University of Exeter

About: University of Exeter is a education organization based out in Exeter, United Kingdom. It is known for research contribution in the topics: Population & Context (language use). The organization has 15820 authors who have published 50650 publications receiving 1793046 citations. The organization is also known as: Exeter University & University of the South West of England.

Water-use efficiency and transpiration across European forests during the Anthropocene

Abstract: Considering the combined effects of CO2 fertilization and climate change drivers on plant physiology leads to a modest increase in simulated European forest transpiration in spite of the effects of CO2-induced stomatal closure. The Earth’s carbon and hydrologic cycles are intimately coupled by gas exchange through plant stomata1,2,3. However, uncertainties in the magnitude4,5,6 and consequences7,8 of the physiological responses9,10 of plants to elevated CO2 in natural environments hinders modelling of terrestrial water cycling and carbon storage11. Here we use annually resolved long-term δ13C tree-ring measurements across a European forest network to reconstruct the physiologically driven response of intercellular CO2 (Ci) caused by atmospheric CO2 (Ca) trends. When removing meteorological signals from the δ13C measurements, we find that trees across Europe regulated gas exchange so that for one ppmv atmospheric CO2 increase, Ci increased by ∼0.76 ppmv, most consistent with moderate control towards a constant Ci/Ca ratio. This response corresponds to twentieth-century intrinsic water-use efficiency (iWUE) increases of 14 ± 10 and 22 ± 6% at broadleaf and coniferous sites, respectively. An ensemble of process-based global vegetation models shows similar CO2 effects on iWUE trends. Yet, when operating these models with climate drivers reintroduced, despite decreased stomatal opening, 5% increases in European forest transpiration are calculated over the twentieth century. This counterintuitive result arises from lengthened growing seasons, enhanced evaporative demand in a warming climate, and increased leaf area, which together oppose effects of CO2-induced stomatal closure. Our study questions changes to the hydrological cycle, such as reductions in transpiration and air humidity, hypothesized to result from plant responses to anthropogenic emissions.

Evaporation of the planet HD 189733b observed in H I Lyman-α

Abstract: We observed three transits of the extrasolar planet HD 189733b in H i Lyman-α and in a few other lines in the ultraviolet with HST/ACS, in the search for atmospheric signatures. We detect a transit signature in the Lyman-α light curve with a transit depth of 5.05 ± 0.75%. This depth exceeds the occultation depth produced by the planetary disk alone at the 3.5σ level (statistical). Other stellar emission lines are less bright, and, taken individually, they do not show the transit signature, while the whole spectra redward of the Lyman-α line has enough photons to show a transit signature consistent with the absorption by the planetary disk alone. The transit depth's upper limits in the emission lines are 11.1% for O i λ 1305 A and 5.5% for C ii λ 1335 A lines. The presence of an extended exosphere of atomic hydrogen around HD 189733b producing 5% absorption of the full unresolved Lyman-α line flux shows that the planet is losing gas. The Lyman-α light curve is well-fitted by a numerical simulation of escaping hydrogen in which the planetary atoms are pushed by the stellar radiation pressure. We constrain the escape rate of atomic hydrogen to be between 109 and 1011 g s-1 and the ionizing extreme UV flux between 2 and 40 times the solar value (1-σ ), with higher escape rates corresponding to larger EUV flux. The best fit is obtained for dM /dt = 1010 g s-1 and an EUV flux F EUV = 20 times the solar value. HD 189733b is the second extrasolar planet for which atmospheric evaporation has been detected.

Anxiety selectively disrupts visuospatial working memory.

Abstract: On the basis of a review of the extant literature describing emotion-cognition interactions, the authors propose 4 methodological desiderata for studying how task-irrelevant affect modulates cognition and present data from an experiment satisfying them. Consistent with accounts of the hemispheric asymmetries characterizing withdrawal-related negative affect and visuospatial working memory (WM) in prefrontal and parietal cortices, threat-induced anxiety selectively disrupted accuracy of spatial but not verbal WM performance. Furthermore, individual differences in physiological measures of anxiety statistically mediated the degree of disruption. A second experiment revealed that individuals characterized by high levels of behavioral inhibition exhibited more intense anxiety and relatively worse spatial WM performance in the absence of threat, solidifying the authors' inference that anxiety causally mediates disruption. These observations suggest a revision of extant models of how anxiety sculpts cognition and underscore the utility of the desiderata.

Distance metric learning with eigenvalue optimization

Abstract: The main theme of this paper is to develop a novel eigenvalue optimization framework for learning a Mahalanobis metric. Within this context, we introduce a novel metric learning approach called DML-eig which is shown to be equivalent to a well-known eigenvalue optimization problem called minimizing the maximal eigenvalue of a symmetric matrix (Overton, 1988; Lewis and Overton, 1996). Moreover, we formulate LMNN (Weinberger et al., 2005), one of the state-of-the-art metric learning methods, as a similar eigenvalue optimization problem. This novel framework not only provides new insights into metric learning but also opens new avenues to the design of efficient metric learning algorithms. Indeed, first-order algorithms are developed for DML-eig and LMNN which only need the computation of the largest eigenvector of a matrix per iteration. Their convergence characteristics are rigorously established. Various experiments on benchmark data sets show the competitive performance of our new approaches. In addition, we report an encouraging result on a difficult and challenging face verification data set called Labeled Faces in the Wild (LFW).