������С���˵��Ŷ�����

??????????????????????????????????????????????????????????????????????????????????????????????????μ???С???????????С????????????????????????С??????????????????????????????????????????????С???????????????????????????????????С??????????????????С???????????С????????????????????????????     

Potential of support vector regression for prediction of monthly streamflow using endogenous property

In the recent past, a variety of statistical and other modelling approaches have been developed to capture the properties of hydrological time series for their reliable prediction. However, the extent of complexity hinders the applicability of such traditional models in many cases. Kernel‐based machine learning approaches have been found to be more popular due to their inherent advantages over traditional modelling techniques including artificial neural networks(ANNs ). In this paper, a kernel‐based learning approach is investigated for its suitability to capture the monthly variation of streamflow time series. Its performance is compared with that of the traditional approaches. Support vector machines (SVMs) are one such kernel‐based algorithm that has given promising results in hydrology and associated areas. In this paper, the application of SVMs to regression problems, known as support vector regression (SVR), is presented to predict the monthly streamflow of the Mahanadi River in the state of Orissa, India. The results obtained are compared against the results derived from the traditional Box–Jenkins approach. While the correlation coefficient between the observed and predicted streamflows was found to be 0·77 in case of SVR, the same for different auto‐regressive integrated moving average (ARIMA) models ranges between 0·67 and 0·69. The superiority of SVR as compared to traditional Box‐Jenkins approach is also explained through the feature space representation. Copyright © 2009 John Wiley & Sons, Ltd.     

Capturing channelized reservoir connectivity uncertainty with amalgamation curves

During reservoir characterization all the geological uncertainties affecting the quantity and distribution of hydrocarbons should be captured to assess the risks affecting final recovery.In a typical modeling workflow the geological uncertainties are accounted for through the construction of a sufficiently large set of 3-D static models. Out of this set, a few representative models are selected and dynamically simulated so as to correlate the geological characteristics of the reservoir with its dynamic performance and to propagate the uncertainty onto the final recovery factors, yet maintaining the computational run time acceptable. In channelized depositional environments, which are strongly heterogeneous, the selection approach must also account for channel connectivity, which plays a key role in the possibility of efficiently draining the reservoir for a reasonable number of wells.This study can be seen as a step forward in the assessment of the risks associated to the development of channelized reservoirs under the assumption that a way to express the concept of channel connectivity is channel amalgamation. Channel amalgamation is here defined through amalgamation curves which are numerically described using a set of indexes whose combination provide spatial information of channel intersections. These indexes were calculated for a full set of 3-D geological models and used to steer the selection of a representative model sub-set for subsequent fluid flow simulations.The validity of the index-based selection was verified on different sets of synthetic reservoir models through the evaluation of the representativeness of the model sub-set in reproducing the uncertainty of the original dataset. Eventually, the existence of a strong correlation between channel amalgamation and production performance was proved. From a practical perspective, the possibility to include channel amalgamation in the assessment of the geological models can considerably improve the representativeness of the selected models for uncertainty propagation thus reducing significantly the number of geological models to be considered.     

一种基于Bancroft算法的GPS动态抗差自适应滤波

基于抗差自适应滤波的思想，结合非线性Bancroft算法的特点，提出了一种基于Bancroft算法的GPS动态抗差自适应滤波。计算表明，该算法不仅在一定程度上减弱了由于线性化忽略高次项对导航解的影响，而且再次证实抗差自适应滤波在控制扰动异常的有效性和合理性。     

The impacts of food price and income shocks on household food security and economic well-being: Evidence from rural Bangladesh

This paper examines the combined impacts of food price and income shocks on household food security and economic well-being in low-income rural communities. Using longitudinal survey data of 1800 rural households from 12 districts of Bangladesh over the period 2007–2009, we estimated a three-stage hierarchical logit model to identify the key sources of household food insecurity. The first-difference estimator was then employed to compare pre- and post-shock expenditure for those households that experienced acute food shortages and those that managed to avoid the worst impacts of the shocks. On the basis of our results we conclude that: (1) the soaring food prices of 2007–2009 unequivocally aggravated food insecurity in the rural areas of Bangladesh; (2) the subsequent income shocks of 2007–2009 contributed toward worsening food insecurity; (3) the adverse impacts of these shocks appeared to have faded over time due to labor and commodity market adjustments, regional economic growth, and domestic policy responses, leaving no profound impacts on households’ economic well-being in most cases; and (4) although the immediate adverse consequences of rising food prices were borne disproportionately by the poor, the longer term consequences were distributed more evenly across the rich and poor and were favorable for the day laborers.     

Corrected prediction intervals for change detection in paired watershed studies

Abstract

Hydrological data may be temporally autocorrelated requiring autoregressive process parameters to be estimated. Current statistical methods for hydrological change detection in paired watershed studies rely on prediction intervals, but the current form of prediction intervals does not include all appropriate sources of variation. Corrected prediction intervals for the analysis of paired watershed study data that include variation associated with covariance and linear model parameter estimation are presented. We provide an example of their application to data from the Hinkle Creek Paired Watershed Study located in the western Cascade foothills of Southern Oregon, USA. Research implications of using the correct prediction limits and incorporating the estimation uncertainty of autoregressive process parameters are discussed.

Editor D. Koutsoyiannis

Citation Som, N.A., Zégre, N.P., Ganio, L.M. and Skaugset, A.E., 2012. Corrected prediction intervals for change detection in paired watershed studies. Hydrological Sciences Journal, 57 (1), 134–143.     

总体最小二乘拟合问题求解方法的比较研究

目前对总体最小二乘求解方法的研究,出现了奇异值分解的总体最小二乘法、顾及自变量和因变量误差的总体最小二乘法及正交总体最小二乘法.在模型推导的基础上,本文对3种总体最小二乘法在直线和平面拟合中求解的参数及其精度进行了分析,通过与最小二乘法的比较表明,总体最小二乘法得到的拟合结果更加稳健,且以正交总体最小二乘法的拟合结果为最优.     

Estimation of Areal Frequency and Mean Trace Length of Discontinuities Observed in Non-Planar Surfaces

Summary. Mauldon (1998) suggested end-point estimators of areal frequency and mean trace length for a planar sampling window which were recently proved to be unbiased maximum likelihood estimators by Lyman (2003). The present paper is to expand the concept and applicability of the end-point estimators to those for a general non-planar sampling window. The generalized end-point estimators are verified and its applicability for variable discontinuity orientation is checked by Monte Carlo simulation. Standard deviation of estimation error and estimation efficiency of areal frequency and mean trace length are also considered.     

基于集合经验模态分解的新疆地区温度场预测评估

综合运用经验正交分解法(EOF),集合经验模态分解(EEMD)和最小二乘支持向量机（LS SVM）,构建新疆地区候平均温度距平场预测模型。采用EEMD分别对经过EOF分解得到的前3个模态的时间系数进行分解,对分解得到的结果运用最小二乘支持向量机进行预测并重构得到了各个时间系数的预测结果,将时间系数预测的结果与空间场重构得到了候平均温度距平场的计算结果,在候平均的基础上计算得到了旬平均的结果,在旬平均的基础上计算得到了月平均的结果。通过采用距平相关系数(ACC〖WTBZ〗),预报技巧(SS)〖WTBZ〗和同号率对结果进行评估显示,对于候平均预测,其在前20候内的预测较为理想,平均ACC〖WTBZ〗达到了0.32,平均SS〖WTBZ〗达到了0.70,平均同号率达到了0.80。对于旬平均的预测,其在前10旬内较为理想,10旬以内平均ACC〖WTBZ〗达到了0.50,平均SS〖WTBZ〗达到了0.50,平均同号率达到了0.50。对于月平均的预测,3个月的预测平均ACC〖WTBZ〗达到了0.50,平均SS〖WTBZ〗达到了0.50,平均同号率达到了0.80。3个月内的短期气候预测具有较高的水平。