USGS B 2187: Table 2

Table 2. Results of the Mann-Whitney U-test to determine if El Niño/Southern Oscillation extremes significantly affected flow at five gaging stations.

[U₁ and U₂ are lower and upper sample values of the test statistic U. U_0.05 and U_0.95 are the lower and upper critical values of U used to accept or reject the null hypothesis H_O: median of population 1 (F₁) = median of population 2 (F₂). The alternative hypothesis H₁: median of F₁ < median of F₂ is supported by U₁ values less than U_0.05 and U₂ values greater than U_0.95. See Tamhane and Dunlop (2000) for discussion of technique]

Gaging station U₁ U₂ Decision

Mississippi River at St. Louis 89 135 Accept H_O

Ohio River at Metropolis 85 139 Accept H_O

Mississippi River at Vicksburg 95 145 Accept H_O

Rio Grande at Embudo 39 201 Reject H_O

Rio Grande at Presidio¹ 12 118 Reject H_O

¹At level of significance = 0.05, U_0.5 = 38 and U_0.95 = 92. Presidio comparison for 1930-97; all others for 1915-97.

[View a PDF file suitable for printing]

U.S. Geological Survey, U.S. Department of the Interior
This page is https://pubs.usgs.gov/bulletin/b2187/table2.html
Contact: Harry Dowsett (hdowsett@usgs.gov)
Last modified 09.07.01 (krw)


Gaging station	U₁	U₂	Decision

Mississippi River at St. Louis	89	135	Accept H_O
Ohio River at Metropolis	85	139	Accept H_O
Mississippi River at Vicksburg	95	145	Accept H_O
Rio Grande at Embudo	39	201	Reject H_O
Rio Grande at Presidio¹	12	118	Reject H_O

¹At level of significance = 0.05, U_0.5 = 38 and U_0.95 = 92. Presidio comparison for 1930-97; all others for 1915-97.