Machine learning for pore-water pressure time-series prediction: Application of recurrent neural networks

Knowledge of pore-water pressure(PWP)variation is fundamental for slope stability.A precise prediction of PWP is difficult due to complex physical mechanisms and in situ natural variability.To explore the applicability and advantages of recurrent neural networks(RNNs)on PWP prediction,three variants of RNNs,i.e.,standard RNN,long short-term memory(LSTM)and gated recurrent unit(GRU)are adopted and compared with a traditional static artificial neural network(ANN),i.e.,multi-layer perceptron(MLP).Measurements of rainfall and PWP of representative piezometers from a fully instrumented natural slope in Hong Kong are used to establish the prediction models.The coefficient of determination(R^2)and root mean square error(RMSE)are used for model evaluations.The influence of input time series length on the model performance is investigated.The results reveal that MLP can provide acceptable performance but is not robust.The uncertainty bounds of RMSE of the MLP model range from 0.24 kPa to 1.12 k Pa for the selected two piezometers.The standard RNN can perform better but the robustness is slightly affected when there are significant time lags between PWP changes and rainfall.The GRU and LSTM models can provide more precise and robust predictions than the standard RNN.The effects of the hidden layer structure and the dropout technique are investigated.The single-layer GRU is accurate enough for PWP prediction,whereas a double-layer GRU brings extra time cost with little accuracy improvement.The dropout technique is essential to overfitting prevention and improvement of accuracy.     

Parsivel(R)激光降水粒子谱仪及其在气象领域的应用

降水粒子特性是大气运动和云内微物理过程的综合结果,在云降水物理及人工影响天气领域有着重要的意义.传统的测量方法不适合对大量数据分析寻找规律,德国OTT公司的Parsivel(R)激光降水粒子谱测量系统可以较好解决自动测量难题.该仪器是以激光测量为基础的粒子测量传感器,采用平行激光束和光电管阵列结合,当有降水粒子穿越采样空间时,自动记录遮挡物的宽度,通过穿越时间计算降水粒子的尺度和速度,根据各种参数的综合信息对降水粒子进行分类,并能够以数字形式显示瞬时降水强度、降水粒子总数、累积降水量、降水时的能见度和雷达反射因子,以图形方式显示降水粒子尺度谱、速度谱、降水粒子分类且自动生成天气现象代码,实现天气现象的自动识别.激光降水粒子谱仪主要用于气象水文观测.在雷达气象学领域可用于Z/R关系的拟合修正,比传统的用雨量筒观测数据拟合效果好得多;由降水粒子谱仪测量雨滴的降落速度,可以对天气雷达垂直向上测量的粒子径向速度谱进行校正.人工影响天气的效果检验一直是一个难题,自然降水粒子谱分布形式与人工催化以后的降水粒子谱型理论上应当具有较大的差别,人工增雨作业降水滴谱变化物理响应和降水强度时间变化响应都有明显的区别.如果能够实时检测到这些差别,就能够充分说明人工催化的有效性.未来如果能够进行联网观测记录区域性降水、降雹,就有充分证据表明人影作业的有效性,在定量化作业效果评估以及灾害损失评估等方面应用潜力巨大.利用该仪器已经对一年的自然降水过程进行了连续观测,并将所获得的降水粒子谱、雨滴浓度值随时间变化状况与卫星反演的云顶有效粒子半径时间变化趋势进行了对比,发现有较好的一致性.     

中国北方磷矿矿床类型及成矿规律

通过对中国北方内生磷灰石、变质及沉积型磷矿成矿特征及典型矿床的研究,建立了各种类型磷矿床的成矿模式和找矿模型、确定了各类型矿床的时空分布规律以及矿床的共生规律,为开展磷矿的找矿工作具有重要参考价值。     

中国北方磷矿成矿预测

在中朝准地台、塔里木准地台边缘深断裂附近和台中隆起带以及天山-兴安、祁连-秦岭、额尔古纳几个地槽褶皱区的中间地带,不同程度地具备中国北方三种主要类型磷矿的成矿条件。研究分析认为中国北方的内生磷矿具有良好的找矿前景,绿岩带型磷矿和古元古代沉积变质磷矿显示一定的找矿潜力,沉积型磷矿一般品位低、规模小。通过归纳,在中国北方划分出磷矿的二级城矿区4个、三级成矿带15个、四级成矿亚带19个。圈定成矿预测区A级2个,B级C级各23个。确定了21处可开展进一步工作的地区。     

东海和黄海表层沉积物生物标志物的分布特征及古生态重建潜力

多参数生物标志物法已被广泛用于重建浮游植物生产力和种群结构.但目前还缺乏这一应用的现代证据,尤其是西太平洋边缘海地区.文中作者对东/黄海的表层沉积物中的浮游植物生物标志物的分布进行研究,发现东海表层沉积物中菜子甾醇、甲藻甾醇、长链烯酮的总含量变化范围为357～3 115 ng/g,黄海的总含量变化范围为2 898 ～6 714 ng/g,硅藻、甲藻和颗石藻生物标志物总量基本具有近岸高、离岸低的趋势,与现场调查资料所得到的趋势基本一致.生物标志物比例结果显示在东海的长江口方向离岸越远硅藻相对比例下降,而在长江口以南则是离岸越远硅藻比例升高.在东海和黄海颗石藻所占比例都很低,但是有离岸越远比例升高的趋势.用生物标志物所重建的种群结构的变化趋势与现场调查资料相符,但其绝对值却与现场资料存在差异.本研究为生物标志物重建种群结构提供了现代数据,但是此方法的定量应用还需做进一步的研究.     

��ά�ռ����ά�ռ��߲���ת����������·���

????????????δ?????????????????????????????????·?????????????????????????????????????????????????????????÷?Χ???ó?????:????γ??????Χ??С??8???,??x??y???????????????1 cm??????????侫???????γ??(?????γ??)??Χ????С??????????     

Fresh snow chemistry from high mountain regions in central himalayas

Investigations of atmospheric composition in the Himalayas has been limited in both temporal and spatial scales, mainly due to difficult logistics. Ideal sites for monitoring atmospheric composition and its evolution should be free from local pollution and representative of the remote troposphere (HUEBERT et al., 1980). As the Himalayas are far removed from highly industrialized regions they provide suitable locations to monitor the chemistry of the remote troposphere and to study the evolu…     

New patterns of urban development in China

1　ＩＮＴＲＯＤＵＣＴＩＯＮＣｉｔｙｐｌａｙｓａｎｉｍｐｏｒｔａｎｔｒｏｌｅｉｎｔｈｅｐｒｏｃｅｓｓｏｆｒｅ ｇｉｏｎａｌｅｃｏｎｏｍｉｃｄｅｖｅｌｏｐｍｅｎｔ .Ｆｏｒｉｎｓｔａｎｃｅ ,ｉｎ 1996,ｔｈｅｔｏｔａｌｍｕｎｉｃｉｐａｌａｒｅａｓｏｆ 666ｃｉｔｉｅｓｏｎｌｙｃｏｖｅｒｅｄｔｈｅｐｒｏｐｏｒｔｉｏｎｏｆ 18.12ｐｅｒｃｅｎｔｉｎｎａｔｉｏｎａｌｔｅｒｒｉｔｏｒｙ ,ａｎｄｂｕｉｌｔ ｕｐａｒｅａ 1.8ｐｅｒｃｅｎｔ .Ｂｕｔｔｈｅｙｐｒｏｄｕｃｅｄｔｈｅ 68.63ｐｅｒｃｅｎｔｏｆＧＤＰ(ＧｒｏｓｓＤｏ…     

Spatiotemporal characteristics of Qinghai Lake ice phenology between 2000 and 2016

Journal of Geographical Sciences - Lake ice phenology is considered a sensitive indicator of regional climate change. We utilized time series information of this kind extracted from a series of...     

Evaluation of the integrated multi-satellite retrievals for global precipitation measurement over the Tibetan Plateau

The availability of high-resolution satellite precipitation measurement products provides an opportunity to monitor precipitation over large and complex terrain and thus accurately evaluate the climatic, hydrological and ecological conditions in those regions. The Global Precipitation Measurement(GPM) mission is an important new program designed for global satellite precipitation estimation, but little information has been reported on the applicability of the GPM’s products for the Tibetan Plateau(TP). The object of this study is to evaluate the accuracy of the Integrated Multi-Satellite Retrievals for GPM(IMERG) Final Run product under different terrain and climate conditions over the TP by using 78 ground gauges from April 2014 to December 2017. The results showed the following:(1) the 3-year average daily precipitation estimation in the IMERG agrees well with the rain gauge observations(R~2=0.58, P0.01), and IMERG also has a considerable ability to detect precipitation, as indicated by a high probability of detection(78%-98%) and critical success index(65%-85%);(2) IMERG performed better at altitudes from 3000 m to 4000 m with a small relative bias(RB) of 6.4%. Precipitation change was not significantly affected by local relief;(3) the climate system of the TP was divided into four climate groups with a total of 12 climate types based on the K?ppen climate classification system, and IMERG performed well in all climate types with the exception of the arid-desert-cold climate(Bwk) type. Furthermore, although IMERG showed the potential to detect snowfall, it still exhibits deficiencies in identifying light and moderate snow. These results indicate that IMERG could provide more accurate precipitation data if its retrieval algorithm was improved for complex terrain and arid regions.