Seafloor primary production in a changing Arctic Ocean.

Phytoplankton and sea ice algae are traditionally considered to be the main primary producers in the Arctic Ocean. In this Perspective, we explore the importance of benthic primary producers (BPPs) encompassing microalgae, macroalgae, and seagrasses, which represent a poorly quantified source of Arctic marine primary production. Despite scarce observations, models predict that BPPs are widespread, colonizing ~3 million km2 of the extensive Arctic coastal and shelf seas. Using a synthesis of published data and a novel model, we estimate that BPPs currently contribute ~77 Tg C y-1 of primary production to the Arctic, equivalent to ~20 to 35% of annual phytoplankton production. Macroalgae contribute ~43 Tg C y-1, seagrasses contribute ~23 Tg C y-1, and microalgae-dominated shelf habitats contribute ~11 to 16 Tg C y-1. Since 2003, the Arctic seafloor area exposed to sunlight has increased by ~47,000 km2 y-1, expanding the realm of BPPs in a warming Arctic. Increased macrophyte abundance and productivity is expected along Arctic coastlines with continued ocean warming and sea ice loss. However, microalgal benthic primary production has increased in only a few shelf regions despite substantial sea ice loss over the past 20 y, as higher solar irradiance in the ice-free ocean is counterbalanced by reduced water transparency. This suggests complex impacts of climate change on Arctic light availability and marine primary production. Despite significant knowledge gaps on Arctic BPPs, their widespread presence and obvious contribution to coastal and shelf ecosystem production call for further investigation and for their inclusion in Arctic ecosystem models and carbon budgets.

Explaining Semi-Supervised Text Alignment Through Visualization.

The analysis of variance in complex text traditions is an arduous task when carried out manually. Text alignment algorithms provide domain experts with a robust alternative to such repetitive tasks. Existing white-box approaches allow the digital humanities to establish syntax-based metrics taking into account the spelling, morphology and order of words. However, they produce limited results, as semantic meanings are typically not taken into account. Our interdisciplinary collaboration between visualization and digital humanities combined a semi-supervised text alignment approach based on word embeddings that take not only syntactic but also semantic text features into account, thereby improving the overall quality of the alignment. In our collaboration, we developed different visual interfaces that communicate the word distribution in high-dimensional vector space generated by the underlying neural network for increased transparency, assessment of the tool's reliability and overall improved hypothesis generation. We further offer visual means to enable the expert reader to feed domain knowledge into the system at multiple levels with the aim of improving both the product and the process of text alignment. This ultimately illustrates how visualization can engage with and augment complex modes of reading in the humanities.

A Survey of Text Alignment Visualization.

Text alignment is one of the fundamental techniques text-related domains like natural language processing, computational linguistics, and digital humanities. It compares two or more texts with each other aiming to find similar textual patterns, or to estimate in general how different or similar the texts are. Visualizing alignment results is an essential task, because it helps researchers getting a comprehensive overview of individual findings and the overall pattern structure. Different approaches have been developed to visualize and help making sense of these patterns depending on text size, alignment methods, and, most importantly, the underlying research tasks demanding for alignment. On the basis of those tasks, we reviewed existing text alignment visualization approaches, and discuss their advantages and drawbacks. We finally derive design implications and shed light on related future challenges.

Modeling How Humans Judge Dot-Label Relations in Point Cloud Visualizations.

When point clouds are labeled in information visualization applications, sophisticated guidelines as in cartography do not yet exist. Existing naive strategies may mislead as to which points belong to which label. To inform improved strategies, we studied factors influencing this phenomenon. We derived a class of labeled point cloud representations from existing applications and we defined different models predicting how humans interpret such complex representations, focusing on their geometric properties. We conducted an empirical study, in which participants had to relate dots to labels in order to evaluate how well our models predict. Our results indicate that presence of point clusters, label size, and angle to the label have an effect on participants' judgment as well as that the distance measure types considered perform differently discouraging the use of label centers as reference points.

Predominance Tag Maps.

A predominance map expresses the predominant data category for each geographical entity and colors are used to differentiate a small number of data categories. In tag maps, many data categories are present in the form of different tags, but related tag map approaches do not account for predominance, as tags are either displaced from their respective geographical locations or visual clutter occurs. We propose predominance tag maps, a layout algorithm that accounts for predominance for arbitrary aggregation granularities. The algorithm is able to utilize the font sizes of the tags as visual variable and it is further configurable to implement aggregation strategies beyond visualizing predominance. We introduce various measures to evaluate numerically the qualitative aspects of tag maps regarding local predominance, global features, and layout stability and we comparatively analyze our method to the tag map approach by Thom et al.  [1] on the basis of real world data sets.

Visualization and the Digital Humanities.

For the past two years, researchers from the visualization community and the digital humanities have come together at the IEEE VIS conference to discuss how both disciplines can work together to push research goals in their respective disciplines. In this paper, we present our experiences as a result of this collaboration.

Interactive Visual Profiling of Musicians.

Determining similar objects based upon the features of an object of interest is a common task for visual analytics systems. This process is called profiling, if the object of interest is a person with individual attributes. The profiling of musicians similar to a musician of interest with the aid of visual means became an interesting research question for musicologists working with the Bavarian Musicians Encyclopedia Online. This paper illustrates the development of a visual analytics profiling system that is used to address such research questions. Taking musicological knowledge into account, we outline various steps of our collaborative digital humanities project, priority (1) the definition of various measures to determine the similarity of musicians' attributes, and (2) the design of an interactive profiling system that supports musicologists in iteratively determining similar musicians. The utility of the profiling system is emphasized by various usage scenarios illustrating current research questions in musicology.

Visualization of graph products.

Graphs are a versatile structure and abstraction for binary relationships between objects. To gain insight into such relationships, their corresponding graph can be visualized. In the past, many classes of graphs have been defined, e.g. trees, planar graphs, directed acyclic graphs, and visualization algorithms were proposed for these classes. Although many graphs may only be classified as "general" graphs, they can contain substructures that belong to a certain class. Archambault proposed the TopoLayout framework: rather than draw any arbitrary graph using one method, split the graph into components that are homogeneous with respect to one graph class and then draw each component with an algorithm best suited for this class. Graph products constitute a class that arises frequently in graph theory, but for which no visualization algorithm has been proposed until now. In this paper, we present an algorithm for drawing graph products and the aesthetic criterion graph product's drawings are subject to. We show that the popular High-Dimensional Embedder approach applied to cartesian products already respects this aestetic criterion, but has disadvantages. We also present how our method is integrated as a new component into the TopoLayout framework. Our implementation is used for further research of graph products in a biological context.