No abstract

PySAL is a library for geocomputation and spatial data science. Written in Python, the library has a long history of supporting novel scholarship and broadening methodological impacts far afield of academic work. Recently, many new techniques, methods of analyses, and development modes have been implemented, making the library much larger and more encompassing than that previously discussed in the literature. As such, we provide an introduction to the library as it stands now, as well as the scientific and conceptual underpinnings of its core set of components. Finally, we provide a prospective look at the library's future evolution.

Geography is an intensely visual domain. Its longstanding dependence on visualization and cartography shows as much, with John Snow's cholera map serving as one of the first instances of geovisual analytics in science (Arribas-Bel, de Graaff, & Rey, 2017; Johnson, 2007), and the perennial presence of maps as statistical displays in seminal works on visualization (Tufte, 2001). As such, the existence and continued focus on maps in geographical analysis demands serious, dedicated attention in scientific computing. However, existing methods in Python, specifically for statistical visualization of spatial data, are lacking. General-purpose mapping provided by geopandas is not fine-tuned enough for statistical analysis (Jordahl et al., 2019). The more analytically-oriented views offered by geoplot, while useful, are limited in their statistical applications (Bilogur, Karve, Marsano, & Fleischmann, 2019). Thus, the need remains for a strong, analytically-oriented toolbox for visual geographical analysis.

Biological invasions represent an increasing threat to ecosystems worldwide, with negative ecological and socio-economic impacts, whereas risk assessment and management remain challenging. The development of decision support systems (DSS) has the potential to help decision-makers and managers mitigate invasive species, but few DSS exist for forest invasive alien species (FIAS). The use of DSS in forestry is not new but they represent an asset in decision making in times of increasing complexity of issues foresters face and factors to consider. Yet, few forest DSS address the problem of FIAS. In this review, we identify key elements of the FIAS risk-assessment and management decision-making process, discuss these elements with a model-based DSS development perspective, and summarize outstanding challenges and opportunities for FIAS DSS development. FIAS DSS should not only estimate the probability of FIAS invasion but also consider forest vulnerability and quantify exposure (i.e., value at risk), while allowing different threat scenarios and possible solutions to be compared. Such a complete risk assessment and management calls for integrative modelling approaches that explicitly link different components of FIAS invasion, management, and impact assessment into a DSS. Such integrative modelling is challenging and may require collaboration among experts of different domains. International collaboration is also needed to facilitate data exchange, as the lack of data is one of the main challenges. In many cases, data and ecological knowledge of invasive species are too limited (in quantity or quality) to constitute useful input to DSS or their components (e.g., species distribution model). Another challenge is to better consider the multiple sources of uncertainties inherent to modelling invasions (e.g., host preferences and behavior, forest vulnerability, potential impacts, and cost and benefits of mitigation actions) when assessing FIAS risk and communicating results from risk assessment. Communication with stakeholders and DSS end-users, in fact, appears as one of the keys to successful DSS development and appropriation, not only to ensure that they correspond to end-users’ needs but also to ensure ease of use, functionality, and good visualization of DSS outputs.