My WebLink
|
Help
|
About
|
Sign Out
Home
Browse
Search
FLOOD01900
CWCB
>
Floodplain Documents
>
Backfile
>
1001-2000
>
FLOOD01900
Metadata
Thumbnails
Annotations
Entry Properties
Last modified
11/23/2009 10:40:54 AM
Creation date
10/4/2006 10:26:47 PM
Metadata
Fields
Template:
Floodplain Documents
County
Statewide
Community
Nationwide
Title
Techniques of Weather-Resources Investigations of the USGS Book 4, Chapter A1
Date
1/1/1968
Floodplain - Doc Type
Educational/Technical/Reference Information
There are no annotations on this page.
Document management portal powered by Laserfiche WebLink 9 © 1998-2015
Laserfiche.
All rights reserved.
/
44
PDF
Print
Pages to print
Enter page numbers and/or page ranges separated by commas. For example, 1,3,5-12.
After downloading, print the document using a PDF reader (e.g. Adobe Reader).
Show annotations
View images
View plain text
<br />SOME STATISTICAL TOOLS IN HYDROLOGY <br /> <br />Referring again to figure 2, probability is related <br />to the continuous distribution in the following <br />way: The area under the curve represents the <br />sum of all probabilities and therefore must <br />equal one. Because every item was used in <br />defining the distribution for which the total <br />area is one, then the probability that any item <br />will fall in the distribution is one and the <br />probability that an item will fall in any segment <br />of the distribution is the ratio of the area of <br />that segment to the total area. <br />The distributions just described, both <br />discrete and continuous, are called relative- <br />frequency distributions, probability distribu- <br />tions, or just distributions. However the <br />, <br />probability interpretation is valid only if the <br />data used are random. For example, the daily <br />mean flow of a stream is closely related to the <br />flows of previous days, so the distribution of <br />daily means is not one to which the proba- <br />bility inte,pretation strictly applies. It is <br />possible also to approximate a distribution <br />which merely describes the sample. For instance, <br />the distribu lion of grain sizes of a sample of <br />a streambed is measured to characterize the <br />material; there is no interest in the probability <br />of obtaining a grain in a particular size range <br />by additional sampling. Here the sample is <br />not the individual grain but an aggregate of <br />grains of various sizes. <br />Only a few standard theoretical distributions <br />are widely used. Samplinl! theory and inference <br />are based largely on the normal distribution <br />with which the reader is assumed to be familiar. <br />Other theoretical distributions will be intro- <br />duced in this and succeeding chapters as <br />appropriate. <br /> <br />. <br /> <br />. <br /> <br />Cumulative distributions <br /> <br />'. <br /> <br />Suppose we know the probability density <br />curve (probability distribution) for a variable <br />and are interested in the probability of a <br />random event being greater than some par- <br />ticular value E. This probability can be ob- <br />tained by measuring or computing the propor- <br />tion of the total area above the base value. <br />For instance, the left curve of figure 3 shows an <br />area under the curve to the right of E of 0.1, <br />that is, P=O.1. Thus the probability of a <br />random event exceeding E is 0.1. <br />275--'622 ()........67-2 <br /> <br />3 <br /> <br />E <br /> <br />i; >] <br />=f- <br />~V'i <br />"'Z <br />0"' <br />",0 <br />0. <br /> <br /> <br />w <br />o <br />:J <br />f- <br />Z <br />" <br /><< <br />" <br /> <br /> <br />E <br />MAGNITUDE <br /> <br />0.1 <br />PROBABILITY OF <br />EXCEEDANCE <br /> <br />Figure 3.~robobility density curve (left) and its cumulative <br />form (right). <br /> <br />Another form of the probability curve can <br />be prepared by cumulating the probabilities <br />from one end of the curve and plotting each <br />of these cumulated probablities against the <br />magnitude of its appropriate event. The <br />cumulation is usually done mathematically. <br />The result is the right-hand curve of figure 3. <br />Cumulative distributions are commonly plotted <br />to a probability scale such that the theoretical <br />curve is a straight line. Such a scale can be <br />devised for any two-parameter distribution. <br />Normal probability plotting paper is widely <br />known and used. Gumbel plotting paper is <br />used in many hydrologic frequency analyses. <br />(Although the Gumbel extreme-value distribu- <br />tion is a three-parameter distribution, one <br />parameter, the skew, is constant for the form <br />used and permits the construction of a scale <br />which gives a straight-line plot). Both the <br />normal and Gumbel probability plotting papers <br />are available with either arithmetic or log- <br />arithmic ordinate scales. Thus plotting papers <br />for four distributions-normal, log-normal, <br />Gumbel, and log-Gumbel-are available. <br />When the probability density curve is cumu- <br />lated from the right end, the probabilities of <br />exceeding the various magnitudes are obtained. <br />If cumulated from the left, probabilities of not <br />exceeding those magnitudes are obtained. The <br />appropriate cumulative curve, commonly called <br />a frequency curve in hydrology, depends on <br />the desired use. <br />The various theoretical cumulative distri- <br />butions used in hydrology and methods of <br />estimating their parameters from a sample of <br />data are discussed in another chapter of this <br /> <br />series. <br />
The URL can be used to link to this page
Your browser does not support the video tag.