Building more realistic reservoir optimization models using data mining - A case study of Shelbyville Reservoir

In this paper, we promote a novel approach to develop reservoir operation routines by learning from historical hydrologic information and reservoir operations. The proposed framework involves a knowledge discovery step to learn the real drivers of reservoir decision making and to subsequently build a more realistic (enhanced) model formulation using stochastic dynamic programming (SDP). The enhanced SDP model is compared to two classic SDP formulations using Lake Shelbyville, a reservoir on the Kaskaskia River in Illinois, as a case study. From a data mining procedure with monthly data, the past month’s inflow (Q_t−1), current month’s inflow (Q_t), past month’s release (R_t−1), and past month’s Palmer drought severity index (PDSI_t−1) are identified as important state variables in the enhanced SDP model for Shelbyville Reservoir. When compared to a weekly enhanced SDP model of the same case study, a different set of state variables and constraints are extracted. Thus different time scales for the model require different information. We demonstrate that adding additional state variables improves the solution by shifting the Pareto front as expected while using new constraints and the correct objective function can significantly reduce the difference between derived policies and historical practices. The study indicates that the monthly enhanced SDP model resembles historical records more closely and yet provides lower expected average annual costs than either of the two classic formulations (25.4% and 4.5% reductions, respectively). The weekly enhanced SDP model is compared to the monthly enhanced SDP, and it shows that acquiring the correct temporal scale is crucial to model reservoir operation for particular objectives.     

自重湿陷性黄土下限深度判定方法探讨

以郑西高速铁路沿线6个现场试坑浸水试验资料和大量室内土工试验资料为基础,对自重湿陷性黄土下限深度的判定问题进行探讨.研究表明:自重湿陷性黄土下限深度是一个明确且唯一的值,但由于受应力条件、变形条件、饱和状态、试样尺寸、边界效应等因素的影响,运用现场取样进行室内压缩试验判定自重湿陷性黄土下限深度的方法还存在一定的局限性,...     

A hybrid efficient method to downscale wave climate to coastal areas

Long-term time series of sea state parameters are required in different coastal engineering applications. In order to obtain wave data at shallow water and due to the scarcity of instrumental data, ocean wave reanalysis databases ought to be downscaled to increase the spatial resolution and simulate the wave transformation process. In this paper, a hybrid downscaling methodology to transfer wave climate to coastal areas has been developed combining a numerical wave model (dynamical downscaling) with mathematical tools (statistical downscaling). A maximum dissimilarity selection algorithm (MDA) is applied in order to obtain a representative subset of sea states in deep water areas. The reduced number of selected cases spans the marine climate variability, guaranteeing that all possible sea states are represented and capturing even the extreme events. These sea states are propagated using a state-of-the-art wave propagation model. The time series of the propagated sea state parameters at a particular location are reconstructed using a non-linear interpolation technique based on radial basis functions (RBFs), providing excellent results in a high dimensional space with scattered data as occurs in the cases selected with MDA. The numerical validation of the results confirms the ability of the developed methodology to reconstruct sea state time series in shallow water at a particular location and to estimate different spatial wave climate parameters with a considerable reduction in the computational effort.     

"莫拉克"台风引起的滑坡泥石流灾害HJ-1图像遥感监测研究

利用HJ-1星2009年"莫拉克"台风前后获取的2个时相图像,通过去相关拉伸 (Decorrelation Stretch)、光谱信息增强和最大似然法分类提取滑坡、泥石流区域,并结合TRMM (Tropical Rainfall Measuring Mission)卫星降雨量数据和DEM (Digital Elevat...     

Late Wisconsinan glaciation and postglacial relative sea-level change on western Banks Island,Canadian Arctic Archipelago

The study revises the maximum extent of the northwest Laurentide Ice Sheet (LIS) in the western Canadian Arctic Archipelago (CAA) during the last glaciation and documents subsequent ice sheet retreat and glacioisostatic adjustments across western Banks Island. New geomorphological mapping and maximum-limiting radiocarbon ages indicate that the northwest LIS inundated western Banks Island after ~ 31 ¹⁴C ka BP and reached a terminal ice margin west of the present coastline. The onset of deglaciation and the age of the marine limit (22–40 m asl) are unresolved. Ice sheet retreat across western Banks Island was characterized by the withdrawal of a thin, cold-based ice margin that reached the central interior of the island by ~ 14 cal ka BP. The elevation of the marine limit is greater than previously recognized and consistent with greater glacioisostatic crustal unloading by a more expansive LIS. These results complement emerging bathymetric observations from the Arctic Ocean, which indicate glacial erosion during the Last Glacial Maximum (LGM) to depths of up to 450 m.     

The early Matuyama Diatom Maximum off SW Africa: a conceptual model

An important discovery during Ocean Drilling Program Leg 175, when investigating the record of upwelling off Namibia, was the finding of a distinct Late Pliocene diatom maximum spanning the lower half of the Matuyama reversed polarity chron (MDM, Matuyama Diatom Maximum) and centered around 2.6–2.0 Ma. This maximum was observed at all sites off southwestern Africa between 20°S and 30°S, and is most strongly represented in sediments of Site 1084, off Lüderitz, Namibia. The MDM is characterized by high biogenic opal content, high numbers of diatom valves, and a diatom flora rich in Southern Ocean representatives (with Thalassiothrix antarctica forming diatom mats) as well as coastal upwelling components. Before MDM time, diatoms are rare until ca. 3.6 Ma. After the MDM, in the Pleistocene, the composition of the diatom flora points to increased importance of coastal upwelling toward the present, but is accompanied by a general decrease in opal and diatom deposition. Here we present a simple conceptual model as a first step in formalizing a possible forcing mechanism responsible for the record of opal deposition in the upwelling system off Namibia. The model takes into account Southern Ocean oceanography, and a link with deepwater circulation and deepwater nutrient chemistry which, in turn, are coupled to the evolution of North Atlantic Deep Water (NADW). The model proposes that between the MDM and the Mid-Pleistocene climate revolution, opal deposition off Namibia is not directly tied to glacial–interglacial fluctuations (as seen in the global δ¹⁸O record), but that, instead, a strong deepwater link exists with increased NADW production (as seen in the deepwater δ¹³C record) accounting for higher supply of silicate to the thermocline waters that feed the upwelling process. The opal record of Site 1084 shows affinity to eccentricity on the 400-kyr scale but not for the 100-kyr scale. This points toward long-term geologic processes for delivery of silica to the ocean.     

Mixture model based multivariate statistical analysis of multiply censored environmental data

Environmental data are commonly constrained by a detection limit (DL) because of the restriction of experimental apparatus. In particular due to the changes of experimental units or assay methods, the observed data are often cut off by more than one DL. Measurements below the DLs are typically replaced by an arbitrary value such as zeros, half of DLs, or DLs for convenience of analysis. However, this method is widely considered unreliable and prone to bias. In contrast, maximum likelihood estimation (MLE) method for censored data has been developed for better performance and statistical justification. However, the existing MLE methods seldom address the multivariate context of censored environmental data especially for water quality. This paper proposes using a mixture model to flexibly approximate the underlying distribution of the observed data due to its good approximation capability and generation mechanism. In particular, Gaussian mixture model (GMM) is mainly focused in this study. To cope with the censored data with multiple DLs, an expectation–maximization (EM) algorithm in a multivariate setting is developed. The proposed statistical analysis approach is verified from both the simulated data and real water quality data.     

Evolution of sand encroachment using supervised classification of Landsat data during the period 1987–2011 in a part of Laâyoune-Tarfaya basin of Morocco

Abstract

The study anticipated to understand sand encroachment evolution through analysis of sand contribution across space and time using remote sensing in Laâyoune-Tarfaya basin, Morocco, over the period from 1987 to 2011. The assessment based on supervised classifications of Landsat imagery orthorectified data, using Maximum Likelihood (ML), Minimum Distance (MD) and Support Vector Machine (SVM) classifiers. In order to ameliorate the information, principal components analysis (PCA) and co-occurrence measurement algorithm were used for choosing bands and data transformation. Images differencing was applied on image pairs derived from classification to analyze sand encroachment evolution. All classifiers present enhanced performances, and revealed that area covered by sand was increased by 7%, 4.66% and 4.59% for ML, MD and SVM, respectively. Consequently, images differencing results confirmed that sand material increasing arise not only from coastal area contribution but also mostly from erosion of complicated sand dunes exist in the middle part of the studied area. Evaluating of the presented phenomenon dimensions and its consequences are extremely important to increase the local authorities awareness and mainly for avoiding or minimizing the consequences of the future sand dunes threats.     

Parallelizing maximum likelihood classification on computer cluster and graphics processing unit for supervised image classification

ABSTRACT

Supervised image classification has been widely utilized in a variety of remote sensing applications. When large volume of satellite imagery data and aerial photos are increasingly available, high-performance image processing solutions are required to handle large scale of data. This paper introduces how maximum likelihood classification approach is parallelized for implementation on a computer cluster and a graphics processing unit to achieve high performance when processing big imagery data. The solution is scalable and satisfies the need of change detection, object identification, and exploratory analysis on large-scale high-resolution imagery data in remote sensing applications.     

A comparison of simulated monsoon circulations and snow accumulation in Asia during the mid-Holocene and at the Last Glacial Maximum

Tropical climatology through the last glacial cycle is believed to have ranged from colder, windier conditions at the Last Glacial Maximum (LGM) to relatively warm, stable conditions during the Holocene. Changes in strength of the South Asian monsoon have previously been determined from a variety of proxy data and have been attributed primarily to changes in radiative forcing, although tropical sea surface temperature (SST) is known to play a fundamental role in regulating monsoon strength and is also believed to have changed throughout the late Quaternary.In this study, the monsoons simulated in a coupled atmosphere–ocean general circulation model (GCM) configured for the mid-Holocene (6000 years B.P.) and for the LGM (21,000 years B.P.) are compared. The colder and windier conditions simulated for the LGM produced a summer monsoon whose westerly winds are stronger and whose precipitation and snowfall into the eastern Himalaya are increased, with drier conditions over the rest of the Indian subcontinent and over most of southwest Asia.The mid-Holocene monsoon circulation is stronger than today, and annual mean snow accumulation is increased over the northwestern Himalaya. These changes in precipitation and snow accumulation are analyzed in terms of the altered atmospheric circulations, which are in turn driven by changes in radiative forcing, sea surface temperatures, and sea surface height. All of these factors are therefore demonstrated to be important in governing the spatial distribution of snow and ice deposition in the Himalaya during the late Quaternary, and are likely to have contributed to the observed asynchroneity of Himalayan glaciation and Northern Hemisphere ice sheet volume.