H24E-01
Climate and Management of the Colorado River: What Does the SSD Study Tell Us About Scenario Analysis?
The "Severe Sustained Drought in the Southwestern US" (SSD) research project conducted between 1989-1994 focused on the likely size and distribution of water shortages in the Colorado River Basin should the most severe, long-term drought found in the tree-ring record reoccur in the modern era. The research approach, generally, was to (a) utilize tree-ring studies to reconstruct prehistoric flows, (b) select the drought from this record that was most likely to catastrophically stress the system, (c) estimate future water demand in the basin, (d) utilize institutional studies and water system modeling to identify the timing, location and magnitude of likely water shortages should this drought reoccur, (e) assess impacts and damages associated with these shortages, and (f) investigate a small set of coping strategies. Each step in this process raised a variety of technical challenges and drew upon the talents of approximately 40 researchers including engineers/hydrologists, attorneys, economists, environmental scientists, sociologists, and policy specialists. The goal was to stimulate thinking and reform prior to a drought crisis; however, the study findings were largely ignored until the recent drought crisis, and then once discovered, some of the key conclusions of the study regarding substantive issues of water shortage vulnerability and distribution were found to be inconsistent with (and thus potentially irrelevant to solving) the current drought. Understanding why some key results of the research scenario differ significantly from the current crisis is important in informing future scenario analyses. Specifically, this case study highlights the importance and difficulty of (a) effectively linking scientific modeling efforts with social analyses, and (b) linking scenario analyses to real-world policy-making settings. In this basin, improving the understanding of drought is likely less about improved scholarship in climate and hydrology, and more about understanding the role of water demand, risk allocation, legal and political uncertainties, distorted economic incentives, and the lack of adequate forums, tools and incentives for dispute resolution. Until the institutional problems are better understood and integrated into scientific assessments, efforts such as the SSD study are unlikely to achieve their full theoretical benefits.
H24E-02
The Value of Scenario Development in Environmental and Socio-economic Policy Applications
Increasing scarcity, growing demand, and a burdened water supply have generated concerns about the sustainability of the southwest's regional water infrastructure. This necessitates the adoption of improved water management practices and policies better suited to contemporary water resource dilemmas. Scenarios introduce an innovative aspect to strategic long-term planning that is currently absent from current decision- making and resource management activities. For the purpose of assessing future water resources management and sustainability needs within the region, the formal approach to scenario development adopted by scientists and researchers at the University of Arizona's SAHRA (Sustainability of semi-Arid Hydrology and Riparian Areas) center is adopted. Through workshops and meetings with regional and state stakeholders, several dominant themes of interest for water resources management emerged. A historical analysis of several key variables associated with these major themes provided insight on the future uncertainty in projections and assumptions adopted in early examples of future planning in the southwest. Analysis of these key variables indicates that historical assumptions and projections in the dimensions of the environment, climate, and socio-economics lacked the dynamic planning foresight that tools such as scenarios can provide.
H24E-03
Paleoflood Data, Extreme Floods and Frequency: Data and Models for Dam Safety Risk Scenarios
Extreme floods and probability estimates are crucial components in dam safety risk analysis and scenarios for water-resources decision making. The field-based collection of paleoflood data provides needed information on the magnitude and probability of extreme floods at locations of interest in a watershed or region. The stratigraphic record present along streams in the form of terrace and floodplain deposits represent direct indicators of the magnitude of large floods on a river, and may provide 10 to 100 times longer records than conventional stream gaging records of large floods. Paleoflood data is combined with gage and historical streamflow estimates to gain insights to flood frequency scaling, model extrapolations and uncertainty, and provide input scenarios to risk analysis event trees. We illustrate current data collection and flood frequency modeling approaches via case studies in the western United States, including the American River in California and the Arkansas River in Colorado. These studies demonstrate the integration of applied field geology, hydraulics, and surface-water hydrology. Results from these studies illustrate the gains in information content on extreme floods, provide data- based means to separate flood generation processes, guide flood frequency model extrapolations, and reduce uncertainties. These data and scenarios strongly influence water resources management decisions.
H24E-04
Strategy of Water Resources Planning Under Risk
In water resources systems analysis, risk, caused by uncertainty, is an important issue to consider, whereas definition of risk and its measure is controversial (many definitions are available in different research fields). The problem of computing the degree of risk in water resources planning is very difficult, and has received more and more attentions from more hydrologists. This study discussed the necessity of risk analysis on decision-making associated with problems of managing regional water quantity. A new concept of risk function for regional water resource planning was introduced, and a theory of risk analysis of water resource systems was developed and implemented numerically. The developed methodology is general and can be used to tackle many kinds of decision-making problems. When loss (or benefit) volumes of an action set and probabilities of nature state of decision environments are given, non-inferior planning strategy or strategies can be derived by ordering the size of risk degrees calculated by the proposed risk function. This method was illustrated in a case study at the Huanghuaihai basin, China, one of the major food-producing areas in north China. In the last several decades, problems of water shortage and pollution are severe, and extreme weather conditions frequently occur. How to reasonably allocate the limited fresh water in the future under uncertainty is an urgent task. In this research, alternative strategies of water resource planning were investigated and risk of the strategies was assessed to facilitate the decision-making of Chinese government. The developed methodology selected the optimum choice of water resources planning strategies to avoid the risk of water shortage. This research has practicably provided support of decision-making of the Chinese central and local governments and organizations in their regional and national planning.
H24E-05
Climate Scenarios for the Western US: Incorporating ENSO Variability Into Downscaled Temperature and Precipitation Projections from Coupled Climate Models
In this study we incorporate the effects of ENSO on the future climate projections of the Western United States. To determine future climate scenarios we use the coupled climate models participating in the IPCC fourth assessment report, as they are currently the best quantitative climate projections available. Unfortunately, the coarse spatial resolution of the climate models makes their raw output unsuitable for decision support systems at the level of local water managers. To overcome this difficulty we downscale the projected values of precipitation and temperature from coupled climate models using observed mean high-resolution values (as in traditional downscaling techniques) but in addition, we incorporate the second moments associated to ENSO spatial and temporal variability. ENSO projections are obtained from the two coupled climate models that best represent the climate in the Southwest. Our work provides improved downscaled calculations as it considers the spatio- temporal variability of precipitation and temperature associated with ENSO while preserving the mean projected changes from the coupled climate models.
H24E-06
Evaluating Recent 20th Century Changes in Cool Season Precipitation Variability in the Western U.S. in the Context of Paleoclimatic Reconstructions
A number of previous studies have identified substantial changes in the variability of cool season precipitation and annual streamflow in the western U.S. Since about 1975, systematic increases in the coefficient of variation (CV), temporal persistence (i.e. lag 1 autocorrelation), and inter-regional correlation of cool season precipitation have been observed across the West. A key question that emerges in response to these observations is how unusual the recent changes in precipitation variability are in the context of normal patterns of variability in longer records. Here we examine long paleoreconstructions of annual river flow (1858-1977) in three large river basins in the Pacific Northwest, California, and the Colorado River basin as a proxy for cool season precipitation in the three regions, and compare them to more recent 20th century records of cool season precipitation from gridded data sets (1916-2003). We begin by creating metrics that identify periods in the time series when the CV, temporal persistence and inter-regional correlation are simultaneously high or low in all three regions. Analysis over the long paleoclimatic records shows that although the CV has varied considerably through time, prior to the mid 1970s temporal persistence and inter-regional correlation have been strongly out of phase (i.e. when lag 1 autocorrelation is high in all three regions, inter-regional correlation tends to be low and vice versa). By contrast, for post 1975 records the CV, temporal persistence, and inter-regional correlation are all simultaneously at very high positive values, a condition unprecedented in the 150 year record. We conclude that recent changes in precipitation variability are unusual in the context of a 150 year record, and in particular that a stable pattern of anticorrelation of temporal persistence and inter-regional correlation seems to have been disrupted since about 1975.
H24E-07
Simulation of Daily Rainfall Scenarios for S. Florida That Reflect Interannual and Multidecadal Climate Cycles
Concerns with the potential effects of anthropogenic climate change have led to a closer examination of how climate varies in the long run, and how such variations may impact rainfall variations at daily to seasonal time scales. For S. Florida in particular, the influences of the El Nino Southern Oscillation, the North Atlantic Oscillation, and the Atlantic Multidecadal Oscillation have been identified with aggregate annual or seasonal rainfall variations. Since the combined effect of these variations is manifest as persistent multi-year variations in rainfall, the question of modeling these variations at the time and space scales relevant for use with the daily time step driven hydrologic models in use by the SFWMD has arisen. This is the problem addressed in this study. A newly developed methodology called Wavelet Autoregressive Modeling (WARM) is used in the first step after suitable climate proxies for regional rainfall are identified. These proxies typically have data available for a century to 4 centuries so that long term quasi-periodic climate modes of interest can be identified more reliably. Correlation analyses with seasonal rainfall in the region are used to identify the specific proxies considered as candidates for subsequent conditioning of daily rainfall attributes using a Nonhomogeneous Hidden Markov Model (NHMM).
H24E-08
A smoothed statistical regionalization approach for modelling monthly precipitation in complex terrain
Using rain-gauge station records for the statistical characterization and simulation of spatio-temporal precipitation fields involves many issues and simplifying assumptions. One major issue is related to dealing with uncertainty at-site sample statistical inference, because of the limited length of records. Regional frequency analysis uses substituting space for time in order to reduce uncertainty by assuming equal shapes of the precipitation statistical distributions in a region. However, this assumption limits the area of the analyzed region where this assumption is valid. The extension is dependent on terrain complexity. This work presents a new approach for the statistical regionalization of a large precipitation fields, replacing the constant shape assumption by using a smooth spatial variation. The approach accounts for every uncertainty on site information, using an L-moment method for inference analysis. Additionally, the orographic effect is introduced in the regionalization, which substantially improves the interpolation performance and estimation of areal precipitation. The approach is used for modelling the monthly precipitation field in the Júcar River Basin Authority Demarcation (Spain), incorporating its stochastic structure, and spatial dependency from a geostatistical analysis. Issues related to the estimation of regional precipitation, and mean areal precipitation are also discussed.