Adaptive management studies of invasive plants on non-agricultural lands typically employ an empirical approach based on designed field experiments that permit rigorous statistical analysis of results to quantify outcomes and assess the efficacy of management practices. When habitat restoration is the primary goal of a project, traditional plot-based study designs (e.g., the randomized complete-block design) are sometimes infeasible (this is often true in aquatic habitats) or inappropriate (e.g., when the goal is to assess effects of management practices on survival or resprouting of individual plants, such as trees or shrubs). Moreover, the assumptions of distribution-specific parametric statistical methods such as ANOVA often cannot be convincingly verified or are clearly untenable when properly assessed. For these reasons, it is worthwhile to be aware of alternative study designs that do not employ plots as experimental units and nonparametric statistical methods that require only weak distributional assumptions. The purpose of this paper is to review several of these alternative study designs and nonparametric statistical methods that we have found useful in our own studies of invasive aquatic and terrestrial plants. We motivate each statistical method by a research question it is well suited to answer, provide corresponding references to the statistical literature, and identify at least one R function that implements the method. In the Supplementary Material, we present additional technical information about the statistical methods, numerical examples with data, and a set of complete R programs to illustrate application of the statistical methods.

Fast and efficient identification is critical for reducing the likelihood of weed establishment and for appropriately managing established weeds. Traditional identification tools require either knowledge of technical morphological terminology or time-consuming image matching by the user. In recent years, deep learning computer vision models have become mature enough to enable automatic identification. The major remaining bottlenecks are the availability of a sufficient number of high-quality, reliably identified training images and the user-friendly, mobile operationalization of the technology. Here, we present the first weed identification and reporting app and website for all of Australia. It includes an image classification model covering more than 400 species of weeds and some Australian native relatives, with a focus on emerging biosecurity threats and spreading weeds that can still be eradicated or contained. It links the user to additional information provided by state and territory governments, flags species that are locally reportable or notifiable, and allows the creation of observation records in a central database. State and local weed officers can create notification profiles to be alerted of relevant weed observations in their area. We discuss the background of the WeedScan project, the approach taken in design and software development, the photo library used for training the WeedScan image classifier, the model itself and its accuracy, and technical challenges and how these were overcome.