Corals in the Persian/Arabian Gulf are the most thermally tolerant in the world, but live very near the thresholds of their thermal tolerance. Warming sea temperatures associated with climate change have resulted in numerous coral bleaching events regionally since the mid-1990s, but it has been unclear why unusually warm sea temperatures occur some years but not others. Using a combination of 5 years of observed sea-bottom temperatures at three reef sites and a meteorologically linked hydrodynamic model that extends through the past decade, we show that summer sea-bottom temperatures are tightly linked to regional wind regimes, and that strong 'shamal' wind events control the occurrence and severity of bleaching. Sea bottom temperatures were primarily controlled by latent heat flux from wind-driven surface evaporation which exceeded 300 W m −2 during shamal winds, double that of typical breeze conditions. Daily temperature change was highly correlated with wind speed, with breeze winds (<4 m s −1) resulting in increased warming, while faster winds caused cooling, with the magnitude of temperature decline increasing with wind speed. Using observed and simulated data from 2012 to 2017, we show that years with reported bleaching events (2012, 2017) were characterized by low winds speeds that resulted in temperatures persisting above coral bleaching threshold temperatures for >5 weeks, while the cooler intervening years (2013-2016) had summers with more frequent and/or strong shamal events which repeatedly cooled temperatures below bleaching thresholds for days to weeks, providing corals temporary respite from thermal stress. Using observed data from 2012 onward and simulated data from 2008 to 2011, we show that the severity of bleaching events over the past decade was linked to both the number of cumulative days above bleaching thresholds (rather than total days, which obfuscates the cooling effects of occasional brief shamal events), as well as to maxima. We show that winds of 4 m s −1 represents a critical threshold for whether or not corals cross bleaching threshold temperatures, and provide simulations to forecast sea-bottom temperature change and recovery times under a range of wind conditions. The role that wind-driven cooling may play on coral reefs globally is discussed.

No abstract

In this work, we present the construction of a multilayered PtSe2/Ge heterostructure-based photodetectors array comprising 1×10 device units operating in the short-wavelength infrared (SWIR) spectrum region. The as-fabricated heterostructures show...

This review provides an overview of the basic concepts and operation mechanisms of ultraviolet (UV) photodetectors (PDs), the main research status, and future outlooks of II–VI group compound semiconductor-based UVPDs.

No abstract

Federated learning (FL) is experiencing a fast booming with the wave of distributed machine learning and ever-increasing privacy concerns. In the FL paradigm, global model aggregation is handled by a centralized aggregate server based on local updated gradients trained on local nodes, which mitigates privacy leakage caused by the collection of sensitive information. With the increased computing and communicating capabilities of edge and IoT devices, applying FL on heterogeneous devices to train machine learning models becomes a trend. The synchronous aggregation strategy in the classic FL paradigm cannot effectively use the resources, especially on heterogeneous devices, due to its waiting for straggler devices before aggregation in each training round. Furthermore, in real-world scenarios, the disparity of data dispersed on devices (i.e. data heterogeneity) downgrades the accuracy of models. As a result, many asynchronous FL (AFL) paradigms are presented in various application scenarios to improve efficiency, performance, privacy, and security. This survey comprehensively analyzes and summarizes existing variants of AFL according to a novel classification mechanism, including device heterogeneity, data heterogeneity, privacy and security on heterogeneous devices, and applications on heterogeneous devices. Finally, this survey reveals rising challenges and presents potentially promising research directions in this under-investigated field. CCS Concepts: • Computer systems organization → Distributed architectures; • Computing methodologies → Learning paradigms; • Networks → Mobile networks.

The fast proliferation of edge computing devices brings an increasing growth of data, which directly promotes machine learning (ML) technology development. However, privacy issues during data collection for ML tasks raise extensive concerns. To solve this issue, synchronous federated learning (FL) is proposed, which enables the central servers and end devices to maintain the same ML models by only exchanging model parameters. However, the diversity of computing power and data sizes leads to a significant difference in local training data consumption, and thereby causes the inefficiency of FL. Besides, the centralized processing of FL is vulnerable to single-point failure and poisoning attacks. Motivated by this, we propose an innovative method, federated learning with asynchronous convergence (FedAC) considering a staleness coefficient, while using a blockchain network instead of the classic central server to aggregate the global model. It avoids real-world issues such as interruption by abnormal local device training failure, dedicated attacks, etc. By comparing with the baseline models, we implement the proposed method on a real-world dataset, MNIST, and achieve accuracy rates of 98.96% and 95.84% in both horizontal and vertical FL modes, respectively. Extensive evaluation results show that FedAC outperforms most existing models.

No abstract