Discussion Questions (Week 6)

 

Pax-Lenney et al. 2001. Forest mapping with a generalized classifier and Landsat TM data.

 

1.      What was the generalization that is mentioned throughout the article?

2.      What type of satellite image was used for the study? Is there a commonly accepted type of atmospheric correction for this type of satellite image on a regional scale?

3.      What were the various types of atmospheric corrections used for the study?  What ways were they used for correcting the data?

4.      Why was a relative transfer method used and why specifically the 6S code?

5.      What is the test used to verify the atmospheric corrections and generalization levels?

  1. Answering one of the thesis questions put forth by the authors, are atmospheric corrections necessary for landcover mapping and which method produces the highest classification?

 

Kaufman and Remer 1994. Detection of forests using mid-IR reflectiance – An application for aerosol studies.

 

 

  1. Why are the authors of this paper interested in the differences and similarities between the mid-IR, red, and near-IR wavelengths?

 

  1. Why are the authors trying to detect dense, dark, vegetation (DDV) and what is the significance of this land cover type to detection of atmospheric phenomena?

 

  1. Why must correcting for emissivity of mid-IR be a part of the methodology for this study?

 

  1. What is NDVI and how does it work?

 

  1. What data is used for this study?

 

  1. What are the differences between the mid-IR, red, and NDVI in detecting seasonal variation of forest cover?

 

 

Simpson and Stitt 1998. A procedure for the detection and removal of cloud shadow from AVHRR data over land.

 

1)      What kinds of conditions (atmospherical, latitude, terrain type) is this process best suited for?

2)       How do the authors determine the height and vertical extent of clouds?

3)      What channel is used for cloud shadow detection, and why?

4)      What are the channel 2 albedo differences between areas covered by cloud, cloud shadow, and clear land?

5)      Can the authors’ approach be used to find areas of cloud shadow without first detecting the cloud that cast it?

6)      How does a cloud’s vertical extent affect its shadow and how does the authors’ method overcome the problem?