(Table S1) Age determination of ODP Site 202-1240

Interannual-decadal variability in the equatorial Pacific El Niño-Southern Oscillation (ENSO) induces climate changes at global scale, but its potential influence during past global climate change is not yet well constrained. New high-resolution eastern equatorial Pacific proxy records of thermocline conditions present new evidence of strong orbital control in ENSO-like variability over the last 275,000 years. Recurrent intervals of saltier thermocline waters are associated with the dominance of La Niña-like conditions during glacial terminations, coinciding with periods of low precession and high obliquity. The parallel dominance of d13C-depleted waters supports the advection of Antarctic origin waters toward the tropical thermocline. This "oceanic tunneling" is proposed to have reinforced orbitally induced changes in ENSO-like variability, composing a complex high- and low-latitude feedback during glacial terminations.

Idiopathic multifocal choroiditis/punctate inner choroidopathy with acute photoreceptor loss or dysfunction out of proportion to clinically visible lesions.

PURPOSE To report acute/subacute vision loss and paracentral scotomata in patients with idiopathic multifocal choroiditis/punctate inner choroidopathy due to large zones of acute photoreceptor attenuation surrounding the chorioretinal lesions. METHODS Multimodal imaging case series. RESULTS Six women and 2 men were included (mean age, 31.5 ± 5.8 years). Vision ranged from 20/20-1 to hand motion (mean, 20/364). Spectral domain optical coherence tomography demonstrated extensive attenuation of the external limiting membrane, ellipsoid and interdigitation zones, adjacent to the visible multifocal choroiditis/punctate inner choroidopathy lesions. The corresponding areas were hyperautofluorescent on fundus autofluorescence and were associated with corresponding visual field defects. Full-field electroretinogram (available in three cases) showed markedly decreased cone/rod response, and multifocal electroretinogram revealed reduced amplitudes and increased implicit times in two cases. Three patients received no treatment, the remaining were treated with oral corticosteroids (n = 4), oral acyclovir/valacyclovir (n = 2), intravitreal/posterior subtenon triamcinolone acetate (n = 3), and anti-vascular endothelial growth factor (n = 2). Visual recovery occurred in only three cases of whom two were treated. Varying morphological recovery was found in six cases, associated with decrease in hyperautofluorescence on fundus autofluorescence. CONCLUSION Multifocal choroiditis/punctate inner choroidopathy can present with transient or permanent central photoreceptor attenuation/loss. This presentation is likely a variant of multifocal choroiditis/punctate inner choroidopathy with chorioretinal atrophy. Associated changes are best evaluated using multimodal imaging.

Influence of fruit turgidity and firmness on apple bruise susceptibility.

Bruise damage is a major cause of quality loss for apples. It would be very useful to establish a method of characterizing bruise susceptibility in order to improve fruit handling, sometimes Magness-Taylor firmness is used as an indirect guide to handling requirements. The objective of the present work was to achieve a better bruise susceptibility prediction.

Toward a climate-dependent paradigm of ammonia emission & deposition.

Existing descriptions of bi-directional ammonia (NH3) land–atmosphere exchange incorporate temperature and moisture controls, and are beginning to be used in regional chemical transport models. However, such models have typically applied simpler emission factors to upscale the main NH3 emission terms. While this approach has successfully simulated the main spatial patterns on local to global scales, it fails to address the environment- and climate-dependence of emissions. To handle these issues, we outline the basis for a new modelling paradigm where both NH3 emissions and deposition are calculated online according to diurnal, seasonal and spatial differences in meteorology. We show how measurements reveal a strong, but complex pattern of climatic dependence, which is increasingly being characterized using ground-based NH3 monitoring and satellite observations, while advances in process-based modelling are illustrated for agricultural and natural sources, including a global application for seabird colonies. A future architecture for NH3 emission–deposition modelling is proposed that integrates the spatio-temporal interactions, and provides the necessary foundation to assess the consequences of climate change. Based on available measurements, a first empirical estimate suggests that 5°C warming would increase emissions by 42 per cent (28–67%). Together with increased anthropogenic activity, global NH3 emissions may increase from 65 (45–85) Tg N in 2008 to reach 132 (89–179) Tg by 2100.

Efficient Monte Carlo optimization for multi-label classifier chains

Multi-label classification (MLC) is the supervised learning problem where an instance may be associated with multiple labels. Modeling dependencies between labels allows MLC methods to improve their performance at the expense of an increased computational cost. In this paper we focus on the classifier chains (CC) approach for modeling dependencies. On the one hand, the original CC algorithm makes a greedy approximation, and is fast but tends to propagate errors down the chain. On the other hand, a recent Bayes-optimal method improves the performance, but is computationally intractable in practice. Here we present a novel double-Monte Carlo scheme (M2CC), both for finding a good chain sequence and performing efficient inference. The M2CC algorithm remains tractable for high-dimensional data sets and obtains the best overall accuracy, as shown on several real data sets with input dimension as high as 1449 and up to 103 labels.

Multi-dimensional classification with super-classes

The multi-dimensional classification problem is a generalisation of the recently-popularised task of multi-label classification, where each data instance is associated with multiple class variables. There has been relatively little research carried out specific to multi-dimensional classification and, although one of the core goals is similar (modelling dependencies among classes), there are important differences; namely a higher number of possible classifications. In this paper we present method for multi-dimensional classification, drawing from the most relevant multi-label research, and combining it with important novel developments. Using a fast method to model the conditional dependence between class variables, we form super-class partitions and use them to build multi-dimensional learners, learning each super-class as an ordinary class, and thus explicitly modelling class dependencies. Additionally, we present a mechanism to deal with the many class values inherent to super-classes, and thus make learning efficient. To investigate the effectiveness of this approach we carry out an empirical evaluation on a range of multi-dimensional datasets, under different evaluation metrics, and in comparison with high-performing existing multi-dimensional approaches from the literature. Analysis of results shows that our approach offers important performance gains over competing methods, while also exhibiting tractable running time.