Lab exercise 4, due right before the next class
GEO5083: Remote Sensing Image Processing and Analysis, UTSA

Student name: ______________

Radiometric correction of ETM+ and comparison between ETM+ and MODIS

Purpose

            This lab is to continue familiarize you in working with ETM+ and MODIS data, as well as in using IDL. In the last lab, you compared the temperatures of the two. In this lab, you will compare the reflectivity and NDVI. You will do an ETM+ radiometric correction, and transfer to reflectivity. the ETM+ reflectivity will then be compared with the MODIS reflectivity product.

            We will use the Dark Subtraction to do a relative radiometric correction, but you are encouraging to explore the FLASSH and Empirical Line methods for an absolute radiometric correction if you will have enough time or you can do this in your term project. This lab also supposes your area is flat. However, if you would like, you can use a DEM to do a topographic correction, again, if you have time or you may do so in your term project. I would be happy to help you out in any way.

Step 1. Preparation

        You will need to download the 6 reflectivity bands of the ETM+ image that you used for Lab 2 from the TexasView website at http://www.crgsc.org/Data/Remote.aspx, or from the USGS Global Visualization Viewer at http://glovis.usgs.gov. You need to download the files of *.H1, *.I1, *.I2, *.I3, *.I4, *.I5, *.I7, which are band 1, 2, 3, 4, 5, and 7, respectively. (notice: the *.H3 is for the *.I8 image of 15 meter resolution). The *.H1 is the headfile. When you open the *.H1 in ENVI, you should be able to open the stacked 6 images.

     Following the same steps in Lab 3 to download the MODIS/TERRA SURFACE REFLECTANCE DAILY L2G GLOBAL 1KM and 500M SIN GRID V005 for exact the same day as for the ETM+ image. This is the MODIS reflectivity product (MOD09A1): daily and 500 m resolution. Then you use HEGTool and ENVI to reproject and stack them as you did in Lab3.

MODIS band1-7 and spectral ranges, with corresponding ETM+ bands in the middle column

ETM+ band1-7 and spectral ranges, with corresponding MODIS bands in the middle column

          In your report, please do a short discussion about the spatial, spectral, temporal, and radiometric resolutions of these two types of reflectivity bands.  Use ENVI statistical tool to find the range of DN, which indicates you the radiometric resolution.

 Step 2. Do a simple relative radiometric correction to ETM+ using ENVI

        Get the minimum DNs of each bands of your stacked ETM+ image by using a Mask to mask the unwanted outside area. Those minimum values will be used as inputs for the Dark Subtraction for radiometric correction (see figure below).

Step 3. Relative atmospheric correction and reflectivity calculation

        Use the similar IDL code as you used in Lab 3, change the equations based on the gains and offsets you can find from the *.H1 of the ETM+ imagery or from lab4 in the remote sensing class http://www.utsa.edu/lrsg/Teaching/EES5053_ES4093/Lab4.doc. The code should include two steps (DN to radiance, radiance to reflectivity). Please submit the code as part of the Lab submission.

        To convert the MODIS DN or scaled integers to reflectivity, a scale factor of 0.0001 will be applied. you can use either band math or IDL code to do this work.

Step 4. Compare the reflectivity from both ETM+ and MODIS

To compare the reflectivity of band by band, I suggest you to upscale the 30 meter ETM+ image to 500 meter as the MODIS resolution. You can stack them together and resize them to cover the exact same area (no zero value around), then you can easily use the statistic tool to find their correlation coefficient of any ETM+/MODIS pair.

Step 5. Compare the NDVI from both ETM+ and MODIS

First calculate the NDVI from both ETM+ and MODIS reflectivity (tell me which bands you should use to do so). then do the similar comparison as step 4.

 Write a report about the study area, the data, the method used, and your analyzing results, including the discussion mentioned in step 1. also please submit the IDL code as part of the Lab submission.