This is a testing site only. See the live Public Lab site here »

Public Lab Wiki documentation



Cape Cod Bay Watch Landsat Tutorial notes 7/14/14

This is a revision from July 26, 2014 15:39. View all revisions
1 | 12 | | #10980

These are the notes from a tutorial that Ned Horning gave to the members of Cape Cod Bay Watch and the Jones River Watershed Association, (Pine duBois, Alex Mansfield, Karen Vale, and Adam Augello) and Don Blaire from Public Lab, on how to download and edit GeoTIFF Landsat images on QGIS. Cape Cod Bay Watch and the JRWA want to use this program for thermal mapping of Cape Cod Bay, specifically to measure the thermal plume generated from the Pilgrim Nuclear Power Station in Plymouth.

Website: glovis.usgs.gov

-must have Java 7 or older to run program -won’t work on Google Chrome if you have a Mac

Steps below use Pilgrim Nuclear Power Station in Plymouth, MA as an example:

• Choose the image of Cape Cod Bay by typing in 11/31 next to “path/row”

• Click on “Resolution” then “240 m”

• Can click on “Collection” then “Landsat Archives” then “L7 SLC off (2003>)” (this has images taken every eight days. However, there are data gaps within this information so that 22% of the dates are missing)

• or Can click on “Collection” then “Landsat Archives” then “Landsat 8 OLI” (this only has images every sixteen days. However, the dates offset Landsat 7’s so that you can supplement the days of both to get 4 data points in a twenty-four day period [not taking into account missing data points]) Data goes back to 1984

• Can change the date underneath the “Scene Information” box. Click “next scene” to see next available image in the series. Only want to download images without cloud cover.

• In the “Scene Information” box you can see the date, the % of the image that is covered by clouds (“CC”), and the Scene ID

• If you scroll your cursor over the image, below the image it will tell you the corresponding latitude and longitude

• To download a particular image, click the “Add” button on the bottom left hand side of the screen. The scene ID will show up in a box above it. You can go through more images and add them to this box, until you are ready to “Send to Cart.” This is where you will have to sign in. Once you do so, click the green button on the far right of the screen, next to the red X. Then, for Landsat 8, click on GeoTIFF Data Product.

• To open images, you will need GIS, QGIS (this was the recommended program), photoshop, or gimp (open source); Extract like a zip file. Then click on either B10 or B11

• TXT file= metadata file= all of the data about that image

• 1 pixel= 30 m squared

• darker images are colder

• lighter images are warmer

• zoom in

• make sure you don’t confound cloud cover and water temperature

• tools to enhance the image:

➢ right click image name

➢ properties

➢ style

➢ actual pixel values

➢ mean plus or minus standard deviation

➢ load (changes max and min)

➢ apply

• fine detail of plume:

➢ zoom in

➢ properties (recalculate statistics)

➢ click pixel value for land and then water to check what colors correspond to in terms of hot and cold

• Linear calibration: (pixel value multiplied by band value)+ “add” value times 0.1 = top of the atmosphere radients units (similar to Watts)

• K1 constant band 10: plug in “top of the atmosphere radients” to get degrees Kelvin.

• can assign brightness temperature (water temp. and surface temp. not adjusted to atmosphere) to pixels. To create a formula for doing this, double click.

• “As far as calculating brightness temperature you can use these two equations:

Top of the atmosphere radiance: TOA Radiance = DN * RADIANCE_MULT_BAND_10 + RADIANCE_ADD_BAND_10

DN is the pixel values in the image.

Here is the formula that I used in QGIS Raster Calculator: "LC80120312014173LGN00_B10@1" * 0.0003342 + 0.1

You would replace "LC80120312014173LGN00_B10@1" with your image.

To calculate brightness temperature use this equation:

T = K2_CONSTANT_BAND_10 / (ln((K1_CONSTANT_BAND_10 / TOA Radiance) + 1))

If you want to combine both of these so you only have to run it once the equation is: T = K2_CONSTANT_BAND_10 / (ln((K1_CONSTANT_BAND_10 / (DN * RADIANCE_MULT_BAND_10 + RADIANCE_ADD_BAND_10)) + 1))

Here is the formula that should work in the QGIS Raster Calculator but it doesn't since it appears that the natural logarithm, "ln", isn't supported yet in this new version of the raster calculator. The user guide mentions that it will be supported in a future release.

• 1321.08 / ln(774.89 / ("LC80120312014173LGN00_B10@1" * 0.0003342 + 0.1) + 1)

I'm not sure of the best way for you to calculate brightness temperature since this isn't working at this time. The problem is that there are several options and they all require using another software package. One option that might be useful would be to do the processing using an R script. I could write a script that would require you to enter the image directory path and name and the script would clip out your area of interest and create a GeoTiff brightness temperature image. I think the script would be able to read the metadata text file so you wouldn't have to enter those TOA radiance and K coefficients by hand. You could run the script outside of QGIS and I can probably get it to work inside of QGIS if that's easier for you.” –Ned Horning

• could create legend, but would have to specify what brightness temperature is or that the difference between the plume and the bay is in relative measurements • you could also draw a transect (draw a line, graph above it with relative temp. above the plume)