ce 397 statistics in water resources, lecture 2, 2009

26
CE 397 Statistics in Water Resources, Lecture 2, 2009 David R. Maidment Dept of Civil Engineering University of Texas at Austin 1

Upload: josephine-strong

Post on 31-Dec-2015

46 views

Category:

Documents


2 download

DESCRIPTION

CE 397 Statistics in Water Resources, Lecture 2, 2009. David R. Maidment Dept of Civil Engineering University of Texas at Austin. Key Themes. Statistics Parametric and non-parametric approach Data Visualization Distribution of data and the distribution of statistics of those data - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: CE 397 Statistics in Water Resources, Lecture 2, 2009

1

CE 397 Statistics in Water Resources, Lecture 2, 2009

David R. MaidmentDept of Civil Engineering

University of Texas at Austin

Page 2: CE 397 Statistics in Water Resources, Lecture 2, 2009

2

Key Themes

• Statistics– Parametric and non-parametric approach

• Data Visualization• Distribution of data and the distribution of statistics

of those data• Reading: Helsel and Hirsch p. 17-51 (Sections 2.1 to

2.3• Slides from Helsel and Hirsch (2002) “Techniques of

water resources investigations of the USGS, Book 4, Chapter A3.

Page 3: CE 397 Statistics in Water Resources, Lecture 2, 2009

3

Characteristics of Water Resources Data

• Lower bound of zero• Presence of “outliers”• Positive skewness• Non-normal distribution

of data• Data measured with

thresholds (e.g. detection limits)

• Seasonal and diurnal patterns

• Autocorrelation – consecutive measurements are not independent

• Dependence on other uncontrolled variables e.g. chemical concentration is related to discharge

Page 4: CE 397 Statistics in Water Resources, Lecture 2, 2009

4

Normal Distribution

From Helsel and Hirsch (2002)

Page 5: CE 397 Statistics in Water Resources, Lecture 2, 2009

5

Lognormal Distribution

From Helsel and Hirsch (2002)

Page 6: CE 397 Statistics in Water Resources, Lecture 2, 2009

6

Method of Moments

From Helsel and Hirsch (2002)

Page 7: CE 397 Statistics in Water Resources, Lecture 2, 2009

7

Statistical measures

• Location (Central Tendency)– Mean– Median– Geometric mean

• Spread (Dispersion)– Variance– Standard deviation– Interquartile range

• Skewness (Symmetry)– Coefficient of skewness

• Kurtosis (Flatness)– Coefficient of kurtosis

Page 8: CE 397 Statistics in Water Resources, Lecture 2, 2009

8From Helsel and Hirsch (2002)

y = xθ

Page 9: CE 397 Statistics in Water Resources, Lecture 2, 2009

9

Histogram

From Helsel and Hirsch (2002)

Annual Streamflow for the Licking River at Catawba, Kentucky 03253500

Page 10: CE 397 Statistics in Water Resources, Lecture 2, 2009

10

Histogram revised

From Helsel and Hirsch (2002)

Page 11: CE 397 Statistics in Water Resources, Lecture 2, 2009

11

Quantile Plot

From Helsel and Hirsch (2002)

Page 12: CE 397 Statistics in Water Resources, Lecture 2, 2009

12

Plotting positions

i = rank of the data with i = 1 is the lowestn = number of datap = cumulative probability or “quantile” of the data value (its percentile value)

Page 13: CE 397 Statistics in Water Resources, Lecture 2, 2009

13

Box and Whisker Plot(Tukey (1977) “Exploratory data analysis”)

Step = 1.5 * IQR

IQR = 75th % – 25th %

Box

Whisker

From Helsel and Hirsch (2002)

Page 14: CE 397 Statistics in Water Resources, Lecture 2, 2009

14

Variants on the Box Plot

Largest value

Largest value within“one step”

90th %

From Helsel and Hirsch (2002)

Page 15: CE 397 Statistics in Water Resources, Lecture 2, 2009

15

Normal Distribution Quantile Plot

From Helsel and Hirsch (2002)

Page 16: CE 397 Statistics in Water Resources, Lecture 2, 2009

16

Probability Plot with Normal Quantiles (Z values)

qzsqq

q

z

q

From Helsel and Hirsch (2002)

Page 17: CE 397 Statistics in Water Resources, Lecture 2, 2009

17

Probabilty Plot on Probability Paper

From Helsel and Hirsch (2002)

Page 18: CE 397 Statistics in Water Resources, Lecture 2, 2009

18

Time Series Analyst for Licking Riverhttp://tsa.usu.edu/nwisanalyst/

00060 = Daily streamflow

SiteID

Time Series Plot

Page 19: CE 397 Statistics in Water Resources, Lecture 2, 2009

19

Probability Plot

Page 20: CE 397 Statistics in Water Resources, Lecture 2, 2009

20

Histogram

Page 21: CE 397 Statistics in Water Resources, Lecture 2, 2009

21

Box and Whisker Plot

Page 22: CE 397 Statistics in Water Resources, Lecture 2, 2009

22

GetSiteInfo in HydroExcelhttp://river.sdsc.edu/wateroneflow/NWIS/DailyValues.asmx?WSDL

Page 23: CE 397 Statistics in Water Resources, Lecture 2, 2009

23

Daily Flows in HydroExcel

Page 24: CE 397 Statistics in Water Resources, Lecture 2, 2009

24

“Dimensions” of Time in HydroExcel

Page 25: CE 397 Statistics in Water Resources, Lecture 2, 2009

25

Annual Flows From HydroExcel

Annual Flows produced using Pivot Tables in Excel

Page 26: CE 397 Statistics in Water Resources, Lecture 2, 2009

26

0 10 20 30 40 50 60 70 80 900

1000

2000

3000

4000

5000

6000

Estimating the mean annual discharge

MeanMean +SeMean - Se

Number of years of Data

Mea

n Di

scha

rge

(cfs

)

Licking River at Catawba, Kentucky, 1934-2008 (75 years of data)