HomeWork 4 - Sampling, Resolution and Nocturnal Emissions

1) Explain, as concisely as possible (using complete english sentences), how temporal aliasing causes the apparent reversal of rotation commonly observed in motion picture video images of accelerating wagon and bicycle wheels.

What additional piece of information would you need to calculate the speed of the wagon or bicycle from the period of the wheels apparent rotation? You may use an equation to summarize your answer if you like.

2) Given optimal nadir-looking viewing geometry and illumination, approximately what spatial resolution (in meters) would be required to:

a) detect the presence or absence of a person against a high contrast background?

b) read the license plate number on a New York car license plate?

c) read 12 point typeface?

3) If data storage (rather than spatial resolution) were the primary limitation of your remote sensing system, what simple on-board operation could your instrument perform that would minimize aliasing effencts without sacrificing the accuracy of the brightness measurement?

The following questions refer to the Elvidge et. al. paper distributed in class.

4) Why didn't this study use higher resolution Landsat or Spot imagery?

Give 2 reasons and explain.

5) Why did this study only use images with less than half lunar illumination?

6) Why was it necessary to use a time series of images rather than just one cloud free image from each area?

7) Why doesn't the processed data product used in this study indicate the actual brightness of the light sources?

8) Why are there non-island light sources observed over the ocean offshore Argentina?

What about the non-island sources in the North Sea (between Scotland and Norway)??

What about the Gulf of Mexico??

9) Do you believe that the relationship between area lit and population is linear? Explain why, or why not.

Could the approach used in this study be used to estimate global population?

If so, explain briefly how the estimate would be made?

Give an order of magnitude estimate of the error (in # people) that would result from using this method.

10) Do you believe that the relationship between area lit and carbon emissions is linear?

Could the method described in this paper be used to predict carbon emissions of countries for which no other estimates exist? Explain why or why not.

11) What is the primary source of information provided by this type of study?

That is, what does this study tell us about population or energy emission that we don't already know?