Tree identity and diversity directly affect soil moisture and temperature but not soil carbon ten years after planting.

Soil C is the largest C pool in forest ecosystems that contributes to C sequestration and mitigates climate change. Tree diversity enhances forest productivity, so diversifying the tree species composition, notably in managed forests, could increase the quantity of organic matter being transferred to soils and alter other soil properties relevant to the C cycle.A ten-year-old tree diversity experiment was used to study the effects of tree identity and diversity (functional and taxonomic) on soils. Surface (0-10 cm) mineral soil was repeatedly measured for soil C concentration, C:N ratio, pH, moisture, and temperature in twenty-four tree species mixtures and twelve corresponding monocultures (replicated in four blocks).Soil pH, moisture, and temperature responded to tree diversity and identity. Greater productivity in above- and below-ground tree components did not increase soil C concentration. Soil pH increased and soil moisture decreased with functional diversity, more specifically, when species had different growth strategies and shade tolerances. Functional identity affected soil moisture and temperature, such that tree communities with more slow-growing and shade-tolerant species had greater soil moisture and temperature. Higher temperature was measured in communities with broadleaf-deciduous species compared to communities with coniferous-evergreen species.We conclude that long-term soil C cycling in forest plantations will likely respond to changes in soil pH, moisture, and temperature that is mediated by tree species composition, since tree species affect these soil properties through their litter quality, water uptake, and physical control of soil microclimates.

Article-level classification of scientific publications: A comparison of deep learning, direct citation and bibliographic coupling.

Classification schemes for scientific activity and publications underpin a large swath of research evaluation practices at the organizational, governmental, and national levels. Several research classifications are currently in use, and they require continuous work as new classification techniques becomes available and as new research topics emerge. Convolutional neural networks, a subset of "deep learning" approaches, have recently offered novel and highly performant methods for classifying voluminous corpora of text. This article benchmarks a deep learning classification technique on more than 40 million scientific articles and on tens of thousands of scholarly journals. The comparison is performed against bibliographic coupling-, direct citation-, and manual-based classifications-the established and most widely used approaches in the field of bibliometrics, and by extension, in many science and innovation policy activities such as grant competition management. The results reveal that the performance of this first iteration of a deep learning approach is equivalent to the graph-based bibliometric approaches. All methods presented are also on par with manual classification. Somewhat surprisingly, no machine learning approaches were found to clearly outperform the simple label propagation approach that is direct citation. In conclusion, deep learning is promising because it performed just as well as the other approaches but has more flexibility to be further improved. For example, a deep neural network incorporating information from the citation network is likely to hold the key to an even better classification algorithm.

Edaphic factors influencing vegetation colonization and encroachment on arsenical gold mine tailings near Sudbury, Ontario.

Mine tailings are found worldwide and can have significant impacts on ecosystem and human health. In this study, natural vegetation patterns on arsenical (As) gold (Au) mine tailings located in Sudbury, Ontario were assessed using transects located at the edge of the tailings and on the tailings. Vegetation communities were significantly different between the edge and open tailings areas of the site. Arsenic concentrations in both areas were extremely variable (from 285-17,567 mg/kg) but were not significantly correlated with vegetation diversity at the site. Nutrients (carbon (C), phosphorus (P)) and organic matter concentrations were associated with higher diversity and with the presence of climax vegetation on the tailings, but there were no significant relationships between tailings chemistry and vegetation indices on the edge. Encroachment onto the tailings from the edge occurred in conventional succession patterns, with a clear gradient from grasses (Agrostis gigantea) to trees such as Picea glauca. On the tailings, a nucleation pattern was visible, distinct from conventional succession. Trees and shrubs such as Betula papyrifera and Diervilla lonicera were associated with higher diversity and higher nutrient concentrations in the underlying tailings, whereas grasses such as A. gigantea were not. We concluded that at all areas of the site, vegetation - particularly trees - was facilitating amelioration of the underlying tailings. Despite high concentrations of As, nutrients appeared to have a greater influence than metals on vegetation diversity.