7 research outputs found
Applications of stacking/blending ensemble learning approaches for evaluating flash flood susceptibility
No full textFlash floods are a type of catastrophic disasters which cause significant losses of life and property worldwide. In recent years, machine learning techniques have become powerful tools for evaluating flash flood susceptibility. This research applies stacking and blending ensemble learning approaches to assess the flash flood potential in Jiangxi, China. Four base models – linear regression, K-nearest neighbours, support vector machine, and random forest – are adopted to build the two ensemble models. All models are evaluated by three metrics (accuracy, true positive rate, and the area under the receiver operating characteristic curve) and compared with a Bayesian approach. The results suggest that the blending approach is superior to all the other models, which has then been selected to evaluate the vulnerability of flash floods for all the catchments in Jiangxi. The derived maps of flash flood susceptibility suggest that over half of the province, in terms of either area or the number of catchments, are prone to flash floods, in particular the north, northeast and south. These empirical findings can help to develop plans for disaster prevention and control, as well as improving public knowledge of flash flood hazards
Characteristics of Speciation of Heavy Metals in Municipal Sewage Sludge of Guangzhou as Fertilizer
No full textAbstractApplication of municipal sewage sludge to agricultural land especially in shallow soils, demands to validate the maximum amount of heavy metal, monitoring its effects on soil and on plants. The use of sewage sludge as a fertilizer and soil amendment has resulted in high concentrations of heavy metals in the soil and food limiting its use. Controlling the pollution of heavy metals is the key factor to realize the safe utilization of sewage sludge. In the present study, the heavy metal stabilizers were added to sludge contained in used plastic containers, through artificially watering or naturally rain falling, the nutrient components flowed out with leaching water and fertilized plants but the heavy metals retained in the sludge within container. The potential toxic risks from heavy metals of sludge depend on their chemical speciation. The contents of heavy metals in different treatments were analyzed and their speciation was determined. The pot experiments with plants (Zea mays and Laetuca satiuali) showed that the positive effects of the mixture of the sludge and K2SO4 on plant production and reduction of heavy metal contents in plants were significant. The BCR sequential extraction procedure was applied for measurement of heavy metals in the experimental sludge. The results showed that the concentrations of Zn were predominant in acid exchangeable and reducible fractions, and Cu was principally distributed in oxidizable fractions. Metals-bound sludge could be collected easily after treatment to prevent the secondary pollution, provided the heavy metals were fixed within the container and reduced obviously the leaching of heavy metals to soil
Flavor anomalies at the LHC and the R-parity violating supersymmetric model extended with vectorlike particles
No full textIn this paper, we consider a solution to explain the three discrepancies with the standard model (SM) predictions in flavor observables, i.e., anomalies in B -> K*mu(+)mu(-) and R-K = B(B -> K mu(+)mu(-))/B(B -> Ke(+)e(-)) at the LHCb and an excess in h -> mu tau at the CMS in the context of R-parity violating (RPV) supersymmetry. We demonstrate that these anomalies can be explained within a unified framework: the minimal supersymmetry model (MSSM) extended with 5 + (5) over bar vectorlike (VL) particles. The new trilinear RPV couplings involving VL particles in our model can solve the b -> s anomalies, and the mixing between the SM-like Higgs boson and the VL sneutrino can yield the extra h -> mu tau decay mode.SCI(E)[email protected]; [email protected]
Axial wind effects on stratification and longitudinal sediment transport in a convergent estuary during wet season
No full textAuthor Posting. © American Geophysical Union, 2020. This article is posted here by permission of American Geophysical Union for personal use, not for redistribution. The definitive version was published in Journal of Geophysical Research-Oceans 125(2), (2020): e2019JC015254, doi:10.1029/2019JC015254.The Coupled Ocean‐Atmosphere‐Wave‐Sediment Transport (COAWST) modeling system was used to examine axial wind effects on vertical stratification and sediment transport in a convergent estuary. The model demonstrated that stratification dynamics in the upper estuary (Kelvin number <1; Ke= fB/√ g'hs) are dominated by longitudinal wind straining, whereas the dominant mechanism governing estuarine stratification in the lower estuary (Kelvin number ~1) is lateral wind straining. Barotropic advection contributes to seaward sediment transport and peaks during spring tides, whereas estuarine circulation causes landward sediment transport with a maximum during neap tides. Down‐estuary winds impose no obvious effects on longitudinal sediment flux, whereas up‐estuary winds contribute to enhanced seaward sediment flux by increasing the tidal oscillatory flux. The model also demonstrates that bottom friction is significantly influenced by vertical stratification over channel regions, which is indirectly affected by axial winds.This research was funded by the National Natural Science Foundation of China (Grants 41576089, 51761135021, and 41890851), the National Key Research and Development Program of China (2016YFC0402603) and the Guangdong Provincial Water Conservancy Science and Technology Innovation Project (Grant 201719). We thank Professor Liangwen Jia at the Sun Yat‐sen University for his kindly providing the surficial sediment samples data in 2011. We also thank graduate students Guang Zhang and Yuren Chen from the Sun Yat‐sen University for their help in data analysis. We are grateful to two anonymous reviewers for their insightful comments to help improve this manuscript. The data related to this article is available online at the Zenodo website (https://zenodo.org/record/3606471).2020-07-1
Multiscale temporal response of salt intrusion to transient river and ocean forcing
No full textSalt intrusion in surface waters endangers freshwater availability, influences water quality, and affects estuarine ecosystem services with high economic and social values. Salt transport and the resulting salinity distributions result from the non-linear interactions between salt and water dynamics. Estuaries are often considered under (quasi)-steady assumption or by focusing on specific timescales. Our understanding of their temporal multiscale response to transient forcing is limited, which hinders the implementation of effective mitigation strategies. We apply wavelet analyses to quantify the variability of salt intrusion from hourly to seasonal timescales and unravel the temporal variability of its response across scales. We focus on an estuary that undergoes significant transient forcing, the Modaomen estuary in the Pearl River Delta, and apply the wavelet analyses to year-long data generated by a coastal ocean numerical model. Our results show that this estuary responds to changes in tidal and riverine forcing throughout the year over interwoven timescales. Our results highlight the temporal variability of the salt intrusion response time both within a given regime and for the transition between regimes. They also suggest that tides control the response time more strongly than river discharge, even though river discharge determines the magnitude of the salt intrusion, and thus modulates the evolution of the salt intrusion response time. We propose a broadly applicable framework to calculate response times with simple data. These results can provide a first-order guidance for design and implementation of estuarine management strategies and mitigation measures that ensure water access and facilitate sustainable development
Evaluation of corneal thickness using a Scheimpflug–Placido disk corneal analyzer and comparison with ultrasound pachymetry in eyes after laser in situ keratomileusis
No full textPurposeTo evaluate the repeatability and reproducibility of corneal thickness measurements in post-laser in situ keratomileusis (LASIK) eyes using a rotating Scheimpflug camera combined with a Placido disk corneal topographer (Sirius) and compare the results with those of ultrasound (US) pachymetry.SettingEye Hospital of Wenzhou Medical College, Wenzhou, China.DesignComparative evaluation of a diagnostic test or technology.MethodsPatients were examined 3 times with the Scheimpflug–Placido topographer by 2 examiners. The central pupil corneal thickness (CTpupil), apical corneal thickness (CTapex), and thinnest corneal thickness (CTthinnest) were recorded. After noncontact examinations, US pachymetry was used to obtain the central corneal thickness (CCT).ResultsThe Scheimpflug–Placido topographer showed high intraoperator repeatability as indicated by a test–retest repeatability of less than 8.5 μm for CTpupil, CTapex, and CTthinnest, The coefficients of variation (CoV) were less than 0.7%, and the intraclass correlation coefficient was higher than 0.99. Excellent results were also obtained for interoperator reproducibility. All CoVs were less than 0.5%. The 95% limits of agreement between the Scheimpflug–Placido measurement and the US pachymetry measurements were narrow (−16.62 to 12.44 μm for CTpupil versus US pachymetry CCT; −17.49 to 12.16 μm for CTapex versus US pachymetry CCT; −18.59 to 10.90 μm for CTthinnest versus US pachymetry CCT).ConclusionsThe Scheimpflug–Placido topographer showed excellent intraoperator repeatability and interoperator reproducibility of CTpupil, CTapex, and CTthinnest measurements in post-LASIK eyes. The CCT measurements obtained using the device were in high agreement with those obtained by US pachymetry, suggesting that the 2 devices are interchangeable.Financial DisclosureNo author has a financial or proprietary interest in any material or method mentioned
