Multi-step ahead short-term predictions of storm surge level using CNN and LSTM network

Storm surges pose significant danger and havoc to the coastal residents’safety,property,and lives,particularly at offshore locations with shallow water levels.Predictions of storm surges with hours of warning time are important for evacuation measures in low-lying regions and coastal management plans.In addition to experienced predictions and numerical models,artificial intelligence(AI)techniques are also being used widely for short-term storm surge prediction owing to their merits in good level of prediction accuracy and rapid computations.Convolutional neural network(CNN)and long short-term memory(LSTM)are two of the most important models among AI techniques.However,they have been scarcely utilised for surge level(SL)forecasting,and combinations of the two models are even rarer.This study applied CNN and LSTM both individually and in combination towards multi-step ahead short-term storm surge level prediction using observed SL and wind information.The architectures of the CNN,LSTM,and two sequential techniques of combining the models(LSTM–CNN and CNN–LSTM)were constructed via a trial-and-error approach and knowledge obtained from previous studies.As a case study,11 a of hourly observed SL and wind data of the Xiuying Station,Hainan Province,China,were organised as inputs for training to verify the feasibility and superiority of the proposed models.The results show that CNN and LSTM had evident advantages over support vector regression(SVR)and multilayer perceptron(MLP),and the combined models outperformed the individual models(CNN and LSTM),mostly by 4%–6%.However,on comparing the model computed predictions during two severe typhoons that resulted in extreme storm surges,the accuracy was found to improve by over 10%at all forecasting steps.

Sea-water-level prediction via combined wavelet decomposition,neuro-fuzzy and neural networks using SLA and wind information

Sea-water-level(SWL)prediction significantly impacts human lives and maritime activities in coastal regions,particularly at offshore locations with shallow water levels.Long-term SWL forecasts,which are conventionally obtained via harmonic analysis,become ineffective when nonperiodic meteorological events predominate.Artificial intelligence combined with other data-processing methods can effectively forecast highly nonlinear and nonstationary inflow patterns by recognizing historical relationships between input and output.These techniques are considerably useful in time-series data predictions.This paper reports the development of a hybrid model to realize accurate multihour SWL forecasting by combining an adaptive neuro-fuzzy inference system(ANFIS)with wavelet decomposition while using sea-level anomaly(SLA)and wind-shear-velocity components as inputs.Numerous wavelet-ANFIS(WANFIS)models have been tested using different inputs to assess their applicability as alternatives to the artificial neural network(ANN),wavelet ANN(WANN),and ANFIS models.Different error definitions have been used to evaluate results,which indicate that integrated wavelet-decomposition and ANFIS models improve the accuracy of SWL prediction and that the inputs of SLA and wind-shear velocity exhibit superior prediction capability compared to conventional SWL-only models.

Study on the Simulation Technique for Urban Electromagnetic Environment Supported by GIS

Electromagnetic radiation environment is one of the important environmental factors. In resent 20 years, the applications of electromagnetic technology became more and more popular. However, electromagnetic environment makes some negative impacts on the surrounding environments and human health, and Environmental Management and Research Department pay a lot attention to it. It will make a great significance if we study of the simulation technique for urban electromagnetic environment which supported by GIS. This paper presents some key technologies of the electromagnetic environment simulation, including terrain rendering, city building modeling and volume data superposition, basically realize the visualization of volume data (electromagnetic fields) based on city building model and real time mapping of virtual scene data.

The O2-ZmGRAS11 transcriptional regulatory network orchestrates the coordination of endosperm cell expansion and grain filling in maize

Maize(Zea mays)endosperm filling is coordinated with cell expansion to enlarge the grain size,but the mechanism coupling the two processes is poorly understood.Starchy endosperm cells basically contain no visible vacuoles for cell expansion.During grain filling,efficient synthesis of storage compounds leads to reduced cytoplasm and thus lowered cell turgor pressure.Although bioactive gibberellins(GAs)are essential for cell expansion,they accumulate at a low level at this stage.In this study,we identified an endosperm-specific GRAS domain-containing protein(ZmGRAS11)that lacks the DELLA domain and promotes cell expansion in the filling endosperm.The zmgras11 loss-of-function mutants showed normal grain filling but delayed cell expansion,thereby resulting in reduced kernel size and weight.Overexpression of ZmGRAS11 led to larger endosperm cells and therefore increased kernel size and weight.Consistent with this,ZmGRAS11 positively regulates the expression of ZmEXPB12,which is essential for cell expansion,at the endosperm filling stage.Moreover,we found that Opaque2(O2),a central transcription factor that regulates endosperm filling,could directly bind to the promoter of ZmGRAS11 and activate its expression.Taken together,these results suggest that endosperm cell expansion is coupled with endosperm filling,which is orchestrated by the O2-ZmGRAS11-centered transcriptional regulatory network.Our findings also provide potential targets for maize yield improvement by increasing the storage capacity of endosperm cells.

LjCYC Genes Constitute Floral Dorsoventral Asymmetry in Lotus japonicus

Previous study shows that LjCYC2,a CYC-like TCP （TB1,CYC and PCFs） gene in the model legume,Lotus japonicus,is involved in dorsal petal development,which together with the other two homologous genes,LjCYC1 and LjCYC3,belongs to an LjCYC gene cluster.In this report,we modified the transformation system in L.japonicus,and constructed different RNAi transgenes to target different LjCYC genes.The expression of three endogenous LjCYC genes was specifically suppressed by different specific RNAi transgenes,and a chimerical RNAi transgene that contains the specific sequences from LjCYC1 and LjCYC2 was found to downregulate the expression of both endogenous genes simultaneously.Effects of silencing three LjCYC genes were mainly restricted on either dorsal or lateral petals,demonstrating their dorsalizing and lateralizing activities during the development of zygomorphic flower.Furthermore,abolishing the expression of three LjCYC genes could give rise to complete loss of dorsoventral （DV） differentiation in the flower whose petals all resembled the ventral one in the wild type and displayed intact organ internal （IN） asymmetry.Our data demonstrate that during zygomorphic flower development,the DV asymmetry is constituted by the LjCYC genes,while the floral organ IN asymmetry is independently determined by other genetic factors.

A new trapezoidal-mesh based data model for spatial operations

This paper presents a new spatial data model based on trapezoidal-mesh for implementing spatial operations within geographical information systems(GIS).Based only on the solid foundation of spatial operations,diversified application models can be established to bridge the gap between Digital Earth models and the real world with its real-world problems(‘connecting through location’).In this paper,the involved polygon features are decomposed into a series of trapezoidalmeshes.Then,geo-processing operations are employed on these meshes rather than the original polygon features,resulting in a relatively simple spatial computation.As a kind of model designed by integrating raster with vector,the model presented here has advantages over other models when carrying out spatial operations insofar as providing a solid foundation for achieving the grand goal of Digital Earth.The concept of this data model and the two extensive examples of its application in spatial operations are elaborated upon in this article.As a result,this article and the research that supports it,proves that the adoption of the trapezoidal-mesh model greatly improves the efficiency of spatial operations in GIS.