The Drought In Satellite Images

Goals/Outcomes

1. Analyze effects of drought in the Dakotas over a four-year span using satellite imagery.
2. Import MODerate-resoluton Imaging Spectroradiometer (MODIS) data into ArcGIS.
3. Determine the relationship between temperature range and land cover type.
4. Determine the two warmest and two coolest land classification types for the assigned time period.

Teacher Preparation

1. Access the drought files animation
2. Perform MRT installation. Directions at:
   
   http://space.hsv.usra.edu/TRESTE/06workshop_modis.html


3. Download appropriate MODIS data for student use either from cd received at workshop or downloaded from zipped files below (a free unzip program for PCs can be downloaded at http://www.pc-shareware.com/unzipwiz.htm)
   
   
   • Base map data
   • Classification data
   • Land-cover land-use data
   • Dry period data
   • Wet period data
   
   or accessed from:
   
   http://delenn.gsfc.nasa.gov/~imswww/3.6.5/imswelcome/imswwwsites.html


4. Have ArcGIS loaded on lab computers


5. If Alternate activity II is going to be used, download appropriate PowerPoint slides
   
   http://space.hsv.usra.edu/TRESTE/papers/
   06workshop_drought_files/wet-dry_modis_images.ppt

Assumptions About Students’ Prior Knowledge/Experience

1. Students can create maps using ArcGIS
• In the interest of time, the instructor may choose to use already created maps for this portion of the lesson.
2. Students can import data to existing ArcGIS project
• If the instructor is not teaching a GIS class, s/he may wish to use the information provided on the TRESTE website.
3. Students have basic understanding of remote sensing principles
• For more background on remote sensing principles, use Maury Estes’ PowerPoint presentation found at:
  
  http://space.hsv.usra.edu/TRESTE/papers/06workshop_rs_
  fundamentals_files/ rs_fundamentals.ppt

Activities

1. Access the following animation:
   
   http://space.hsv.usra.edu/TRESTE/papers/06workshop_drought_files/ Animation_LCLU_Type5.gif


• What changes in the color patterns do you see?
• What do those changes mean?
• What does this suggest about the effects of the drought from 2001-2004 in the Dakotas?
2. Select the Missouri River watershed area in North Dakota from the MODIS Wet/Dry period data on the cd or downloaded from above addresses.
• Produce a mosaic with the MRT for one wet and one dry day
• Bring the data into ArcGIS, determine the appropriate number of classes (data ranges) for display
• Scale the digital data to compute Kelvin temperature
• Use the “temperature converter tool” to establish a Fahrenheit temperature scale
• Determine the land surface maximum, minimum temperature range spatial distribution for the two days
• Load the land cover data into ArcView for the same area
• Use the visual overlay technique to determine the relationship between temperature range and land cover type
• Determine the warmest and coolest land classification types for the wet and dry days
3. Alternate activity I
• Use the created MODIS overlay to determine the relationship between temperature range and land cover type.



4. Alternate activity II
• Use the created MODIS images for various regions to observe the relationship between land surface temperature range and land cover type during wet and dry periods.
  
  The MODISimages depict two of the products produced by the MODIS sensor orbiting Earth on the Terra Satellite, Land Surface Temperature (LST) and Land Cover Land Use (LCLU). The LST product provides 1 km pixel surface temperatures on a daily basis for both a day and night observation. The data is collected in Kelvin units.
  
  The MODIS LCLU is an annual product whose classification schemes are multitemporal classes describing land cover properties as observed during the year (12 months of input data). Slides comparing surface temperature during a wet and dry climatological period in North Dakota and LCLU for 2001 versus 2004 are provided for the Great Plains Region, the Bismarck, N.D. area, the Mission, S.D. area, the central Minnesota area, the Black Hills area, and the Chief Mountain, MT area.
  
  Dry period temperatures are from the summer of 2004 (Julian Day 207 or July 25) and the wet period temperatures from the summer of 2005 (Julian Day 209 or July 28).
  
  As noted above, the climatological data was geographically based on the state of North Dakota, which had a very dry spring in 2004, a heavy rain period from July 4-8 and very little rain from July 8-August 4. A wet period in 2005 was chosen for comparison in part because Sun angles would be similar rather than a period earlier or later in 2004.
  
  Notes
  
  Since rainfall did not occur at all or at the same rate in all the regions, LST comparisons between wet and dry will not always show the anticipated temperature differences. LCLU changes from 2001 to 2004 are likely influenced by the drought and are significant in some regions. Changes in crop type, agriculture to grassland and grassland to shrub are notable in the region.

Assessment

1. Answers to instructor questions
2. Completed ArcGIS project showing relationship between land cover and temperature range.
3. Revisit the original problem sheets for additions, deletions, corrections.