新疆阿克苏河流域年径流时序特征分析

阿克苏河是塔里木河的主要支流,依据阿克苏河流域内5个代表站的40余年年径流实测资料,选取多种径流统计参数,分析流域范围内不同径流补给来源的径流年内分配规律和多年变化特征。结果表明:阿克苏河流域径流补给具有垂直地带性和多样化特点,径流时序特征与径流的补给来源有密切关系;径流年内分配极不均匀,集中程度高;而径流的多年变化变差系数小,丰、枯频率密度近似正态分布,无特大丰水年和枯水年,径流量多年变化趋势比较稳定。

基于S变换的潘家口水库入库径流时序多时间尺度分析

以1956—2003年间的潘家口入库径流时序为研究对象,采用S变换对该径流时序的演变特征进行了时频分析,得到了该水文时序频谱在时-频平面上的三维分布图,它清晰地显示了该时间序列在不同时段内的多时间尺度演变特征。然后再对得到的频谱图进行拓扑结构分析,总结了潘家口水库入库径流时序的宏观演变规律。结果表明,潘家口水库的年入库天然年径流近50年来总体上呈减少的趋势,其演变过程中存在16年和多个10年以下的多重变化周期：时间尺度为16年的周期分量波动特征较为平稳;但是对该径流时序演变特征起主要作用的是3~10年的周期分量,并且3~10年的多个周期分量在不同时段内对该径流时序波动特征的贡献不同。

新疆呼图壁河流域径流时序变化特征

依据呼图壁河石门水文站1956—2005年50 a径流量资料,选取多种径流量统计参数,分析流域范围内不同径流补给来源的径流年内分配规律和多年变化特征。结果表明:径流量时序特征与径流的补给来源有密切关系;径流量年内分配极不均匀,集中度高达66.47%;而径流量的多年变化变差系数小,仅为0.15。Kendall秩次检验显示在过去的50年,呼图壁河年径流表现出显著的递增趋势;而不同时间分辨率下的Hurst指数,尤其年径流的Hurst指数为0.8,表明径流具有强劲的持续性趋势,即年径流量在未来变化周期内将表现为持续增加的趋势。

汭河流域径流空间分布及时序变化分析

收集了汭河流域华亭、安口和袁家庵3处水文站建站以来到2018年年径流量资料,采用常规水文分析法对汭河径流空间分布及时序变化进行分析。得出汭河干流径流量分布随海拔增高而增大,海拔变化范围在1000~2900 m间,下游径流深变化80 mm左右,上游最高值在300 mm以上;汭河年径流量整体呈减少趋势,趋势显著;采用有序聚类法分析,汭河年径流量在1990年左右发生突变。

水土保持地区人类活动对汛期径流影响的估算

人类活动对于流域水文要素时变过程的影响显著,确定人类活动影响下水文要素时序的显著转折点、选出相对'天然'的序列,对于开展水土保持效益评价研究具有重要意义。以延河流域为例,采用数理统计方法推估获取人类活动影响下汛期径流时序的显著转折点,以显著转折点前的数据建立预测模型,对比分析汛期径流量的实测值与预测值,其差值即为人类活动对汛期径流的影响程度。较之传统研究方法,本文以汛期水文数据取代年均数据,以汛期降水径流综合系数代替年径流系数,从而减弱了黄土高原地区降水、径流变化幅度巨大的干扰影响。结果表明,从20世纪50年代末至80年代末,1970年为人类活动下延河流域径流时序的显著转折点,1970年以后人类活动对流域水文的影响更显著,1981年达到最大值,为72.04%。20世纪70、80年代人类活动对延河汛期径流量的影响一直处于负面减流状态。人为因素是延河径流量演变的主要驱动因子。

鄱阳湖流域天然径流变化特征与水旱灾害

为探讨鄱阳湖流域河川径流的变化过程和规律,深入了解径流特征及其与鄱阳湖水旱灾害的关系,利用时间序列分析方法,对鄱阳湖流域五河水系干流河段主要控制水文测站的天然径流系列进行了研究。研究显示,鄱阳湖五河径流年内分配集中程度在0.43到0.56之间,集中期为每年的5月底6月初,比鄱阳湖汛期提前1-2个月。径流多年变化变差系数变化在0.28～0.33之间,径流年际变率较大,在年代际变化上20世纪90年代径流增加尤其突出;径流序列呈长期的增加趋势,1998年后增加趋势变缓;鄱阳湖流域五河水系出现特大枯水年和丰水年的概率较大,出现平水年的概率略小。以上结果表明,五河汛期来水是影响鄱阳湖洪水的重要因素,而其形成、发展过程还受到长江中上游洪水的控制;径流的长期变化中,气候因素是引起鄱阳湖流域径流变化主导因素,水土流失和水利工程的建设等人为因素起着一定的辅助作用;径流序列枯、丰循环周期的交替变化过程,与过去几十年间流域内出现的干旱、洪水现象具有较好的一致性。

Wet-Dry Runoff Correlation in Western Route of South-to-North Water Diversion Project,China

The Western Route of the South-to-North Water Diversion Project is an important trans-basin diversion project to transfer water from the upstream Yangtze River and its tributaries （water-exporting area）, to the upstream of the Yellow River （water- importing area）. The long-term hydrologieal data from 14 stream gauging stations in the Western Route area and techniques including the pre-whitening approach, non-parametric test, Bayes, law, variance analysis extrapolation, and Wavelet Analysis are applied to identify the streamflow eharacteristics and trends, streamflow time series cross-correlations, wetness-dryness encountering probability, and periodicities that occurred over the last 50 years. The results show that the water-exporting area, water- importing area, and the streteh downstream of the water-exporting have synehronization in high-low flow relationship, whereas they display non- synchronization in long-term evolution. This corresponds to the complicated and variable climate of the plateau region. There is no obvious increasing or decreasing trend in runoff at any gauging station. The best hydrological eompensation probability for rivers where water is diverted is about 25% to lO%, and those rivers influenced significantly by diversion are the Jinsha and Yalong rivers. Proper planning and design of compensation reservoirs for the water-exporting area and stretch downstream of the water- exporting area can increase the hydrological compensation possibility from water-exporting area to the water-importing area, and reduce the impact on the stretch of river downstream of the water- exporting area.

Runoff Change of Naoli River in Northeast China in 1955–2009 and Its Influencing Factors

Runoff change and trend of the Naoli River Basin were studied through the time series analysis using the data from the hydrological and meteorological stations. Time series of hydrological data were from 1957 to 2009 for Bao′an station, from 1955 to 2009 for Baoqing station, from 1956 to 2009 for Caizuizi station and from 1978 to 2009 for Hongqiling station. The influences of climate change and human activities on runoff change were investigated, and the causes of hydrological regime change were revealed. The seasonal runoff distribution of the Naoli River was extremely uneven, and the annual change was great. Overall, the annual runoff showed a significant decreasing trend. The annual runoff of Bao′an, Baoqing, and Caizuizi stations in 2009 decreased by 64.1%, 76.3%, and 84.3%, respectively, compared with their beginning data recorded. The wet and dry years of the Naoli River have changed in the study period. The frequency of wet year occurrence decreased and lasted longer, whereas that of dry year occurrence increased. The frequency of dry year occurrence increased from 25.0%-27.8% to 83.9%-87.5%. The years before the 1970s were mostly wet, whereas those after the 1970s were mostly dry. Precipitation reduction and land use changes contributed to the decrease in annual runoff. Rising temperature and water project construction have also contributed important effects on the runoff change of the Naoli River.

Monthly discharge forecasting using wavelet neural networks with extreme learning machine

Accurate and reliable hydrological forecasting is essential for water resource management. Feedforward neural networks can provide satisfactory forecast results in most cases, but traditional gradient-based training algorithms are usually time-consum- ing and may easily converge to local minimum. Hence, how to obtain more appropriate parameters for feedforward neural networks with more precise prediction within shorter time has been a challenging task. Extreme learning machine （ELM）, a new training algorithm for single-hidden layer feedforward neural networks （SLFNs）, has been proposed to avoid these disad- vantages. In this study, a conjunction model of wavelet neural networks with ELM （WNN-ELM） is proposed for 1-month ahead discharge forecasting, The ~ trous wavelet transform is used to decompose the original discharge time series into several sub-series. The sub-series are then used as inputs for SLFNs coupled with ELM algorithm （SLFNs^ELM）; the output is the next step observed discharge. For comparison, the SLFNs-ELM and support vector machine （SVM） are also employed. Monthly discharge time series data from two reservoirs in southwestern China are derive] for validating the models. In addi- tion, four quantitative standard statistical performance evaluation measures are utilized to evaluate the model performance. The results indicate that the SLFNs-ELM performs slightly better than the SVM for peak discharge estimation, and the proposed model WNN-ELM provides more accurate forecast precision than SLFNs-ELM and SVM