvasive (Table 4.4) due largely to its sterility and, thus, dramatically reduced risk of escaping into natural environments [92, 105, 106, 109]. This contrasts with M. sinensis, which identifies as invasive at most locations. However, not all genotypes of M. sinensis would be expected to present the same invasiveness risk, since sterile triploid clones exist within this species [110], but unfortunately the genotypes used in the M. sinensis study were not specified. Nevertheless, many invasive species do not produce fertile seeds but are serious invaders as Arundo donax, Polygonum cuspidatum and so on [104-106, 109].

In addition, M. x giganteus rhizomes have a slow rate of lateral growth (only spread at a space around 10 cm per year) compared to M. sacchariflorus, which has extensive creeping rhizomes (can spread several meters in a few years), which increased the risk of spreading due to erosion and water transport [93, 110]. Therefore, M. sacchariflorus is defined as invasive.

It is important to note that invasiveness evaluations using risk assessment systems are not absolute, as they must incorporate subjectivity and uncertainty [116]. For example, the WRA can produce incorrect answers and provides non definitive answers in almost one — third of all cases [106, 116]. Therefore, it is recommended that a combined assessment approach be used that incorporates pre-entry and post-entry evaluation tools [105,106, 116].

1. Pre-entry evaluation

• WRA to identify invaders and benign species and reject or accept them for intro­duction. The protocol used would be based on the biology of the target species and its ecology, climatic requirements, history and biogeography relative to the target regions. [106, 109]

• Climate matching analysis (e. g. CLIMEX model [119]) to determine the climatic and agronomic regions at risk of a potential invasion.

• Evaluate the potential for the proposed bioenergy crop to hybridize with related species or taxa.

2. Post entry evaluations

Any species conditionally accepted would then require in situ ecological analyses:

• Agronomic trials in quarantined field trials in the new environments.

• Long-term experiments to determine the competitiveness of the proposed bioenergy crop within native or managed ecosystems in the new environment.

• An efficient management plan covering the eradication of each feedstock in case of invasiveness after commercialization [106, 109].

Breeding and management programs will also help to minimize the invasive risk of Miscanthus [109] by minimizing seed production and rhizome spread.