挠力河流域河流径流量对气候变化的敏感性分析

流域径流量对气候变化的敏感性分析是理解气候变化对流域水资源影响的重要手段。本文利用非更新式人工神经网络(ANN)模型,以年平均降雨、年最低气温和最高气温为输入参数,年平均径流量为输出变量,构建了三江平原挠力河流域的径流量预测ANN模型;并根据IPCC第四次报告的气候变化模式,设定了9种不同的气候变化情景,利用构建的ANN模型分析了流域径流量对气候变化的敏感性。结果表明:构建的人工神经网络模型能够较好的模拟径流量,可用于气候变化的敏感性分析;挠力河流域上游径流量对气候变化的敏感性要大于中游区域的,降水变化对径流量的影响大于气温对其产生的影响。     

基于PCA和BP神经网络的径流预测

径流预测为流域水资源的合理开发利用与统筹配置提供依据。运用多元线性回归、主成分回归、BP神经网络及主成分分析和BP神经网络相结合的方法,对新疆呼图壁河流域石门水文站2009-2011年各月径流量进行预测,并采用相关系数、确定性系数及均方根误差对各模型预测精度进行比较。结果表明：（1）神经网络等智能算法具有高速寻优的能力,对短时间尺度的月径流量的预测结果较好;（2）主成分回归等常规算法能充分反映出某地区径流的年际的稳定性,对全年径流总量的模拟精度较高;（3）主成分分析和BP神经网络相结合的方法,提高了神经网络的收敛速度,同时降低了局部极值的影响,优于简单的BP神经网络,适用于呼图壁河月径流量预测。     

Comparison of Volterra Model and Artificial Neural Networks for Rainfall–Runoff Simulation

This study evaluates the performances of two distinct linear and non-linear models for simulating non-linear rainfall–runoff processes and their applications to flood forecasting in the Navrood River basin, Iran. Due to the excellent capacity of the artificial neural networks [multilayer perceptron (MLP)] and Volterra model, these models were used to approximate arbitrary non-linear rainfall–runoff processes. The MLP model was trained using two different training algorithms. The Volterra model was applied as a linear model [the first-order Volterra (FOV) model] and solved using the traditional ordinary least-square (OLS) method. Storm events within the Navrood River basin were used to verify the suitability of the two models. The models’ performances were evaluated and compared using five performance criteria namely coefficient of efficiency, root mean square error, error of total volume, relative error of peak discharge, and error of time for peak to arrive. Results indicated that the non-linear MLP models outperform the linear FOV model. The latter was ineffective because of the non-linearity of the rainfall–runoff process. Moreover, the OLS method is inefficient when the FOV model has many parameters that must be estimated.     

陇中黄土丘陵沟壑区人工草地土壤水蚀预测模型

针对陇中黄土丘陵沟壑区土壤水蚀过程复杂且难以有效预测的问题,以定西市安家沟水土保持试验站2005—2016年1~12月人工草地径流场试验数据为主要来源,将流域月降雨量、月侵蚀性降雨量、月径流量、月降雨强度、径流场面积、径流场坡度、土壤砂粒含量、土壤粘粒含量8个因子作为输入因子,月土壤水蚀量作为输出,运用偏最小二乘法(Partial Least-Squares Regression,PLSR)和长短期记忆(Long Short-Term Memory,LSTM)循环神经网络建立人工草地土壤水蚀预测模型,并利用BP(Back Propagation)、RNN(Recurrent Neural Network)、LSTM常见神经网络模型,对模型的有效性进行评估。结果表明：PLSR将模型8个输入因子减少为4个,从而有效解决LSTM神经网络模型对样本数量要求过高的问题; PLSR和LSTM神经网络模型的结合可以有效提高模型对人工草地土壤水蚀过程的预测精度和收敛速度,预测结果的平均相对误差小于4%,相关系数高于其他3种神经网络模型,而迭代次数、均方根误差和平均绝对误差均低于其他3种模型;研究发现坡度对人工草地土壤水蚀过程影响较为明显,降雨量小于25 mm时,人工草地土壤水蚀量不会随坡度增加而明显增长,但当降雨量超过25 mm时,人工草地土壤水蚀量会随坡度明显增加。 PLSR LSTM神经网络土壤水蚀预测模型可以准确预测陇中黄土丘陵沟壑区人工草地土壤水蚀量,为该地区水土流失的准确预报提供新的思路和方法。     

Natural runoff changes in the Yellow River Basin

The driving factors of runoff changes can be divided into precipitation factor and non-precipitation factor, and they can also be divided into natural factor and human activity factor. In this paper, the ways and methods of these driving factors impacting on runoff changes are analyzed at first, and then according to the relationship between precipitation and runoff, the analytical method about impacts of precipitation and non-precipitation factors on basin's natural runoff is derived. The amount and contribution rates of the two factors impacting on natural runoff between every two adjacent decades during 1956-1998 are calculated in the Yellow River Basin (YRB). The results show that the amount and contribution rate of the two factors impacting on natural runoff are different in different periods and regions. For the YRB, the non-precipitation impact is preponderant for natural runoff reduction after the 1970s. Finally, by choosing main factors impacting on the natural runoff, one error back-propagation (BP) artificial neural network (ANN) model has been set up, and the impact of human activities on natural runoff reduction in the YRB is simulated. The result shows that the human activities could cause a 77×108 m3·a-1 reduction of runoff during 1980-1998 according to the climate background of 1956-1979.     

策勒河出山径流特征及其趋势

根据策勒河近50年水文及气象资料,利用经验频率、Mann-Kendall非参数检验和非线性回归分析等方法,分析策勒河径流量的变化特征、变化趋势及其对气候的响应.结果表明:策勒河径流年内分配不均,夏、秋季偏丰,5～9月径流量占全年88.70%;冬、春季偏枯,10月到次年4月径流量占全年11.30%,策勒河变差系数为0.21,径流最大值与最小值之比为2.53,河流年际变化相对稳定;Mann-Kendall检验得出近十年出山径流量有减少的趋势;月平均流量与月平均降水量、月平均蒸发量和月平均气温之间存在复杂的非线性关系,非线性回归模型模拟的月平均流量与实测值对比,确定系数为0.73,模拟效果较好,月平均蒸发量与月平均流量呈负相关.     

Natural runoff changes in the Yellow River Basin

1IntroductionThe driving factors of runoff changes can be divided into precipitation factor and non-precipitation factor, and they can also be divided into natural factor and human activity factor. The influence of the natural factor includes precipitation reduction, precipitation features (for example, spatio-temporal distribution and intensity), landuse natural changes and so forth. All of these can cause runoff changes. Temperature, evaporation, topography, soil and geological environment i…     

50年来秦岭金钱河流域水文特征及其对降水变化的响应

运用集中度和集中期、Kendall秩相关系数、R/S分析法、降水—径流双累积曲线模型及其他数理统计方法,分析了金钱河流域径流的变化特征,探讨了年际、季节及月尺度上径流变化的趋势并预测了未来趋势,用集中期指标反映了径流对降水响应的滞后效应,并定量分析了降水变化和人类活动对径流变化的贡献率。结果表明:50 a来径流量呈现出显著的递减趋势(p<0.05),递减率为34.33 m3(/s 10a),Hurst指数H=0.669>0.5,表明未来的一段时间内变化趋势与现在相同;1~12月各月径流均表现为下降趋势。流域内径流对降水的响应存在滞后效应,50 a径流对降水平均每年滞后23.6 d,且滞后天数具有明显上升趋势。50 a来径流系数呈极显著减小趋势,降水量转化为径流的部分在逐年减少,被植物截留、填洼、入渗和蒸发的部分增加;径流发生突变后比突变前径流系数降低了35.2%。50 a来降水变化对径流变化的影响率为53.4%,高于人类活动影响率46.6%,是导致径流变化的主要原因,人类活动为次要原因。     

黄河流域径流变化

The driving factors of runoff changes can be divided into precipitation factor and non-precipitation factor, and they can also be divided into natural factor and human activity factor. In this paper, the ways and methods of these driving factors impacting on runoff changes are analyzed at first, and then according to the relationship between precipitation and runoff, the analytical method about impacts of precipitation and non-precipitation factors on basin's natural runoff is derived. The amount and contribution rates of the two factors impacting on natural runoff between every two adjacent decades during 1956-1998 are calculated in the Yellow River Basin (YRB). The results show that the amount and contribution rate of the two factors impacting on natural runoff are different in different periods and regions. For the YRB, the non-precipitation impact is preponderant for natural runoff reduction after the 1970s. Finally, by choosing main factors impacting on the natural runoff, one error back-propagation (BP) artificial neural network (ANN) model has been set up, and the impact of human activities on natural runoff reduction in the YRB is simulated. The result shows that the human background of 1956-1979.     

基于主成分分析的BP神经网络在城市建成区面积预测中的应用——以北京市为例

城市建成区规模受社会、经济、城市环境等诸多因素影响, 传统统计方法难以准确预测城 市建成区的面积。人工神经网络具有良好的非线性映射逼近性能, 在各类预测研究中得到了广泛 的应用, 尤其是BP 神经网络。主成分分析可以在有效保留数据信息前提下对数据进行降维, 它 与BP 神经网络的结合主要在数据输入端, 通过减少输入层神经元个数, 增强网络性能, 提高预 测精度。本文以北京市为例, 综合运用主成分分析和BP 神经网络方法建立预测模型, 以1986~ 2003 年数据为学习样本, 以2004 年数据为检验样本, 对2005 年北京市城市建成区面积进行模 拟预测。预测结果表明, 基于主成分分析的BP 神经网络预测结果与实际值的相对误差为2.8%, 比传统BP 神经网络预测精度提高1.8 个百分点, 网络训练收敛速度也更快, 其预测精度和效率 都有不同程度的改善。     

Climate transformation to warm-humid and its effect on river runoff in the source region of the Yellow River

The change characteristics and trends of the regional climate in the source region of the Yellow River, and the response of runoff to climate change, are analyzed based on observational data of air temperature, precipitation, and runoff at 10 main hydrological and weather stations in the region. Our results show that a strong signal of climate shift from warm-dry to warm-humid in the western parts of northwestern China （Xinjiang） and the western Hexi Corridor of Gansu Province occurred in the late 1980s, and a same signal of climate change occurred in the mid-2000s in the source region of the Yellow River located in the eastern part of northwestern China. This climate changeover has led to a rapid increase in rainfall and stream runoff in the latter region. In most of the years since 2004 the average annual precipitation in the source region of the Yellow River has been greater than the long-term average annual value, and after 2007 the runoff measured at all of the hydrologic sections on the main channel of the Yellow River in the source region has also consistently exceeded the long-term average annual because of rainfall increase. It is difficult to determine the prospects of future climate change until additional observations and research are conducted on the rate and temporal and spatial extents of climate change in the region. Nevertheless, we predict that the climate shift from warm-dry to warm-humid in the source region of the Yellow River is very likely to be in the decadal time scale, which means a warming and rainy climate in the source region of the Yellow River will continue in the coming decades.     

基于小波变换和GRNN神经网络的黑河出山径流模型

对黑河山区流域月降水量和气温做Harr小波变换,并作为GRNN神经网络的输入,对黑河出山径流进行模拟和预测验证,效果较好。应用全球变化成果,在不同的气候情景下,对黑河出山径流进行预测。结果表明,黑河出山径流在未来一段时间内,径流量会有一定程度的增加,最终会减少。但模型对气温反应不敏感。去除气温重构的细节系数后,气温也成为一个敏感因素,但径流量却随气温的增加而增加。可推断,引进Haar小波变换的GRNN神经网络模型可应用于径流量对气温不敏感的流域。     

多源再分析降水数据在拉萨河流域应用对比研究

针对资料稀缺地区水文模拟计算难题,开展多源再分析降水数据在拉萨河流域应用对比研究,本文基于HIMS系统构建了拉萨河流域分布式水文模型,以气象站实测数据为参照,对比分析了中国区域地面降水格点日值数据集和中国区域高时空分辨率地面气象要素驱动数据集两套遥感再分析数据集的气象数据在拉萨河流域的径流模拟效果。结果表明：在日和月时间尺度上,气象站实测降水数据的径流模拟精度最好,驱动集降水数据径流模拟结果要好于网格点降水数据。总体上,基于气象站实测降水数据的径流模拟纳西效率系数为0.86(日过程)和0.93(月过程),相关系数均在0.9以上。基于两类再分析数据的降水径流模拟纳西效率系数均在0.7(日过程)和0.8(月过程)以上,相关系数均在0.9左右。对于资料稀缺地区,多源再分析降水数据是重要的可用数据来源。借助于降水—径流模型,探讨多源再分析降水数据对径流模拟精度的影响,是完善多源再分析降水数据产品质量的一个重要环节。     

Precipitation-runoff simulation for a Himalayan River Basin, India using artificial neural network algorithms

In the Himalayan regions, precipitation-runoff relationships are amongst the most complex hydrological phenomena, due to varying topography and basin characteristics. In this study, different artificial neural networks (ANNs) algorithms were used to simulate daily runoff at three discharge measuring sites in the Himalayan Kosi River Basin, India, using various combinations of precipitation-runoff data as input variables. The data used for this study was collected for the monsoon period (June to October) during the years of 2005 to 2009. ANNs were trained using different training algorithms, learning rates, length of data and number of hidden neurons. A comprehensive multi-criteria validation test for precipitation-runoff modeling has been undertaken to evaluate model performance and test its validity for generating scenarios. Global statistics have demonstrated that the multilayer perceptron with three hidden layers (MLP-3) is the best ANN for basin comparisons with other MLP networks and Radial Basis Functions (RBF). Furthermore, non-parametric tests also illustrate that the MLP-3 network is the best network to reproduce the mean and variance of observed runoff. The performance of ANNs was demonstrated for flows during the monsoon season, having different soil moisture conditions during period from June to October.     

人工智能计算技术在新疆干旱区典型绿洲土壤盐分预测中的应用

针对新疆渭干河-库车河三角洲绿洲土壤盐分动态监测中存在的方法问题,首先用灰色关联度模型分析影响形成土壤盐渍化的各因子,并确定其与土壤盐分之间的关联度,然后将人工智能计算技术引入土壤盐分的预测中,经过多次调整网络结构和参数,建立了预测表层土壤盐分的BP神经网络模型和RBF神经网络模型。结果表明：以潜在蒸散量、地下水埋深、地下水矿化度、土壤电导率、总溶解固体、pH值、坡度和土地利用类型8个因素为输入因子,以土壤含盐量为输出因子的BP网络模型和RBF网络模型可有效模拟土壤盐分与其影响因子之间的内在复杂关系,并且有较高的精度。BP网络模型预测误差略低于RBF神经网络。本研究可为分析和预测土壤盐渍化动态规律提供一种有效可行的新途径,是对传统土壤盐分动态研究的补充。     

基于SWAT的典型沼泽性河流径流演变的气候变化响应研究——以三江平原挠力河为例

SWAT（Soil and Water Assessment Tool）是流域尺度的分布式水文模型,具有评价气候变化对径流影响的优势。利用SWAT模拟了三江平原典型沼泽性河流——挠力河流域3个水文站（上游的宝清站、保安站和中游的菜嘴子站）1974～1992年年径流量演变特征及变化趋势。在对模型参数敏感性分析的基础上,对模型的参数进行了率定和验证,率定期为1975～1982年,验证期为1983～1992年。率定期的模型效率指数ENS都大于0．85,皮尔逊相关系数都大于0．9,相对误差都小于10％;验证期,模型效率指数ENS有所减小,但也都大于0．61,模型对年径流的模拟结果令人满意。将率定的SWAT应用于气候变化的水文响应研究,结果发现,1995～2004年相对1975—1985年的年径流量变化只有部分是由气候因素引起的,气候因素对3个水文站（宝清站、保安站和菜嘴子站）的年径流量变化的影响率分别为25．7％、11．4％、39．9％,说明还有其他因素影响研究区的年径流量。     

三江平原沼泽性河流径流演变的驱动力分析

在分析三江平原典型沼泽性河流挠力河径流演变特征及趋势的基础上，应用灰色关联分析和径向基函数网络等方法，探讨了引起径流量减少和发生突变的原因，分析结果表明：当地河川径流演化与沼泽化流的地理特征以及近50年来沼泽及沼泽化土地的大规模开垦和水资源的开发利用有密切的关系，人类活动是本区河川径流演变的主要驱动力，气候变化在径流演变中所起的作用相应减少，沼泽湿地对区域水系统的水量平衡产生着重要的影响，在流域下垫面已明显变化，水文循环出现变异的情况下，必须加强沼泽湿地保护的水的调控和管理。     

策勒河流域荒漠类型特征研究

策勒河流域荒漠类型多样,在干旱区具有典型代表性。基于Landsat5的TM遥感影像数据、数字高程模型（DEM）数据,利用GIS技术,结合野外调查、样品采集及实验室粒度、有机质、盐分分析,对策勒河流域荒漠类型分类,并编制荒漠类型分布图。据地貌类型、物质组成、植被盖度等指标将流域内荒漠分为2个一级类,7个二级类,14个三级类,建立策勒河流域荒漠类型分类系统。流域荒漠自下游向上游具有依次是沙漠、砾漠、岩漠、土漠和寒漠空间分布的规律;荒漠总面积2 416.76 km²,占流域总面积的70.45%,砾漠占荒漠总面积的45.52%,为最大一类,泥漠所占比重最小。可为流域环境保护、荒漠化防治提供依据,为其他干旱区荒漠类型研究提供参考。     

黄河上游地区近千年气候变化的模拟重建

运用BP人工神经网络较好地建立了全球气候模式模拟数据与区域气候之间的关系,拟合了黄河上游沙漠河谷地区的近千年温度、降水序列。在气候信号年代际和百年际变化特征上,拟合结果较为理想,但对极值的拟合能力较差,尤其是冬季温度和夏季降水的拟合极值偏差较大。拟合结果表明该地区近千年存在中世纪暖期、小冰期和现代暖期,且小冰期降温在冬季更为明显,冬季平均气温小冰期比中世纪暖期低2 ℃。降水的千年变化趋势较温度略微平缓,尤其冬季降水无明显趋势变化。空间分布显示20世纪暖期在近千年是最暖的,但降水较中世纪暖期偏少。     

气候与土地利用变化对汉江流域径流的影响

作为联结大气圈和地圈的纽带,水文循环同时承受气候变化和土地利用/覆被变化(LUCC)的双重影响,然而大多数的水文响应研究主要关注未来气候变化对径流的影响,忽略了未来LUCC的作用。因此,本文的研究目的是评估未来气候变化和LUCC对径流的共同影响。首先采用2种全球气候模式(BCC-CSM1.1和BNU-ESM)输出,基于DBC降尺度模型得到未来气候变化情景;然后,利用CA-Markov模型预测未来LUCC情景;最后,通过设置不同的气候和LUCC情景组合,采用SWAT模型模拟汉江流域的未来径流过程,定量评估气候变化和LUCC对径流的影响。结果表明:①未来时期汉江流域的年降水量、日最高、最低气温相较于基准期(1966—2005年),在RCP 4.5和RCP 8.5浓度路径下,分别增加4.0%、1.8℃、1.6℃和3.7%、2.5℃、2.3℃;②2010—2050年间,流域内林地和建设用地的面积占比将分别增加2.8%和1.2%,而耕地和草地面积占比将分别减少1.5%和2.5%;③与单一气候变化或LUCC情景相比,气候变化和LUCC共同影响下的径流变化幅度最大,在RCP 4.5和RCP 8.5浓...