首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献，搜索用时 31 毫秒
1.
In the current study, the efficiency of Wavelet-based Least Square Support Vector Machine(WLSSVM) model was examined for prediction of daily and monthly Suspended Sediment Load(SSL) of the Mississippi River. For this purpose, in the first step, SSL was predicted via ad hoc LSSVM and Artificial Neural Network(ANN) models; then,streamflow and SSL data were decomposed into subsignals via wavelet, and these decomposed sub-time series were imposed to LSSVM and ANN to simulate discharge-SSL relationship. Finally, the ability of WLSSVM was compared with other models in multistep-ahead SSL predictions. The results showed that in daily SSL prediction, LSSVM has better outcomes with Determination Coefficient(DC)=0.92 than ad hoc ANN with DC=0.88. However unlike daily SSL, in monthly modeling, ANN has a bit accurate upshot.WLSSVM and wavelet-based ANN(WANN) models showed same consequences in daily and different in monthly SSL predictions, and adding wavelet led to more accuracy of LSSVM and ANN. Furthermore,conjunction of wavelet to LSSVM and ANN evaluated via multi-step-ahead SSL predictions and, e.g.,DC LSSVM=0.4 was increased to the DC WLSSVM=0.71 in 7-day ahead SSL prediction. In addition, WLSSVM outperformed WANN by increment of time horizon prediction.  相似文献   

2.
?????С???????????????????LS-SVM??????????μ???????????????????μ????????н???С???????????C-C??????????????????????????????????????????????????????????????????LS-SVM??????н?????????BP??????????????????????????????????????С???????LS-SVM????????????????????н??????Ч????  相似文献   

3.
基于机器学习的参考作物蒸散量估算研究   总被引：2，自引：0，他引：2  
参考作物蒸散量（Reference Evapotranspiration, ET0）的准确估算对区域水资源管理和分配、流域水量平衡以及气候变化等研究具有重要作用。新疆地处我国西北干旱地区,水资源供需矛盾尖锐,精确估算该地区的ET0有助于其科学合理地调配水资源,缓解水资源供需压力。FAO推荐的Penman-Monteith法是计算ET0的标准方法,但该方法需要多项气象因子,而新疆地区气象站点较少且分布不均,精确完备的气象数据在新疆大部分区域难以获取。因此,如何使用有限气象因子获取高精度的ET0在新疆地区备受关注。本文基于中国气象数据网提供的新疆地区1980—2019年的地面气候资料日值数据集,在日和月尺度下,通过对最高气温Tmax、最低气温Tmin、平均气温Tavg、风速U2、相对湿度RH和日照时数n共6项气象因子进行敏感性分析,形成不同的气象因子组合;然后使用SVM、RF、GBDT和ELM 4种机器学习算法,以FAO-56 PM计算值为标准值,对新疆地区的ET0进行了拟合建模;最后,从拟合精度、稳定性和计算代价3个方面对模型进行评价。研究表明：① 在新疆地区,ET0对RH、Tmax和U2敏感系数级别为高,平均敏感系数分别为-0.516、0.283和0.266;n为中等,平均敏感系数为0.124;Tmin和Tavg为低,平均敏感系数分别为-0.016和-0.003;② 在日尺度,各算法在RH、Tmax、U2和n这4项气象因子为输入时精度较高（RMSE<0.5 mm/day,R2>0.95）,可对ET0进行精确估算;在月尺度,各算法使用RH、Tmax和U2这3项参数便可对ET0进行精确估算。SVM和GBDT这2种算法在日尺度和月尺度都有较好的适用性,可在相应尺度下使用较少气象因子替代FAO-56 PM公式对ET0进行估算。  相似文献   

4.
Spatial downscaling methods are widely used for the production of bioclimatic variables(e.g. temperature and precipitation) in studies related to species ecological niche and drainage basin management and planning. This study applied three different statistical methods, i.e. the moving window regression(MWR), nonparametric multiplicative regression(NPMR), and generalized linear model(GLM), to downscale the annual mean temperature(Bio1) and annual precipitation(Bio12) in central Iran from coarse scale(1 km × 1 km) to fine scale(250 m ×250 m). Elevation, aspect, distance from sea and normalized difference vegetation index(NDVI) were used as covariates to create downscaled bioclimatic variables. Model assessment was performed by comparing model outcomes with observational data from weather stations. Coefficients of determination(R2), bias, and root-mean-square error(RMSE) were used to evaluate models and covariates. The elevation could effectively justify the changes in bioclimatic factors related to temperature and precipitation. Allthree models could downscale the mean annual temperature data with similar R2, RMSE, and bias values. The MWR had the best performance and highest accuracy in downscaling annual precipitation(R2=0.70; RMSE=123.44). In general, the two nonparametric models, i.e. MWR and NPMR, can be reliably used for the downscaling of bioclimatic variables which have wide applications in species distribution modeling.  相似文献   

5.
The quality of debris flow susceptibility mapping varies with sampling strategies. This paper aims at comparing three sampling strategies and determining the optimal one to sample the debris flow watersheds. The three sampling strategies studied were the centroid of the scarp area(COSA), the centroid of the flowing area(COFA), and the centroid of the accumulation area(COAA) of debris flow watersheds. An inventory consisting of 150 debris flow watersheds and 12 conditioning factors were prepared for research. Firstly, the information gain ratio(IGR) method was used to analyze the predictive ability of the conditioning factors. Subsequently, 12 conditioning factors were involved in the modeling of artificial neural network(ANN), random forest(RF) and support vector machine(SVM). Then, the receiver operating characteristic curves(ROC) and the area under curves(AUC) were used to evaluate the model performance. Finally, a scoring system was used to score the quality of the debris flow susceptibility maps. Samples obtained from the accumulation area have the strongest predictive ability and can make the models achieve the best performance. The AUC values corresponding to the best model performance on the validation dataset were 0.861, 0.804 and 0.856 for SVM, ANN and RF respectively. The sampling strategy of the centroid of the scarp area is optimal with the highest quality of debris flow susceptibility maps having scores of 373470, 393241 and 362485 for SVM, ANN and RF respectively.  相似文献   

6.
提出一种基于主成分分析（PCA）的ZTD时空建模方法,并利用GNSS连续运行参考站获取的ZTD数据,建立香港、云南、中国3个区域范围的ZTD时空模型。结果表明,所建立的区域对流层延迟时空模型不仅精度明显高于Saastamoinen、EGNOS和UNB3m模型,而且建模过程简单,模型参数较少,使用方便。  相似文献   

7.
针对SVM模型在基坑沉降预测领域存在参数选择困难和基于单点数据建模的缺点,建立顾及邻近点的PSO-SVM模型。采用PSO-SVM模型进行最优训练样本数量研究,结果表明短期样本的预测效果最优。将邻近点沉降变形值作为影响基坑沉降的因素引入到改进的PSO-SVM模型中,实例表明,在短期样本数据下顾及邻近点的PSO-SVM模型的拟合精度优于PSO-SVM模型,而在中长期样本条件下预测精度不佳。针对该缺点提出组合多尺度一维小波分解函数和柯西分布函数来改进顾及邻近点的PSO-SVM模型,实验结果表明,顾及邻近点的改进PSO-SVM模型可有效解决参数选择困难和单点建模的问题,适用于不同样本数量下的沉降变形预测,具有较高的预测精度。  相似文献   

8.
快速、准确地从卫星影像中提取水体信息一直是遥感应用的热点问题,在水资源管理、水环境监测和灾害应急管理等领域极具应用价值。虽然目前已有多种针对Landsat系列影像的水体提取方法,但由于地理位置、地形和水体形态等环境背景因素的影响,导致同种方法在不同的环境背景中呈现出不同的提取效果。本文针对人为影响严重、影像明暗对比强烈的城区（北京怀柔县城周边）以及地形起伏明显、水体细小的非城区（北京密云水库周边） 2种典型背景环境,选择波段设置略有差异的Landsat 5（2009年）和Landsat 8（2019年）卫星影像,对比了常用的指数法（NDWI和MNDWI）和分类法（最大似然法和支持向量机）在水体信息提取方面的优势和不足。结果表明：在城区背景中,SVM的准确性最高（总体精度>97%）;在非城区背景中,MNDWI与SVM的精度相当（总体精度>95%）,前者更适用于水体的快速提取,而后者提取的山间细碎河流更完整,且在Landsat 8中应用的效果更好。该研究为不同环境背景下水体提取方法的选择提供了参考。  相似文献   

9.
Recently, study in past trends of climate variables gained significant consideration because of its contribution in adaptions and mitigation strategies for potential future changes in climate, primarily in the area of water resource management. Future interannual and inter-seasonal variations in maximum and minimum temperature may bring significant changes in hydrological systems and affect regional water resources. The present study has been performed to observe past(1970-2010) as well as future(2011-2100)spatial and temporal variability in temperature(maximum and minimum) over selected stations of Sutlej basin located in North-Western Himalayan region in India. The generation of future time series of temperature data at different stations is done using statistical downscaling technique. The nonparametric test methods, modified Mann-Kendall test and Cumulative Sum chart are used for detecting monotonic trend and sequential shift in time series of maximum and minimum temperature. Sen’s slope estimator test is used to detect the magnitude of change over a period of time on annual and seasonal basis. The cooling experienced in annual TMax and TMin at Kasol in past(1970-2010) would be replaced by warming in future as increasing trends are detected in TMax during 2020 s and 2050 s and in TMin during 2020 s, 2050 s and 2080 s under A1 B and A2 scenarios. Similar results of warming are also predicted at Sunnifor annual TMin in future under both scenarios which witnessed cooling during 1970-2010. The rise in TMin at Rampur is predicted to be continued in future as increasing trends are obtained under both the scenarios. Seasonal trend analysis reveals large variability in trends of TMax and TMin over these stations for the future periods.  相似文献   

10.
A multilayer study of pCO2 for the Yellow and South China Seas in the surface waters was conducted based on data from four cruises sponsored by the China SOLAS Project in 2005 and 2006, including data for the surface microlayer (SML), sub- surface layer (SSL) and surface layer (SL). The carbon fluxes across the air-sea interface were calculated. The results showed that the pCO2 values in the surface waters of the study area decreased in the following order: pCO2 SML pCO2 SSL pCO2 SL. The highest val- ues were found in March for all SML, SSL and SL, followed by those in April, and the lowest were in May. The pCO2 values had a significant positive correlation with temperature or salinity. While there was no relationship between pCO2 and longitude, there was a significant negative correlation between it and latitude, i.e., 'high latitude low pCO2'. By using four calculation models, the carbon dioxide fluxes ( FCO2 ) in spring in the Yellow and South China Seas, which were found to act as a 'sink' of atmospheric CO2, were preliminarily estimated on the basis of the pCO2 data in the SML to be -7.00×106t C and -22.35×106t C, respectively. It is suggested that the FCO2 calculated on the basis of pCO2 data in the SML is more reliable than that calculated on the basis of those in the SL.  相似文献   

11.
山区降水较集中,但降水测站多位于山谷或人口密集区,代表性差。遥感和再分析降水产品能提供时空分布连续的数据,不受地形条件限制。柴达木盆地中心属干旱荒漠区,水是制约该区开发的首要条件,其四周属高寒山区,降水相对较多,但降水监测十分薄弱。为获取该区相对精确的降水时空分布信息,本文评估了4套高分辨率降水产品（CMADS、TRMM、GPM和MSWEP）的适用性。首先基于地面站点数据评估它们在不同时空尺度上的精度,并分析它们在柴达木盆地的空间分布和年内分配特征。然后,以盆地东南隅的无测站山区香日德河流域为研究区,利用降水产品驱动SWAT模型来评估它们的分布式水文模拟适用性。结果表明：① MSWEP在年、月尺度上与站点降水的吻合程度最高（R ≥ 0.79,PBIAS = 0.5%）,其次是GPM和TRMM,CMADS精度最低（R ≥ 0.64,PBIAS = 5.8%）;② 从降水精度与站点高程的关系来看,降水产品在相对低海拔区容易高估站点降水,而在相对高海拔区常低估实际降水;③ 在香日德河流域,MSWEP（NSE = 0.64）在基准期（2009—2012年）的径流模拟表现明显好于其它降水产品（NSE = 0.36~0.59）,变化期（2013—2016年）表现最好的是CMADS（NSE = 0.75,其余产品NSE = 0.53~0.68）。本研究可为缺资料干旱山区获取精确的降水时空信息和后续水资源的科学管理与规划提供重要支撑。  相似文献   

12.
针对现有区域天顶对流层延迟(ZTD)模型属于函数或格网型,参数固定,且难以表达ZTD时空快速变化特性等问题,提出一种基于小波变换、傅里叶级数拟合、自回归(AR)、支持向量回归(SVR)的组合预报新模型构建方法。该模型在时域内对ZTD序列进行小波变换,分解出低频和高频序列。低频序列采用傅里叶级数拟合成时间函数,高频序列则由AR进行预报。在空间域内利用SVR建立位置参数向傅里叶级数参数的映射。在该模型中输入时间与位置信息即可获取ZTD预报值。利用94个GNSS基站2 a的ZTD数据进行建模,24个GNSS基站1 a的ZTD数据进行预测对比。结果表明,实测值与模型预报值之间的平均偏差为-2.02 mm,均方根误差为3.07 cm,优于大部分区域ZTD模型。在伪距单点定位测试中,该模型能够显著提高定位精度。实验表明,该组合模型具有较高的预报精度和可靠性,具有一定的应用价值。  相似文献   

13.
Daily meteorological data are the critical inputs for distributed hydrological and ecological models. This study modified mountain microclimate simulation model （MTCLIM） with the data from 19 weather stations, and compared and validated two methods （the MTCLIM and the modified MTCLIM） in the Qilian Mountains of Northwest China to estimate daily temperature （i.e., maximum temperature, minimum temperature） and precipitation at six weather stations from i January 2000 to 31December 2009. The algorithm of temperature in modified MTCLIM was improved by constructing the daily linear regression relationship between temperature and elevation, aspect and location information. There are two steps to modify the MTCLIM to predict daily precipitation： firstly, the linear regression relationship was built between annual average precipitation and elevation, location, and vegetation index; secondly, the distance weight for measuring the contribution of each weather station on target point was improved by average wind direction during the rainy season. Several regression analysis and goodness-of-fit indices （i.e., Pearson＇s correlation coefficient, coefficient of determination, mean absolute error, root-mean-square error and modelingefficiency） were used to validate these estimated values. The result showed that the modified MTCLIM had a better performance than the MTCLIM. Therefore, the modified MTCLIM was used to map daily meteorological data in the study area from 2000 to 2009. These results were validated using weather stations with short time data and the predicted accuracy was acceptable. The meteorological data mapped could become inputs for distributed hydrological and ecological models applied in the Qilian Mountains.  相似文献   

14.
Fishery-independent surveys can provide high-quality data and support fishery assessment and management.Optimization of sampling design is crucial to increase the quality of fishery surveys.Crab pots are important fishing gears used to catch crabs.We analyzed the impacts of sampling design of crab pots on the abundance of Portunus trituberculatus in the Changjiang(Yangtze) River estuary to the Hangzhou Bay and its adjacent waters in East China Sea.The crab pots were cylindrical,240 mm in height and 600 mm in diameter of the iron ring.Our sampling designs(including fixed-station sampling,simple random sampling,stratified fixed-station sampling,and stratified random sampling),three number of stations(9,16,and 24),and three numbers of crab pots(500,1 000,and 3 000) were simulated and compared with the "true" abundance that obtained from bottom trawl survey s in the study area in 2007.The scenarios with16 stations were set in stratification as a control group for comparison with unstratified designs.Results show that simple random sampling can obtain more stable results than fixed-station sampling in the abundance estimation of P.trituberculatus.In addition,stratified sampling resulted in more accurate abundance than unstratified sampling.The accuracy of the simulated results improved with the increase of the number of stations.No remarkable differences in the re sults were found among the scenarios of different number of crab pots at each station.However,resource-intensive areas exerted great impacts on simulation results.Thus,prior information or pre-survey results about resource abundance and density distribution are necessary.This study may serve as a reference for future sampling designs of crab pots of P.trituberculatus and other species.  相似文献   

15.
开展三维地质建模的目标,不应当只是实现地质体框架的可视化表达,而应当同时实现地质大数据的聚合、管理、挖掘、分析和共享。然而,传统的方法和技术难以实现顾及地质语义的结构-属性一体化三维地质建模与耦合表达。多点地质统计学方法虽然便于多源数据、地质先验知识、结构-属性的融合建模,却仍然受到数据结构表达能力不足、三维训练图像难以获取和非平稳现象的限制。面向地质大数据集成与管理的要求,详细讨论了三维地质建模中的空间数据模型、基于多点地质统计学的结构-属性一体化集成建模方法、以及基于三维地质模型的地质大数据集成与管理的框架与模式。发展新型的面向地质结构-属性耦合表达的统一空间数据模型,以及知识驱动与数据驱动协同的三维地质结构-属性一体化集成建模技术体系,着力构建出地质大数据的聚合、集成、管理、挖掘和分析的可视化环境与操作平台,是未来三维地质建模领域的研究热点和前沿方向。   相似文献   

16.
选取中国大陆构造环境监测网（陆态网）提供的155个测站2014~2018年对流层延迟产品,基于BP-Adaboost算法将多个弱神经网络预测器集成为强预测器,建立新的无气象参数对流层延迟计算模型。利用陆态网2019年参与建模的141个建模测站、未参与建模的62个测站的对流层延迟产品和中国区域86个无线电探空站解算出的对流层延迟精确值对BP-Adaboost模型进行精度评定,结果表明,新模型的平均偏差分别为0.62 mm、-1.16 mm和12.32 mm,均方根误差分别为25.30 mm、26.72 mm和46.29 mm,优于常见的无气象参数模型;BP-Adaboost模型在内陆地区或海拔2 km以上地区具有更高的精度,能够满足中国大陆区域卫星导航用户实时对流层延迟改正的需求。  相似文献   

17.
为评价新西兰海潮负荷位移建模精度,利用新西兰189个GPS站11 a的实测数据,基于静态精密单点定位测定8个半日潮波及周日潮波的海潮负荷位移参数,并将其与7种全球海潮模型及4种地球模型计算的海潮负荷位移改正值进行比较。结果表明：1）TPXO7.2模型负荷位移改正值与GPS解算的海潮负荷位移参数最符合,M2、N2、O1和Q1潮波均方根误差在水平方向小于0.5 mm,垂直方向小于0.7 mm;2）不同地球模型对确定海潮负荷位移的影响主要体现在M2和N2潮波;3）GPS估值和海潮模型值之间的残差矢量呈现出大小及方向上的区域一致性,部分站点异常的残差值可能反映出当前SNREI地球模型的缺陷。  相似文献   

18.
Studies on susceptibility to debris flows at regional scale(100-1000 km~2) are important for the protection and management of mountain areas. To reach this objective, routing models, mainly based on land topography, can be used to predict susceptible areas rapidly while necessitating few input data. In this research, Flow-R model is implemented to create the susceptibility map for the debris flow of the Vizze Valley(BZ, North-Eastern Italy; 134 km~2). The analysis considers the model application at local scale for three sub-catchments and then it explores the model upscaling at the regional scale by verifying two methods to generate the source areas of debris-flow initiation. Using data of an extreme event occurred in the Vizze Valley(4 August 2012) and historical information, the modeling verification highlights that the propagation parameters are relatively simple to set in order to obtain correct runout distances. A double DTM filtering-using a threshold for the upslope contributing area(0.1 km~2) and a threshold for the terrain-slope angle(15°)-provides a satisfactory prediction of source areas and susceptibility map within the geological conditions of the Vizze Valley.  相似文献   

19.
???????е?Delaunay ???????????????????CORS???????繹????????????CORS????????桢?????????????????????????????????????????????????????????????о???????????????????????????CORS???????????????????????????????????????????о???????μ??????????????????????????????????Ч???  相似文献   

20.
In this paper, we use the optimum interpolation sea surface temperature (OISST) provided by the National Center for Environmental Prediction (NCEP) to replace the temperature in the top three layers in the ISHII data, and make use of the modified ISHII temperature data to calculate the thermosteric sea level (called modified steric sea level (SSL) hereafter). We subtract the modified SSL and the steric sea level (called ordinary SSL hereafter) derived from the ISHII temperature and salinity from the steric sea level (SSL) provided by the Gravity Recovery and Climate Experiment (GRACE), respectively, and find that the rms error of the difference of the former is obviously smaller than that of the latter. Therefore we reach the conclusion that under the assumption that the GRACE SSL is accurate, the modified SSL can reflect the true steric sea level more accurately. Making use of the modified SSL, we can find that the modified SSL in sea areas of different spatial scales shows an obvious rising trend in the upper 0-700 m layer for the period 1982-2006. The global mean SSL rises with a rate of 0.6 mm year-1 .The modified SSLs in sea areas of different spatial scales all show obvious oscillations with period of one year. There are oscillations with periods of 4-8 years in global oceans and with periods of 2-7 years in the Pacific. The Empirical Orthogonal Function method is applied to the sea areas of different spatial scales and we find that the first modes all have obvious 1-year period oscillations, the first mode of the global ocean has 4-8 year period oscillations, and that of the Pacific has 2-6 year period oscillations. The spatial distribution of the linear rising trend of the global modified SSL in the upper 0-700 m layer is inhomogeneous with intense regional characteristics. The modified SSL linear trend indicates a zonal dipole in the tropical Pacific, rising in the west and descending in the east. In the North Atlantic, the modified SSL indicates a meridional dipole, rising in the latitude band of 20°N-40°N and 45°N-65.5°N and descending obviously in the latitude band of 40°N-45°N.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号