The warm center of the 2015/2016 extreme El Niño was located further to the west compared to previous extreme events, causing vastly different social‐economic impacts. Though many studies ...investigated ocean‐atmosphere couplings responsible for its large magnitude, the dynamics that govern the unique structure of the event remain elusive. Here we show that a local ocean‐atmosphere feedback among convection‐induced surface winds, oceanic upwelling and sea surface temperature in the far‐eastern Pacific, resulted in a boreal summer cooling tendency along the South American coast. It canceled warming in the far‐eastern equatorial Pacific that would otherwise develop, leading to the distinctive anomaly pattern of the 2015/2016 El Niño. These processes were initially triggered by the leftover anomalous warm sea surface temperature and positive moist static energy over the northeastern tropical Pacific in the following boreal spring of the 2014 warm event.
Plain Language Summary
Extreme El Niño events exert severe impacts on society, but predicting extreme events is challenging. These events are rare and distinctive from one to another in their spatial pattern, generating different regional impacts and disasters. In particular, the 2015/2016 extreme El Niño features a more westward warm center of sea surface temperature anomalies, compared to that of 1982/1983 or 1997/1998. Our analysis of observations relates this unique anomaly pattern to the preceding warm tropical Pacific in 2014. That is, rainfall anomalies over the eastern Pacific were initially fueled by the leftover unstable energy over the northeastern tropical Pacific. Then the center of anomalous rainfall was restricted to the north of the equator through a local ocean‐atmosphere feedback among convection‐induced surface flows, oceanic upwelling and sea surface temperature. The positive feedback resulted in a summer cooling tendency along the South American coast, which offset warm SST anomalies and caused the unusual pattern of the 2015/2016 extreme El Niño. Our further investigation suggested that the state of the eastern Pacific before the developing summer is also essential for the development of other El Niño events.
Key Points
The South American coastal cooling process led to a westward propagation of the 2015/2016 El Niño warm center after the boreal spring
Convective activity relating to the 2014 warm condition induced cooling tendency along the South American coast through a feedback loop
This antecedent state of the eastern Pacific played a crucial role in the unique spatial pattern of the 2015/2016 El Niño
Heterostructured catalysts show outstanding performance in electrochemical reactions owing to their beneficial interfacial properties. However, the rational design of heterostructured catalysts with ...the desired interfacial properties and charge‐transfer characteristics is challenging. Herein, we developed a SrMn3O6−x‐SrMnO3 (SMOx‐SMO) heterostructure through epitaxial growth, which demonstrated excellent electrocatalyst performance for the oxygen reduction reaction (ORR). The formation of high‐valence Mn3+/4+ is beneficial for promoting a positive shift in the position of the d‐band center, thereby optimizing the adsorption and desorption of ORR intermediates on the heterojunction surface and resulting in improved catalytic activity. When SMOx‐SMO was applied as an air‐electrode catalyst in a rechargeable zinc‐air battery, a high output voltage and power density was achieved, with performance comparable to a battery prepared with Pt/C‐IrO2 air‐electrode catalysts, albeit with much better cycling stability.
An epitaxially grown SrMn3O6−x‐SrMnO3 heterostructure has been developed for the oxygen reduction reaction (ORR). Owing to the effect of interfacial chemistry on ORR performance, the formation of high‐valence Mn3+/4+ is beneficial in promoting the positive shift of the d‐band center, thereby optimizing the adsorption and desorption of ORR intermediates on the heterojunction surface and improving catalytic activity.
Both Xp11 translocation renal cell carcinomas and the corresponding mesenchymal neoplasms are characterized by a variety of gene fusions involving TFE3. It has been known that tumors with different ...gene fusions may have different clinicopathologic features; however, further in-depth investigations of subtyping Xp11 translocation-associated cancers are needed in order to explore more meaningful clinicopathologic correlations. A total of 22 unusual cases of Xp11 translocation-associated cancers were selected for the current study; 20 cases were further analyzed by RNA sequencing to explore their TFE3 gene fusion partners. RNA sequencing identified 17 of 20 cases (85%) with TFE3-associated gene fusions, including 4 ASPSCR1/ASPL-TFE3, 3 PRCC-TFE3, 3 SFPQ/PSF-TFE3, 1 NONO-TFE3, 4 MED15-TFE3, 1 MATR3-TFE3, and 1 FUBP1-TFE3. The results have been verified by fusion fluorescence in situ hybridization (FISH) assays or reverse transcriptase polymerase chain reaction (RT-PCR). The remaining 2 cases with specific pathologic features highly suggestive of MED15-TFE3 renal cell carcinoma were identified by fusion FISH assay. We provide the detailed morphologic and immunophenotypic description of the MED15-TFE3 renal cell carcinomas, which frequently demonstrate extensively cystic architecture, similar to multilocular cystic renal neoplasm of low malignant potential, and expressed cathepsin K and melanotic biomarker Melan A. This is the first time to correlate the MED15-TFE3 renal cell carcinoma with specific clinicopathologic features. We also report the first case of the corresponding mesenchymal neoplasm with MED15-TFE3 gene fusion. Additional novel TFE3 gene fusion partners, MATR3 and FUBP1, were identified. Cases with ASPSCR1-TFE3, SFPQ-TFE3, PRCC-TFE3, and NONO-TFE3 gene fusion showed a wide variability in morphologic features, including invasive tubulopapillary pattern simulating collecting duct carcinoma, extensive calcification and ossification, and overlapping and high columnar cells with nuclear grooves mimicking tall cell variant of papillary thyroid carcinoma. Furthermore, we respectively evaluated the ability of TFE3 immunohistochemistry, TFE3 FISH, RT-PCR, and RNA sequencing to subclassify Xp11 translocation-associated cancers. In summary, our study expands the list of TFE3 gene fusion partners and the clinicopathologic features of Xp11 translocation-associated cancers, and highlights the importance of subtyping Xp11 translocation-associated cancers combining morphology, immunohistochemistry, and multiple molecular techniques.
The fast development of synthesis routes and preparation technology of 2D materials has motivated a rapid growth in the micro‐ and nanoelectronic memory devices, which gives rise to the breakthroughs ...in the semiconductor research area. Hexagon boron nitride (h‐BN) with excellent chemical, mechanical, and optical properties has been proven to have potential in overcoming the scaling limit to nanometer, and even sub‐nanometer lengths to replace the use of thick and stiff blocking dielectrics in two‐terminal or three‐terminal devices. The use of atomically thin h‐BN or h‐BN van der Waals heterostructures (vdWhs) can improve the reliability, capability, and functionality of memory devices. This is an encouraging strategy toward high‐density on‐chip integrated circuits, which has recently earned considerable interest. While the research in h‐BN material properties and characterization is comprehensively verified, specified mechanisms of resistive switching have not been analyzed in‐depth. Moreover, recent concern about novel structure design and expanding applications in electronics, optoelectronics, and spintronics has arisen. In this review, recent progress in h‐BN memories with volatile or nonvolatile properties is presented, expanding the memories to functional applications, and further challenges of the development of h‐BN‐based memories and logic circuits are discussed.
Hexagon boron nitride (h‐BN) with excellent chemical, mechanical, and optical properties has the potential to replace the use of thick and stiff blocking dielectrics in two‐terminal or three‐terminal devices. Recent progress in h‐BN memories with volatile or nonvolatile properties is presented, expanding the memories to functional applications, and further challenges in the development of h‐BN‐based memories are discussed.
Learning generic and robust feature representations with data from multiple domains for the same problem is of great value, especially for the problems that have multiple datasets but none of them ...are large enough to provide abundant data variations. In this work, we present a pipeline for learning deep feature representations from multiple domains with Convolutional Neural Networks (CNNs). When training a CNN with data from all the domains, some neurons learn representations shared across several domains, while some others are effective only for a specific one. Based on this important observation, we propose a Domain Guided Dropout algorithm to improve the feature learning procedure. Experiments show the effectiveness of our pipeline and the proposed algorithm. Our methods on the person re-identification problem outperform stateof-the-art methods on multiple datasets by large margins.
In March 2017, sea surface temperatures off Peru rose above 28 °C, causing torrential rains that affected the lives of millions of people. This coastal warming is highly unusual in that it took place ...with a weak La Niña state. Observations and ocean model experiments show that the downwelling Kelvin waves caused by strong westerly wind events over the equatorial Pacific, together with anomalous northerly coastal winds, are important. Atmospheric model experiments further show the anomalous coastal winds are forced by the coastal warming. Taken together, these results indicate a positive feedback off Peru between the coastal warming, atmospheric deep convection, and the coastal winds. These coupled processes provide predictability. Indeed, initialized on as early as 1 February 2017, seasonal prediction models captured the extreme rainfall event. Climate model projections indicate that the frequency of extreme coastal El Niño will increase under global warming.
The state-of-the-art performance for object detection has been significantly improved over the past two years. Besides the introduction of powerful deep neural networks, such as GoogleNet and VGG, ...novel object detection frameworks, such as R-CNN and its successors, Fast R-CNN, and Faster R-CNN, play an essential role in improving the state of the art. Despite their effectiveness on still images, those frameworks are not specifically designed for object detection from videos. Temporal and contextual information of videos are not fully investigated and utilized. In this paper, we propose a deep learning framework that incorporates temporal and contextual information from tubelets obtained in videos, which dramatically improves the baseline performance of existing still-image detection frameworks when they are applied to videos. It is called T-CNN, i.e., tubelets with convolutional neueral networks. The proposed framework won newly introduced an object-detection-from-video task with provided data in the ImageNet Large-Scale Visual Recognition Challenge 2015. Code is publicly available at https://github.com/myfavouritekk/T-CNN .
Genome analysis is one of the main criteria for description of new taxa. Availability of genome sequences for all the actinobacteria with a valid nomenclature will, however, require another decade's ...works of sequencing. This paper describes the rearrangement of the higher taxonomic ranks of the members of the phylum '
', using the phylogeny of 16S rRNA gene sequences and supported by the phylogeny of the available genome sequences. Based on the refined phylogeny of the 16S rRNA gene sequences, we could arrange all the members of the 425 genera of the phylum '
' with validly published names currently in use into six classes, 46 orders and 79 families, including 16 new orders and 10 new families. The order
Prévot 1940 (Approved Lists 1980) emend. Nouioui
. 2018 is now split into 11 monophyletic orders: the emended order
and ten proposed new orders
,
,
,
,
,
,
,
,
and
. Further, the class '
' Stackebrandt
. 1997 emend. Nouioui
. 2018 was described without any nomenclature type, and therefore the name '
' is deemed illegitimate. In accordance to Rule 8 of the International Code of Nomenclature of Prokaryotes, Parker
. 2019, we proposed the name
which is formed by using the stem of the name
Buchanan 1917 (Approved Lists 1980) emend. Zhi
. 2009, to replace the name '
'. The nomenclature type of the proposed new class
is the order
Buchanan 1917 (Approved Lists 1980) emend. Zhi
. 2009.
The eastern tropical Pacific features strong climatic asymmetry across the equator, with the intertropical convergence zone (ITCZ) displaced north of the equator most of time. In February–April ...(FMA), the seasonal warming in the Southern Hemisphere and cooling in the Northern Hemisphere weaken the climatic asymmetry, and a double ITCZ appears with a zonal rainband on either side of the equator. Results from an analysis of precipitation variability reveal that the relative strength between the northern and southern ITCZ varies from one year to another and this meridional seesaw results from ocean–atmosphere coupling. Surprisingly this meridional seesaw is triggered by an El Niño–Southern Oscillation (ENSO) of moderate amplitudes. Although ENSO is originally symmetric about the equator, the asymmetry in the mean climate in the preceding season introduces asymmetric perturbations, which are then preferentially amplified by coupled ocean–atmosphere feedback in FMA when deep convection is sensitive to small changes in cross-equatorial gradient of sea surface temperature. This study shows that moderate ENSO follows a distinct decay trajectory in FMA and southeasterly cross-equatorial wind anomalies cause moderate El Niño to dissipate rapidly as southeasterly cross-equatorial wind anomalies intensify ocean upwelling south of the equator. In contrast, extreme El Niño remains strong through FMA as enhanced deep convection causes westerly wind anomalies to intrude and suppress ocean upwelling in the eastern equatorial Pacific.
Deep learning-based Compressed Sensing (CS) reconstruction attracts much attention in recent years, due to its significant superiority of reconstruction quality. Its success is mainly attributed to ...the employment of a large dataset for pre-training the network to learn a reconstruction mapping. In this paper, we propose a non-locally regularized compressed sensing network for reconstructing image sequences, which can achieve high reconstruction quality without pre-training. Specifically, the proposed method attempts to learn a deep network prior for the reconstruction of an individual instance under the constraint that the network output can well match the given CS measurement. The non-local prior is designed to guide the network to capture the long-range dependencies by exploiting the self-similarities among images, and it can also make the network noise-aware. In order to deal with the compound of non-local prior and deep network prior, we construct a half-quadratic splitting based optimization method for network learning, in which the two priors are decoupled into two simple sub-problems by introducing an auxiliary variable and a quadratic fidelity constraint. Extensive experimental results demonstrate that our method is competitive to the popular methods, including sparsity prior based methods and deep learning based methods, even better than them in the cases of low measurement rates.
PDF
