SO503 Spring 2007

Test 2

Exam makeup

There is one question worth 20 points, and 10 worth 10 points.  Answer questions worth 100 points.

This question, and only this question, is worth 20 points.  The others are worth 10 points.

This record was sampled every six minutes.  What does that say about the interpretations you can make from it?

Is this a cross correlation or an autocorrelation?  What is the difference, and how could you tell?

What would you get as the result if this were an autocorrelation?  If it is,what you do see?

What would you get as the result if this were an cross correlation?  If it is,what you do see?

What is the pattern of negative and positive r values in this graph telling you?

What do the two graphs above show on the two axes, why do we use them, and what is the difference between the two?

These are the values for one of the two graphs above.  What is the pattern for the periods, and why do you think this is the case?  What are the significant period(s), and how confident are you about the values you pick?  Why is this the case?

Period          Power

-------------------------------

204.8000        1.8594

136.5333        1.8092

102.4000        0.5610

81.9200        0.3328

68.2667        0.3063

58.5143        0.1770

51.2000        0.2017

45.5111        0.1093

40.9600        0.0895

37.2364        0.1172

34.1333        0.0707

31.5077        0.1541

29.2571        0.1257

27.3067        0.3196

25.6000        3.2841

24.0941        9.9680

22.7556        1.7449

21.5579        0.0675

20.4800        0.0476

19.5048        0.0179

18.6182        0.0111

17.8087        0.0080

17.0667        0.0077

16.3840        0.0079

15.7538        0.0108

15.1704        0.0166

14.6286        0.0167

14.1241        0.0273

13.6533        0.0609

13.2129        0.0912

12.8000        0.9426

12.4121        8.3166

12.0471        2.5828

11.7029        0.0111

11.3778        0.0581

11.0703        0.0184

10.7789        0.0161

10.5026        0.0104

10.2400        0.0083

The table below shows the results of fitting a Fourier series to a curve.  What does it tell you, and which columns are most important?  Could this table belong the time series shown above?

Component   Period   Phase    Amp   % SS  Cum % SS

1    4560.000    28.8   0.141  70.132   70.13

3    1520.000    28.6   0.057  11.443   81.58

4    1140.000     1.4   0.024   1.994   83.57

2    2280.000    59.4   0.024   1.986   85.55

5     912.000    82.8   0.024   1.981   87.54

21     217.143    31.9   0.022   1.628   89.16

6     760.000   118.4   0.015   0.739   89.90

19     240.000    42.2   0.014   0.662   90.56

23     198.261   145.3   0.012   0.502   91.07

17     268.235   127.0   0.012   0.496   91.56

24     190.000    47.1   0.011   0.389   91.95

9     506.667    71.4   0.010   0.367   92.32

20     228.000   177.7   0.010   0.351   92.67

[Graph from EOS]

Could you run a cross correlation on this data?  Why or why not, and why would you want to try?

What might have caused the 2006 hurricane season to be so mild?

How can we use GIS to predict the effects of rising sea levels?

Is Wikipedia compatible with the aims of science?

You have a byte with a decimal value of 169.  What is the value of the leftmost 4 bits?  The rightmost?

What do you think is the most important rule for using PowerPoint to communicate scientific results?

Insure that you answered questions worth 100 points.  You could leave blank either the first question, or two other questions.