基于CSA-PLS算法的养殖水体水质快速高光谱预测反演模型研究

养殖水体水质的优劣直接影响养殖对象的成长,准确、快速、全面地掌控养殖水环境的水质参数变化情况具有重要意义。传统的水质指标监测方法都通过人工采样的方式,不仅耗费时间长,且只能体现局部水体情况。针对这些问题,提出了一种乌鸦搜索算法(CSA)结合偏最小二乘回归(PLSR)的高光谱特征波段筛选方法,快速构建回归模型,实现光谱数据的精准预测反演。以连片的养殖小区为研究对象,采集养殖水体样本并拍摄同时期的高光谱影像数据。首先对提取的采样点光谱数据利用多种数据变换方法分别预处理;其次利用这些数据,对水质指标总氮(TN)、氨氮(NH₄⁺-N)、总磷(TP)和化学需氧量(COD)分别构建全波段的SVR和AdaBoost回归模型,同时与提出的CSA-PLS自动筛选波段方法和传统的连续投影算法(SPA)筛选波段后构建的模型进行比较分析;最后根据决定系数(R²)和均方根误差(REMS)选出适合各水质指标的最优模型。从实验结果可以看出,所提波段筛选方法的AdaBoost模型预测结果优于SVR和传统SPA方法提取特征波段后构建的模型,与全波段最优模...     

海底斜坡土体失稳的研究进展

根据国内外最新的文献,综述了海底斜坡失稳研究核心问题的进展情况。这些进展主要包括海底斜坡土体失稳的定性和定量研究,以及海底斜坡土体失稳后演变趋势的定性和定量研究。     

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.     

Mechanical properties of calcareous silts in a hydraulic fill island-reef

Abstract

The mechanical characteristics of calcareous silt interlayers play an important role in the stability of island-reef foundations. Direct shear and consolidation tests were performed to study the relationship between the mechanical properties and the physical parameters of calcareous silt. Based on the consolidation test results and analysis of the settling examples, different calculation methods for soil settling were compared. The results show the following. (1) The relationship between the cohesion and water content of calcareous silt can be represented by an M-shaped curve. The water contents corresponding to the two peaks of the M-type curve increase with increasing dry density. (2) When the dry density is less than 1.33?g/cm³, increasing the density significantly improves the internal friction angle of calcareous silts. When the dry density of the calcareous silt is greater than 1.33?g/cm³, the internal friction angle is affected by both the dry density and the water content. (3) The shear strength decreases when the water content exceeds the optimum level. (4) The compressive modulus of calcareous silt is larger than that of terrigenous silt. Specifically, it decreases with decreasing dry density and increasing water content. (5) The stepwise loading method should be used to estimate the soil settling before fill engineering construction.     

A hybrid integrated deep learning model for the prediction of citywide spatio-temporal flow volumes

ABSTRACT

The spatio-temporal residual network (ST-ResNet) leverages the power of deep learning (DL) for predicting the volume of citywide spatio-temporal flows. However, this model, neglects the dynamic dependency of the input flows in the temporal dimension, which affects what spatio-temporal features may be captured in the result. This study introduces a long short-term memory (LSTM) neural network into the ST-ResNet to form a hybrid integrated-DL model to predict the volumes of citywide spatio-temporal flows (called HIDLST). The new model can dynamically learn the temporal dependency among flows via the feedback connection in the LSTM to improve accurate captures of spatio-temporal features in the flows. We test the HIDLST model by predicting the volumes of citywide taxi flows in Beijing, China. We tune the hyperparameters of the HIDLST model to optimize the prediction accuracy. A comparative study shows that the proposed model consistently outperforms ST-ResNet and several other typical DL-based models on prediction accuracy. Furthermore, we discuss the distribution of prediction errors and the contributions of the different spatio-temporal patterns.     

基于Meta-Gaussian模型的中国农业干旱预测研究

在全球气候变化背景下,干旱愈加频发,有效且可靠的农业干旱预测对于保障粮食安全和水资源安全具有重要意义。以标准化降水指数（SPI）和联合标准化土壤湿度指数（JSSI）分别表征气象干旱和农业干旱,以前期的气象干旱和农业干旱指数作为预测因子,在1~3个月预见期下基于Meta-Gaussian（MG）模型对中国1961—2015年6—8月的农业干旱进行预测,并采用Brier Skill Score（BSS）和纳什效率系数（NSE）评价MG模型的预测性能。结果表明：① 将1个月、3个月、6个月、9个月和12个月时间尺度的标准化土壤湿度指数（SSI）结合起来得到的JSSI能够对中国农业干旱的综合状况进行客观评价。② 以中国2010年和2014年遭受严重的干旱事件为例,预见期为1~3个月时,除新疆南部、青海西部以及内蒙古西部等沙漠地区外,MG模型对6—8月农业干旱预测结果的分布范围与实际干旱的分布区域较吻合,预见期越短,吻合越好。③ 预见期为1个月时,6—8月BSS ≥ 0.5的面积比例分别为0.714、0.642和0.640,NSE ≥ 0.5的面积比例分别为0.903、0.829和0.837,表明MG模型能够对中国大部分区域的农业干旱作出可靠的预测。本文结果可为中国农业干旱的监测、预警及干旱决策提供科学指导。     

Reconstruction of the Cenozoic deformation of the Bohai Bay Basin,North China

A well‐constrained plate deformation model may lead to an improved understanding of sedimentary basin formation and the connection between subduction history and over‐riding plate deformation. Building quantitative models of basin kinematics and deformation remains challenging often due to the lack of comprehensive constraints. The Bohai Bay Basin (BBB) is an important manifestation of the destruction of the North China Craton, and records the plate kinematic history of East Asia during the Cenozoic. Although a number of interpretations of the formation of the BBB have been proposed, few quantitative basin reconstruction models have been built to test and refine previous ideas. Here, we developed a quantitative deformation reconstruction of the BBB constrained with balanced cross‐sections and structural, stratigraphic and depositional age data. Our reconstruction suggests that the basin formation process was composed of three main stages: Paleocene‐early Eocene (65–42 Ma) extension initiation, middle Eocene‐early Oligocene (42–32.8 Ma) extension climax and post‐Oligocene (32.8–0 Ma) post‐extensional subsidence. The deformation of the BBB is spatially heterogeneous, and its velocity directions rotated clockwise during the basin formation process. The reconstruction supports the interpretation that the BBB formed via strike‐slip faulting and orthogonal extension and that the basin is classified as a composite extensional‐transtensional basin. We argue that the clockwise rotation of the basin velocity field was driven by the counter‐clockwise rotation in the direction of Pacific Plate subduction. The kinematics of the BBB imply that the Pacific Plate may have been sufficiently coupled to the over‐riding East Asian Plate during the critical period of Pacific Plate reorganization. The new reconstruction provides a quantitative basis for studies of deformation processes not only in the vicinity of the BBB, but also more broadly throughout East Asia.     

Method for UAV-based 3D topography reconstruction of tidal creeks

It is common to obtain the topography of tidal flats by the Unmanned Aerial Vehicle(UAV) photogrammetry,but this method is not applicable in tidal creeks.The residual water will lead to inaccurate depth inversion results,and the topography of tidal creeks mainly de-pends on manual survey.The present study took the tidal creek of Chuandong port in Jiangsu Province,China,as the research area and used UAV oblique photogrammetry to reconstruct the topography of the exposed part above the water after the ebb tide.It also proposed a Trend Prediction Fitting (TPF) method for the topography of the unexposed part below the water to obtain a complete 3D topography.The topography above the water measured by UAV has the vertical precision of 12 cm.When the TPF method is used,the cross-section should be perpendicular the central axis of the tidal creek.A polynomial function can be adapted to most shape of sections,while a Fourier function obtains better results in asym-metrical sections.Compared with the two-order function,the three-order function lends itself to more complex sections.Generally,the TPF method is more suitable for small,straight tidal creeks with clear texture and no vegetation cover.     

地理空间传感网融合服务技术与应用

随着遥感数据网、传感网、物联网和人工智能的发展,逐渐形成空天地海立体化、集成化和一体化的地理空间传感网。地理空间传感网感知资源呈现出多源、异构和分散的特征,面向多层次用户个性化、即时化和智能化应用需求,存在异构资源共享管理、多协议实时接入、时空无缝感知、自动化感知和精准预测等技术挑战。静态地理信息服务由于无法提供鲜活的地理信息,难以满足地理事件的综合监测、决策预警和聚焦应用需求,急需发展地理空间传感网融合服务技术和实时动态地理信息服务平台。本文围绕信息物理网环境下空天地海观测平台的观测高效共享和融合服务问题,提出了传感网观测共享信息模型和点面观测协同无缝重建模型,突破了观测在线接入、集成管理、星地融合、时空预测和聚焦服务等地理空间传感网融合服务关键技术,研制了包含“感—联—知—控”等功能的传感网时空信息网络感知服务系统GeoSensor,介绍了GeoSensor在流域、海洋和城市等典型应用。未来将进一步发展“人—水—城”智能感知认知理论,突破“空天地海人”群智感知、空间智能和认知服务技术,开展长江经济带应用。     

安徽省2021年梅雨期降水预报检验分析

检验梅雨期降水的预报效果,对于提升梅雨期降水预报能力、减少梅雨期降水带来的人员伤亡和经济财产损失有着重要的意义。文章对安徽省2021年梅雨期（6月10日—7月10日）六个客观模式和一个主观订正预报产品进行了检验分析,其中包含了三个区域模式数值预报（中国气象局中尺度天气数值预报系统（简称CMA-MESO）、中国气象局上海数值预报模式系统（简称CMA-SH9）、安徽WRF）、三个全球模式数值预报（中国气象局全球同化预报系统（简称CMA-GFS）、欧洲中期天气预报中心确定性预报模式（简称ECMWF）、美国国家环境预报中心全球预报系统（简称NCEP-GFS））和安徽智能网格主观订正预报的降水产品,进行了检验分析,结果表明：传统检验中安徽智能网格和区域模式对晴雨准确率的预报效果优于全球模式,又以CMA-MESO最优;在暴雨及以上量级的强降水预报中,传统检验表明安徽智能网格预报的得分最高（23.83）,ECMWF模式则是客观模式预报中效果最好的（20.12）,CMA-SH9次之（19.34）;通过对除安徽智能网格以外的各个客观数值模式进行的MODE空间检验可知,不同数值模式间暴雨预报误差原因不尽相同,ECMWF与各区域数值模式主要是由雨区位置的预报偏差,尤其是纬度偏差导致的,NCEP-GFS全球模式对降水强度和雨区面积的预报偏弱偏小比较明显,CMA-GFS在强降水方面的预报可参考性较差;各个主客观预报暴雨及以上量级预报,整体表现出较明显的日变化特征,在午夜前后、上午时段TS评分较高,而午后到傍晚评分较低,这个现象或许是梅雨期的午后降水多以地表太阳加热引起的短历时热对流降水为主造成的。