P. 1
Trend Analysis CD

Trend Analysis CD

|Views: 29|Likes:
Published by Olusola Brossa
Trend Analysis
Trend Analysis

More info:

Categories:Types, Presentations
Published by: Olusola Brossa on Oct 04, 2013
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

07/27/2014

pdf

text

original

Trend Analysis and Presentation

Trend Monitoring
What is it and why do we do it?
Trend monitoring looks for changes in environmental parameters over time periods (E.g. last 10 years) or in space (e.g. as you move downstream)
Box Plot of Workshop Participants Concentration VS Time of Day
100
***

Percent Total Concentration

90 80 70 60 50 40 30 20 10 0
08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00

Time of Day
*** Created with StatPlus™ Software

. StatisticalGood News…Can identify hard to see trends and gives a number that is defensible and repeatable..com/elmer.asp .barbneal.Graphing or mapping data for people to see is the easiest way to communicate trends.Trend Analysis and Presentation Be vewy. vewy quiet…I’m hunting fow a twend How to find a trend VisualGood News. Bad News…No way of “measuring” that there is a trend or how big it is. Bad News…Easy to do wrong and hard to figure out. especially to a non-technical crowd. http://www.

Trend Analysis and Presentation Selecting How to Analyze and Present a Trend • • Did your data meet your data quality objectives? Trend monitoring requires strict monitoring protocols. Who is your audience and what type of analysis or presentation is appropriate for them? • • • If a statistical test is required. etc.)? . time of day. does the data satisfy all the statistic’s assumptions? What is the trend telling you? Determining the cause of the trend is more difficult than determining the trend. Have you considered all the exogenous parameters that could influence your trends (flow.

9 20 18 16 14 12 10 0 5 20.2 15.6 20.9 1500 Concentration (MPN/100mL) 1000 October Sept 0 1 August July 2 3 4 June 5 7 8 May 9 10 500 10 15 20 25 30 35 40 Month River Mile Time/area series scatter plot Bar Graphs Site Number Time series smoothed scatter plot* Time/area series box plot of statistics •From Helsel D.R.M.2 18.5 20. and R.6 20. Hirsch.coli Concentrations Avgerage 7 Day Max for July 2000 22 20. 1991 Statistical Methods in Water Resources. Chapter A3 p.6 18. 287 .Trend Analysis and Presentation Visual methods for displaying trends Average 7 day max by River Mile 24 Summer 2000 E.9 18.9 2000 2500 18. Techniques of Water-Resources Investigations of the USGS Book 4.6 21.

Trend Analysis and Presentation The Trouble with Graphs • People tend to focus on the outliers and not on more subtle changes • Gradual trends are hard to detect by eye • Seasonal variation and exogenous variables can mask trends in a parameter.gov/wlr/waterres/streams/northtrend.metrokc. • Viewers can “see what they want to see” sometimes http://dnr.htm .

Review your data for the following characteristics: • • • • • • • Normal Distribution Abrupt Changes Cycles Outliers Missing Values Censored Data Serial Correlation Select a Statistic that is appropriate for the type of data set you have.Trend Analysis and Presentation Statistical Trend Analysis Methods Picking a statistic can be tricky. .

Includes regression of parameter against time or spatial measure. or (b) conduct tests for normality like the Shapiro-Wilk W test. Generally the safer statistics to use. p.Statistics for normally distributed data. Sen slope test for measure of magnitude of slope. Parametric statistics are rarely appropriate for environmental samples without massaging data. . like river mile. and Wilcoxon-Mann-Whitney step trend analysis. 4-6). Nonparametric. but are not as readily available. or the studentized rage test (EPA.To test if an existing data set is “normally” distributed you may (a) plot a histogram of the results. the Filliben’s statistic. 2000.Statistics that are not as dependent on assumptions about data distribution. See the USGS guide by Helsel and Hirsch for descriptions of using regression. Parametric.Trend Analysis and Presentation Statistical Trend Analysis Methods Type of statistic depends on data characteristics. Test include the Kendall test for presence of consistent trend.

M. Techniques of Water-Resources Investigations of the USGS Book 4. and intercept* * From Helsel D.Trend Analysis and Presentation Things to watch out for with statistical analysis • Verify and document that all assumptions have been tested and met. non-parametric tests like the seasonal Kendall test are not readily available and can be computationally demanding. OLS regression. is usually not appropriate. Four OLS regression charts with the same R2. If you do use a statistical package. Failure to do so. and R. • Infrequent use of statistical methods often requires relearning fairly complicated procedures every year. keep excellent notes of what you do and why.245 * . • Finding no trend may only mean your data was insufficient to find the trend. Hirsch. slope. Chapter A3 p. 1991 Statistical Methods in Water Resources. may lead to misleading results • The easiest test to do.R. • More robust.

** Wilcoxon-Mann-Whitney Step Trend: Seasonal or non-seasonal test to identify changes that take place at a distinct time.R. ** Seasonal Kendall Test: Compares the relationship between points at separate time periods or seasons and determines if there is a trend. Hirsch. 266 ** Created with WQHydro Water Quality/Hydrology Graphics/Analysis System Software . 2001) * From Helsel D.M. 1991 Statistical Methods in Water Resources. and R. p. Combine with HodgesLehmann estimator to determine magnitude of step (Aroner. distribution independent methods * Sen Slope or Kendall –Theil: Compares the change in value vs time (slope) for each point on the chart and takes the median slope as the a summary statistic describing the magnitude of the trend.Trend Analysis and Presentation Statistical Methods for Determining Trends Non-Parametric. Chapter A3. recommended method for most water quality trend monitoring (Aroner. Highly robust and relatively powerful. 2001). Techniques of Water-Resources Investigations of the USGS Book 4.

267 .Trend Analysis and Presentation How to Calculate Non-Parametric Statistics Kendall Test to determine if a significant trend exists see EPA guide QA/G-9 pages 4-16 to 4-20 XY Plot of data vs time 12 10 Some Units 8 6 4 2 0 1998 Year Units 2000 5 2001 6 2002 11 2003 8 2004 10 2000 2002 Yearly Value 2004 2006 Sen Slope or Kendall-Theil line to measure the magnitude of the trend see USGS Statistical Methods for Water Resources pages 266 .

Trend Analysis and Presentation What type of trend analysis is right for you to use? What would you need for temperature trend monitoring? Is there an exogenous variable? What about changes in other water quality parameters? How could changes in monitoring schedule impact trends? Example Bob has been monitoring DO in the morning at a site for 3 years. What impact will this have on your ability to track trends? What could you do to correct the problem? DO Saturation in Willamette R. Railroad Bridge .P. Now samples are collected in the pm. He adds 4 new sites to his route and now visits his old site in the afternoon every visit. & S. at Portland S.

Techniques of Water-Resources Investigations of the USGS Book 4. 279 . * * From Helsel D.R. Chapter A3. Helsel and Hirsch (1991) give a good description of the steps needed to satisfy assumptions and data transformation tools to bring your data within the limits of the assumptions. Use regressions only if you can satisfy the data requirements.Trend Analysis and Presentation Statistical Methods for Determining Trends Regressions Regression methods draw a line as close to all the data as possible. p.M. 1991 Statistical Methods in Water Resources. and R. Hirsch.

M o n t h l y D a t a W i l l a m e t t e R i v e r a t N e w b e r g B r i d g e 9 0 9 0 8 0 8 0 7 0 7 0 OWQI. m o n t h O W Q I .month 6 0 O W Q I ..) • Seasonality.Trend Analysis and Presentation Trend Monitoring. C o m p a r i s o n o f T r e n d s : A n n u a l v s .Parameters that vary naturally in different seasons of the year require special statistics or “de-seasonalization”. 2000 WQHYDRO Environmental Data Analysis Technical App.If your data is not normally distributed. E. .Trend monitoring assumes that the same or equivalent methods and protocols are used for all the monitoring. Based on Aroner. p.R..anul 6 0 5 0 . . a n n u a l 5 0 2 / 1 2 / 1 9 9 0 6 / 2 7 / 1 9 9 1 1 1 / 8 / 1 9 9 2 3 / 2 3 / 1 9 9 4 8 / 5 / 1 9 9 5 D a t e 1.al. 1982).. you need to use a non-parametric statistical test.5 years of monthly data for WQ monotonic trend analysis (Lettenmaier et. (Statistic should be predefined in QAPP to make sure you have everything you need. . at least 2 years of monthly data before and after management change (Hirsch. 1982). 146 * Issue most applicable to statistical analysis OWQI. • Time frame & number of samples. • Type of data distribution (Is your data Normal?)*. The devil is in the details Issues to consider when doing trend monitoring1 • Consistent data quality. for step trends. .

respectively. 2000 WQHYDRO Environmental Data Analysis Technical Appendix p. Based on Aroner. • Outliers. equally spaced across the period) to minimize variance.Parameters like conductivity and pH can be greatly impacted by exogenous parameters. E. 146 * Issue most applicable to statistical analysis . like stream discharge or time of day. • Exogenous variables. Calculate the residuals (the difference between the measured value and value predicted by the regression) and then run the trend analysis on the residuals. constant number of samples per period.Missing observations may invalidate some tests by changing variance.Statistical tests designed for normal distributions are very sensitive to outliers • Serial correlation*.Trend Analysis and Presentation Issues to consider when doing trend monitoring1 • Similar variance across data set*. 1.e.Statistical methods require consistent sampling frequency (i.R. To remove these impacts develop a regression between the two variables. Changes in these exogenous variables may impact trends you measure in the parameter of interest. • Minimal missing observations*.Each result must be independent of other samples (across time and space). • Differences in detection limits*..Changing detection limits might fool the statistic into thinking concentrations are decreasing (make sure you have a way to statistically handle changing detection limits}.

WQHYDRO Environmental Data Analysis Technical Appendix. pp. . 803-813 May. Berk K. WA. Pacific Grove. 1991 Statistical Methods in Water Resources. Seattle.P. C.P. Duxbury Press.R. 1982. E. Smith. Hirsh. CA. Techniques of Trend Analysis for Monthly Water Quality Data.M. Technical Report No. Vol. Alexander. Hughes. QA00 Update Helsel D.M. 75. Routine Streams and Rivers Water Quality Trend Monitoring Review. J. 155-162.. 2000. D. Guidance for Data Quality Assessment: Practical methods for Data Analysis. Harris Hydraulics Laboratory. R.. and R..Trend Analysis and Presentation References Aroner. Water Resources Research.A.R. Data Analysis with Microsoft® Excel. 18 (1) pp.N. and R.A. EPA.M. Selection of Methods for the Detection and Estimation of Trends in Water Quality. Carey. and P. Smith 1991.. Lettenmaier.B. Slack and R.. Techniques of Water-Resources Investigations of the USGS Book 4. Chapter A3 Hirsch R. 2000 (EPA/600/R-96/084). pp. Conquest and J. 2000. 1982. Vol.W.L. 27(5). Hirsch R. EPA QA/G-9.R. 107-121 February. L. Water Resources Research.

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->