Seasonal to Decadal Variability in Discharge & Dissolved Solids in the Colorado River Basin: The Climate-Groundwater System

9805035 duffy本研究拟从科罗拉多河流域（CRB）独特的地貌中调查历史河流流量和溶解固体的记录。研究的目的是为低频率气候强迫（年际和年代际时间尺度）的大尺度流域响应以及自然和人为来源的长期趋势制定探测和建模策略。以盆地上部的科罗拉多高原和盆地下部的盆地和山脉地貌为代表的CRB的特征地形，定义了整个地区流体和溶质储存的自然空间和时间尺度；而灌溉和水库运行的土地利用实践则改变了自然和水化学的“信号”。需要解决的一个重要问题是“作为土地利用、地貌和水文地质条件的函数，溶解固体排放的最小状态空间维度是什么？它们与长期气候强迫有何关系？”数据库包括：从格林河和科罗拉多河上游到加利福尼亚湾的20个站点40-60年的月平均流量和河流化学；月降水-温度记录横跨整个盆地和邻近的高原和山脉；灌溉区和非灌溉区的地下水位以及数字地形和土地利用数据。从有噪声的水化学数据中提取时空信号将使用两种方法：用于空间模态分析的主成分分析和用于检测时间振荡的奇异谱分析。我们最近发表在《水资源研究》（Duffy, 1996； Duffy和Cusumano, 1998； Shun和Duffy， 1998）上的研究涉及从水化学变量的地形平均和潜在的偏微分方程中制定动力模型。该研究将为测试一种实用的低维模拟策略奠定基础，该策略试图平衡模型复杂性和可定义的建模技能水平，并评估年际和年代际气候振荡对CRB水资源的影响。

9805035DuffyThis research proposes to investigate historical river discharge and dissolved solids records from the unique landform of the Colorado River Basin (CRB). The purpose of study is to develop a detection and modeling strategy for large-scale basin response from low-frequency climate forcing (interannual and decadal time scales) and secular trends from natural and anthropogenic origin. The characteristic terrain of the CRB, represented by the Colorado plateau in the upper basin and the Basin and Range physiography in the lower basin, serves to define the natural space and time scales of fluid and solute storage across the region; while landuse practices of irrigation and reservoir operation modify the natural and hydrochemical "signal). An important question to be addressed is "what is the minimum state-space dimension of dissolved solids-discharge as a function of landuse, landform and hydrogeologic conditions, and how do they relate to long term climate forcing. The data base is: 40-60 years of monthly average streamflow and stream chemistry for 20 stations distributed from the upper Green and Colorado rivers to the Gulf of California; monthly precipitation-temperature records across the basin and adjacent plateaus and ranges; groundwater levels in irrigated and nonirrigated regions as well as digital terrain and landuse data. Two methods for extracting space-time signal from noisy hydrochemical data will be used: principal component analysis for spatial modal analysis and, singular spectrum analysis for detection of temporal oscillations. Our recent research published in Water Resources Research (Duffy, 1996; Duffy and Cusumano, 1998; Shun and Duffy, 1998) has involved formulation of dynamical models from terrain averaging of hydrochemcial variables and underlying partial differential equations. The study will form the basis for testing a practical, low-dimensional modeling strategy for the CRB which attempts to balance the degree of model complexity with a definable level of modeling skill, and to evaluate the effect of interannual and decadal climate oscillations on the water resources of the CRB.