中国云南小江流域泥石流暴发与ENSO的关联性

选取典型泥石流发育区云南小江流域为研究对象,利用典型泥石流沟滇北小江流域蒋家沟的长期观测资料,分析了小江流域泥石流暴发与该流域邻近的沾益及会泽常规气象站夏季（6—8月）降水的关系,以及小江流域及其周边地区夏季降水与ENSO的关系。结果表明,蒋家沟泥石流暴发的次数与夏季降水量有显著的正相关关系,而夏季降水和前期的Nino3区海表温度（SST）呈显著的负相关;每年泥石流发生的次数与首场泥石流发生的早晚关系密切,而激发首场泥石流的降水量与冬春Nino3区SST呈负相关;泥石流暴发次数与大雨日数关系密切,而大雨日数与Nino3区SST具有较好的对应关系。这说明冬春季Nino3区SST对小江流域泥石流的暴发次数有显著的影响,形成了ENSO与小江流域及蒋家沟泥石流发生的关联性。分析结果亦表明,Nino3区1月SST与当年蒋家沟泥石流次数具有显著的负相关关系,El Ni?o位相年泥石流少发而La Ni?a位相年泥石流多发。Nino3区SST变化最少要超前泥石流暴发4个月,因而ENSO可以为云南北部泥石流的预测预报提供一种指标信息,从而有可能利用ENSO冬季信息来预测小江流域及其周边地区（滇北）当年夏季泥石流活动。     

2010年汛期多模式对山东降水预报的检验

为提高数值预报模式在山东汛期降水的预报能力,为降水预报及模式物理参数方案选择和调整提供客观依据,对2010年汛期（6-9月）山东区域MM5、WRF-RUC（WRF快速循环同化系统）和T639模式24 h、48 h累积降水预报产品,进行晴雨、一般性降水和分量级降水TS评分及平均绝对误差、平均误差分析。结果表明：从晴雨预报准确率来看,3种模式相差不大;一般性降水和小雨预报,MM5模式评分结果最差,T639模式预报效果最好;中雨以上量级,24 h降水T639模式预报效果最好,特别是24 h大暴雨评分T639模式达到了10.37 %,而48 h降水T639模式预报效果下降明显。无论24 h降水还是48 h降水,除9月WRF-RUC模式平均绝对误差最小外,6-8月T639模式平均绝对误差均为最低,WRF-RUC模式24 h和48 h降水各月平均误差均为负偏差;不同的降水预报检验方案和天气过程类型对检验结果有一定的影响。     

基于余水位订正的海洋潮位推算关键技术研究

从分潮数目选择、调和常数稳定性优化两个关键环节入手,选择11个主要分潮用于潮位推算,并利用邻近长期站对中短期验潮站调和常数做差比订正,显著提高了中短期验潮站的潮位推算精度。     

Dam-break flood simulation under various likely scenarios and mapping using GIS: Case of a proposed dam on River Yamuna, India

The precision modeling of dam break floods can lead to formulation of proper emergency action plan to minimize flood impacts within the economic lifetime of the assets.Application of GIS techniques in integration with hydrological modeling for mapping of the flood inundated areas can play a momentous role in further minimizing the risk and likely damages.In the present study,dam break analysis using DAMBRK model was performed under various likely scenarios.Probable Maximum Flood (PMF)calculated for a return period of 1000 years using deterministic approach was adopted for dam break analysis of the proposed dam under various combinations of breach dimensions.The available downstream river cross-sections data sets were used as input in the model to generate the downstream flood profile.Dam break flow depths generated by the DAMBRK model under various combinations of structural failure are subsequently plotted on Digital Elevation Model(DEM)of the downstream of dam site to map the likely affected area.The simulation results reveals that in one particular case the flood without dam may be more intense if a rainfall of significant intensity takes place.     

Initiation conditions for debris flows generated by runoff at Chalk Cliffs, central Colorado

We have monitored initiation conditions for six debris flows between May 2004 and July 2006 in a 0.3 km² drainage basin at Chalk Cliffs; a band of hydrothermally-altered quartz monzonite in central Colorado. Debris flows were initiated by water runoff from colluvium and bedrock that entrained sediment from rills and channels with slopes ranging from about 14° to 45°. The availability of channel material is essentially unlimited because of thick channel fill and refilling following debris flows by rock fall and dry ravel processes. Rainfall exceeding I = 6.61(D)^− 0.77, where I is rainfall intensity (mm/h), and D is duration (h), was required for the initiation of debris flows in the drainage basin. The approximate minimum runoff discharge from the surface of bedrock required to initiate debris flows in the channels was 0.15 m³/s. Colluvium in the basin was unsaturated immediately prior to (antecedent) and during debris flows. Antecedent, volumetric moisture levels in colluvium at depths of 1 cm and 29 cm ranged from 4–9%, and 4–7%, respectively. During debris flows, peak moisture levels in colluvium at depths of 1 cm and 29 cm ranged from 10–20%, and 4–12%, respectively. Channel sediment at a depth of 45 cm was unsaturated before and during debris flows; antecedent moisture ranged from 20–22%, and peak moisture ranged from 24–38%. Although we have no measurements from shallow rill or channel sediment, we infer that it was unsaturated before debris flows, and saturated by surface-water runoff during debris flows.Our results allow us to make the following general statements with regard to debris flows generated by runoff in semi-arid to arid mountainous regions: 1) high antecedent moisture levels in hillslope and channel sediment are not required for the initiation of debris flows by runoff, 2) locations of entrainment of sediment by successive runoff events can vary within a basin as a function of variations in the thickness of existing channel fill and the rate of replenishment of channel fill by rock fall and dry ravel processes following debris flows, and 3) rainfall and simulated surface-water discharge thresholds can be useful in understanding and predicting debris flows generated by runoff and sediment entrainment.     

城市供水管网在地震时的连通可靠性分析

考虑地震作用效应和管道抗力的随机特性,建立了埋地管道单元的概率预测模型,评估其在地震时的震害状态。把供水管网系统简化为边权有向网络图,通过Monte Carlo随机模拟过程,近似再现管网中各管段的破坏状态,进而分别结合图论理论方法和模糊关系矩阵法,对管网进行连通可靠性分析。由于Monte Carlo模拟算法是以管网各节点与水源点处于连通状态的近似频率计算来代替精确概率分析,为获得稳定的计算结果,对所用算例进行了5000次模拟。算例分析表明,基于图论方法和模糊关系矩阵法得到的管网连通可靠性结果基本相等。