Estimating the Population Size of Razor Clams Using a Model Assisted Sampling Design and Analysis Babineau, D. and Schwarz, C. Department of Mathematics.

Slides:



Advertisements
Similar presentations
Contour Lines.
Advertisements

Mean, Proportion, CLT Bootstrap
Sta220 - Statistics Mr. Smith Room 310 Class #14.
Sampling: Final and Initial Sample Size Determination
Statistics 100 Lecture Set 7. Chapters 13 and 14 in this lecture set Please read these, you are responsible for all material Will be doing chapters
An Overview of the Key Issues to be Discussed Relating to South African Sardine MARAM International Stock Assessment Workshop 1 st December 2014 Carryn.
Sampling Mathsfest Why Sample? Jan8, 2003 Air Midwest Flight 5481 from Douglas International Airport in North Carolina stalled after take off, crashed.
Written Reports Suggestions for Good Scientific Writing John E. Silvius Professor of Biology Cedarville University.
Slide 5-2 Copyright © 2008 Pearson Education, Inc. Chapter 5 Probability and Random Variables.
Regression Analysis Using Excel. Econometrics Econometrics is simply the statistical analysis of economic phenomena Here, we just summarize some of the.
1 Chapter 4 Interpolation and Approximation Lagrange Interpolation The basic interpolation problem can be posed in one of two ways: The basic interpolation.
Analysis of Variance 2-Way ANOVA MARE 250 Dr. Jason Turner.
PPA 415 – Research Methods in Public Administration Lecture 5 – Normal Curve, Sampling, and Estimation.
Excellence Justify the choice of your model by commenting on at least 3 points. Your comments could include the following: a)Relate the solution to the.
16 MULTIPLE INTEGRALS.
Chapter 11 Multiple Regression.
4-1 Statistical Inference The field of statistical inference consists of those methods used to make decisions or draw conclusions about a population.
Statistics for Managers Using Microsoft Excel, 4e © 2004 Prentice-Hall, Inc. Chap 6-1 Chapter 6 The Normal Distribution and Other Continuous Distributions.
Chapter 7 Probability and Samples: The Distribution of Sample Means
Chapter 5 DESCRIBING DATA WITH Z-SCORES AND THE NORMAL CURVE.
IB Chemistry Chapter 11, Measurement & Data Processing Mr. Pruett
Copyright © Cengage Learning. All rights reserved. 10 Inferences Involving Two Populations.
Interactive Graphics Lecture 9: Slide 1 Interactive Graphics Lecture 9: Introduction to Spline Curves.
Chapter 7: Sampling Distributions
STA Lecture 161 STA 291 Lecture 16 Normal distributions: ( mean and SD ) use table or web page. The sampling distribution of and are both (approximately)
© 2008 McGraw-Hill Higher Education The Statistical Imagination Chapter 7. Using Probability Theory to Produce Sampling Distributions.
16-1 Copyright  2010 McGraw-Hill Australia Pty Ltd PowerPoint slides to accompany Croucher, Introductory Mathematics and Statistics, 5e Chapter 16 The.
Probabilistic and Statistical Techniques 1 Lecture 24 Eng. Ismail Zakaria El Daour 2010.
Copyright © 2012 by Nelson Education Limited. Chapter 7 Hypothesis Testing I: The One-Sample Case 7-1.
QBM117 Business Statistics Descriptive Statistics Numerical Descriptive Measures.
COURSE: JUST 3900 INTRODUCTORY STATISTICS FOR CRIMINAL JUSTICE Instructor: Dr. John J. Kerbs, Associate Professor Joint Ph.D. in Social Work and Sociology.
The Examination of Residuals. Examination of Residuals The fitting of models to data is done using an iterative approach. The first step is to fit a simple.
Measures of central tendency are statistics that express the most typical or average scores in a distribution These measures are: The Mode The Median.
Chapter 4 Linear Regression 1. Introduction Managerial decisions are often based on the relationship between two or more variables. For example, after.
PCB 3043L - General Ecology Data Analysis. OUTLINE Organizing an ecological study Basic sampling terminology Statistical analysis of data –Why use statistics?
Time series Decomposition Farideh Dehkordi-Vakil.
+ Chapter 12: More About Regression Section 12.1 Inference for Linear Regression.
Time series Model assessment. Tourist arrivals to NZ Period is quarterly.
Chapter 7 Probability and Samples: The Distribution of Sample Means
Biostatistics Unit 5 – Samples. Sampling distributions Sampling distributions are important in the understanding of statistical inference. Probability.
11/23/2015Slide 1 Using a combination of tables and plots from SPSS plus spreadsheets from Excel, we will show the linkage between correlation and linear.
Simulated data sets Extracted from:. The data sets shared a common time period of 30 years and age range from 0 to 16 years. The data were provided to.

Sampling Design and Analysis MTH 494 Lecture-22 Ossam Chohan Assistant Professor CIIT Abbottabad.
© 2008 McGraw-Hill Higher Education The Statistical Imagination Chapter 8. Parameter Estimation Using Confidence Intervals.
Basic Hydraulics: Channels Analysis and design – I
CHAPTER 27: One-Way Analysis of Variance: Comparing Several Means
Chapter 2: Frequency Distributions. Frequency Distributions After collecting data, the first task for a researcher is to organize and simplify the data.
Sampling Design and Analysis MTH 494 Lecture-21 Ossam Chohan Assistant Professor CIIT Abbottabad.
Estimation by Intervals Confidence Interval. Suppose we wanted to estimate the proportion of blue candies in a VERY large bowl. We could take a sample.
Lecture 8: Measurement Errors 1. Objectives List some sources of measurement errors. Classify measurement errors into systematic and random errors. Study.
Calculating ‘g’ practical
1 Chapter 4 Interpolation and Approximation Lagrange Interpolation The basic interpolation problem can be posed in one of two ways: The basic interpolation.
Lecture 7: Bivariate Statistics. 2 Properties of Standard Deviation Variance is just the square of the S.D. If a constant is added to all scores, it has.
Copyright © Cengage Learning. All rights reserved. 8 9 Correlation and Regression.
Dr.N.K.Tyagi, SAMPLE SIZE The average in the form of estimate ‘p’ or mean should be of known along with its precision and tolerable error,
Comparing survival estimates from a radio-tag mark-recapture study. L. Cowen and C.J. Schwarz Department of Statistics and Actuarial Sciences, Simon Fraser.
The Statistical Imagination Chapter 7. Using Probability Theory to Produce Sampling Distributions.
Copyright © Cengage Learning. All rights reserved. 8 4 Correlation and Regression.
Copyright © Cengage Learning. All rights reserved.
Chapter 4 Basic Estimation Techniques
How do you assign an error to a measurement?
Some examples Noughts and crosses exercise
Sardine Two-Stock Hypothesis: Results at the Posterior Mode
AGE DETERMINATION – INTEGRATED METHOD
Descriptive Analysis and Presentation of Bivariate Data
Methods of Age Determination using Length Frequency Method
Geology Geomath Chapter 7 - Statistics tom.h.wilson
Sampling: How to Select a Few to Represent the Many
These probabilities are the probabilities that individual values in a sample will fall in a 50 gram range, and thus represent the integral of individual.
Presentation transcript:

Estimating the Population Size of Razor Clams Using a Model Assisted Sampling Design and Analysis Babineau, D. and Schwarz, C. Department of Mathematics and Statistics, Simon Fraser University, Burnaby, BC V5A 1S6 Abstract For several decades, beaches near Masset, British Columbia, seen in Figure 1, have been used for commercial and non commercial harvesting of razor clams (Siliqua patula Dixon), seen in Figure 2. In the early 1990’s, the health of the stock seemed to be failing because commercial fishery landings were low and there were high proportions of undersized razor clams in commercial catches. Thus, a quantitative assessment of clam stocks needed to be carried out. This was done using the following model assisted sampling design and analysis. Study Design The beaches located near Masset, British Columbia are one of eight major concentrations of razor clams existing on the Pacific coast. Using a three stage sampling design, the three beaches that were surveyed between 1994 and 1996 were North Beach, South 1 Beach and South 2 Beach. A map of the area surveyed is given in Figure 3. Figure 4 illustrates the following sampling procedure for a specific beach. Stage 1: To begin sampling of a beach, transects (lines laid perpendicular to chart datum (lowest possible tide)) were allocated by spacing them equally along the beach. Transects where then randomly sampled. Stage 2: Once a transect was randomly sampled, distances from chart datum along the transect were systematically sampled, with sampling beginning where the surf line was located at the time of arrival. For each distance sampled, the elevation above chart datum was also recorded. Stage 3: Once distances were sampled, plots of sand called quadrats were located at each sampled distance by drawing a line parallel to chart datum that was within 7 m of the transect line. Quadrats were then randomly sampled along the line using a galvanized steel circular sampling ring with an area of 0.5 m 2. For each sampled quadrat, characteristics such as the number of razor clams in the sample and the length of each razor clam were determined. To illustrate the method about to be developed, the data collected from the 1995 survey of South 1 Beach for razor clams larger than 20 mm is used and is given in Table 1. Estimation of the Number of Razor Clams Along a Transect To begin the analysis, an estimate of the number of razor clams along each sampled transect is found. This is done using biological information that is extracted from the data collected. Due to environmental factors and the physiological requirements of the organism,razor clams are more likely to be found closer to chart datum. This is easily seen in Figure 5 by plotting the number of razor clams found in each sampled quadrat against the distance from chart datum where the quadrat is located. The following analysis determines the number of razor clams along the transect located at 0.8 km along South 1 Beach in 1995 by modeling the above relationship between the number of razor clams sampled at each quadrat and distance from chart datum. This is done using two methods. Method 1: Straight Line Interpolation To model the relationship that exists between the number of razor clams sampled at each quadrat and the distance from chart datum using a straight line interpolation, the following steps are carried out. Step 1:Determine the mean number of razor clams at each sampled distance. Step 2:Interpolate between the points representing the mean number of razor clams at each sampled distance to produce a piecewise smooth curve. To estimate the number of razor clams along the above transect, the area under the above curve must be found and is determined in the following way. Step 1:Separate the piecewise smooth curve into distinct trapezoids. Step 2:Determine the area of each trapezoid under the curve. For example, the area of the hatched section shown in Figure 9 represents razor clams. Step 3:Once the area of all trapezoids under the curve is determined, the areas are summed together. This determines the estimated number of razor clams along thetransect using a straight line interpolation. For the transect located at 0.8 km along South 1 Beach in 1995, it is estimated that there are approximately 585 razor clams with lengths greater than 20 mm. Using standard statistical methodology, the precision of this estimate is found to be within 183 razor clams, 19 times out of 20. In general, to find the number of razor clams along any transect using a straight line interpolation, the estimate is given by Method 2: Cubic Smoothing Spline The non linear relationship between the number of razor clams at each distance and the distance from chart datum is also modeled using a cubic smoothing spline. By fixing the smoothing parameter, a smooth curve is found and is shown in Figure 10. To estimate the area under the above curve, Riemann sums are used. This determines the estimated number of razor clams along the transect using a cubic smoothing spline. For the transect located at 0.8 kmalong South 1 Beach in 1995, it is estimated that there are approximately 583 razor clams greater than 20 mm in length. Using bootstrapping techniques, the precision of this estimate is found to be within 138 razor clams, 19 times out of 20. A comparison of the methods used to determine the number of razor clams along a transect shows that both the estimates and standard errors are quite similar for both methods. However, it should be noted that while a cubic smoothing spline offers a more realistic approach to modeling the relationship between the number of razor clams sampled at a distance and distance from chart datum, the method using a straight line interpolation requires less computational time. The choice that must be made between either method is than a trade off between these two characteristics. Upon applying both a straight line interpolation and a cubic smoothing spline to each of the transects sampled along South 1 Beach in 1995, estimates and associated standard errors of the number of razor clams along each transect are given in Table 2. Estimation of the Number of Razor Clams On a Beach To estimate the number of razor clams greater than 20 mm in length on South 1 Beach in 1995, two different estimators are used. Ratio Estimator This estimator stems from the fact that longer transects contain higher numbers of razor clams than shorter transects. This is shown in Figure 10. Because a straight line fit to the above plot would pass through the origin, the following estimator is suggested. Justification of the above expression begins by noting that  T i /  L i is the estimated number of razor clams per 0.5 m 2. Once this expression is multiplied by 2, an estimate of the number of razor clams per m 2 is determined. It is then multiplied by A, the area of the beach, to determine the total number of razor clams on the beach. Inflation Estimator Because the first stage of sampling involves a simple random sample of transects, an inflation estimator like the one given below is used. Justification of the above expression begins by noting that  T i / n is an estimate of the total number of razor clams along any transect. Once this expression is multiplied by N, the total number of transects on a beach, the total number of razor clams on the beach is determined. The above methods are applied to the data collected from the 1995 survey of South 1 Beach for razor clams with lengths greater than 20 mm. The results are given in Table 3. Comparison of the above estimates indicates that ratio and inflation estimates are similar. However, the estimated standard error for the ratio estimator is approximately half that of the estimated standard error for the inflation estimator. This is expected because the beach surveyed in not rectangular in shape and the inflation estimator is more suitable for rectangular shaped beaches. However, it should be noted that if the relationship shown in Figure 10 is not linear or a straight line fit does not pass through the origin, the ratio estimator will perform poorly. In cases such as these, the inflation estimator is more suitable. For a more detailed discussion of this and other analyses of razor clam data, please refer to the following publications. Babineau, D. (2000). Estimating the Population Size of Razor Clams Using a Model Assisted Sampling Design and Analysis. Master’s Project. Burnaby: Simon Fraser University. Bourne, N. (1969). Population Studies on the Razor Clam at Masset, British Columbia. Fish. Res. Board Can. Tech. Rep. No Szarzi, N.J. (1991). Distribution and abundance of the Pacific razor clam (Siliqua patula Dixon), on theEastside Cook Inlet beaches, Alaska. Master’s thesis. Fairbanks: University of Alaska. The Statistics Group at Simon Fraser University specializes in Applied Statistics. We offer both undergraduate and graduate programs with co-op options for both streams. For further information, visit us at our website, or call us at (604) Figure 3: Map of beaches near Masset, BC. Figure 4: Sampled beach. Figure 2: Pacific razor clam Figure 1:Map of south western coast of Canada TRUE OR... FALSE?? Table 1:Raw data from 1995 survey of South 1 Beach for razor clams greater than 20 mm in length. Figure 5:Plots of the number of razor clams sampled per quadrat versus the distance from chart datum using data collected from the 1995 survey of South 1 Beach for razor clams greater than 20 mm in length. Figure 6:Plot of the mean number of razor clams at each sampled distance versus distance from chart datum. Figure 7:Curve produced using a straight line interpolation. Figure 8: Trapezoidal sections produced. Figure 9: Hatched section displays the area of the first trapezoid. Figure 10: Curve produced a cubic smoothing spline. Table 2:Estimates of the number of razor clams along each transect sampled on South 1 Beach in 1995 for razor clams greater than 20 mm in length. Table 3:Estimates of the number of razor clams greater than 20 mm in length on South 1 Beach in Figure 10: Plot of the estimated number of razor clams along a sampled transect versus the length of the transect.