No abstract

In this paper, several considerations for developing a practical, analytical definition of droughts are discussed. These considerations include (1) selection of the nature of the water deficit to be studied (hydrological, meteorological, or agricultural); (2) selection of the averaging period used to discretize a continuous time series (months, seasons, or years); (3) selection of the truncation level used to separate droughts from the remainder of the time series (mean or median); and (4) method of regionalization or standardization. These decisions are discussed in terms of their impacts on various approaches to drought frequency analysis. In this paper, drought events are considered to be composed of duration, magnitude (average water deficiency), and severity (cumulative water deficiency). An application of the proposed drought definition procedure is presented for the case of a frequency analysis of multiyear hydrologic droughts.

A maximum likelihood estimation procedure is presented through which two aspects of the streamflow measurement errors of the calibration phase are accounted for. First, the correlated error case is considered where a first-order autoregressive scheme is presupposed for the additive errors. This proposed procedure first determines the anticipated correlation coefficient of the errors and then uses it in the objective function to estimate the best values of the model parameters. Second, the heteroscedastic error case (changing variance) is considered for which a weighting approach, using the concept of power transformation, is developed. The performances of the new procedures are tested with synthetic data for various error conditions on a two-parameter model. In comparison with the simple least squares criterion and the weighted least squares scheme of the HEC-I of the U.S. Army Corps of Engineers for the heteroschedastic case, the new procedures constantly produced better estimates. The procedures were found to be easy to implement with no convergence problem. In the absence of correlated errors, as theoretically expected, the correlated error procedure produces the exact same estimates as the simple least squares criterion. Likewise, the self-correcting ability of the heteroschedastic error procedure was effective in reducing the objective function to that of the simple least squares as data gradually became homoscedastic. Finally, the effective residual tests for detection of the above-mentioned error situations are discussed. 0 1 where ov 2 is the variance of the random components for all t > 1 and fi is a constant and, as we shall see later, its presence, from a theoretical standpoint, provides some degree of flexibility in the formation of ft matrix. Now, expressing the covariance matrix of the errors in terms of the distribution parameters of { V.r} will result in (For detail, see $orooshian [1978].) #(1 --p2)-i #p(1 --p2)-i #p2(l --p2)-, + 1 symmetric

No abstract

The relationship between the E1Nifio/Southern Oscillation (ENSO) and unimpaired streamflow over the contiguous United States is studied. The extreme phases of the Southern Oscillation have been linked to fairly persistent classes of atmospheric anomalies over the low and middle latitudes at regional and global scales. Of particular interest in this investigation is the identification of regions of land that appear to have strong and consistent ENSO-related streamflow signals. The first harmonic extracted from a 24-month ENSO composite at each station is assumed to be the ENSO-related signal appearing in streamflow anomalies. These regions were identified by the similarity in phase of the harmonic vectors. The vectorial display of these harmonics over a map of the United States provides the areal extents of ENSO influence on streamflow. Coherent and significant streamflow responses to hypothesized ENSO forcing are found in four regions of the United States: the Gulf of Mexico, the Northeast, the North Central, and the Pacific Northwest. Once an ENSO event sets in, a long-range forecasting utility may be available for these regions. The results of this analysis, which are consistent with previous studies on precipitation and temperature, demonstrate the mid-latitude hydrologic response to the tropical ENSO phenomena. most extensive region in the world of water warmer than 28øC is located in the western tropical Pacific-Indonesia region. This large warm water pool is associated with heavy rainfall and is considered a major source of atmospheric heating, which drives large-scale convective circulation patterns [Rasmusson, 1987]. Normally, the trade winds (easterlies) are driven from the eastern to the western Pacific by the differences in atmospheric pressures. These winds tend to move water westward along the equator, maintaining a buildup of warm surface water and creating, in turn, a higher sea level elevation in the western Pacific. When the Southern Paper number 93WR00744. .0043.1397/93/93WR-00744505.00 and the trade winds usually weaken (or become westerlies as the case in 1982-1983). The warm water pool extends eastward, piling up off the coast of Peru and southern Ecuador. This is the sign of the development of an E1 Nifio event [Nicholls, 1987; Vogel, 1989]. E1 Nifio events have been observed and recorded since 1726 [Cane, 1983]. They occur approximately once every 4 years; however, the time interval between successive events varies from 2 to 10 years. No two events are completely alike: They evolve according to a consistent pattern, but differ in timing, intensity, extent, and duration. Quinn et al. [1978] introduce four E1 Nifio categories: strong, moderate, weak, and very weak. However, there are noticeable differences within each category. The strongest El Nifio episode in this century, in 1982-1983, had a unique pattern in terms of the sequence of the waxming and the time of onset [Cane, 1983]. The El Nifio/Southern Oscillation (ENSO) is a warm event in the tropical Pacific Ocean and is considered a sig...