google-earth-engineremote sensingsentinel-1sentinel-2
In LANDSAT
var mndwi = ee.Image("LANDSAT/LC08/C01/T1_RT_TOA/LC08_148044_20130425").normalizedDifference(['B3', 'B6']).rename('MNDWI')
gives us MNDWI Band
In Sentinel 2 the resolutions of Green Band (B3 is 10m) and SWIR Band (B11 is 20m) are different.
How to change SWIR Band to be finer resolution so that I can apply normalized difference?
I believe the tool you are looking for is reclassify. You can create classes for your different NDVI sets. Furthermore you have the option to choose different classification method such as natural jenks, geometric, percentile, etc. I am working on a similar study in which I have used 4 classes; barren, grass/pasture, woods, and forest.
Reflectance is a values that lies between 0 and 1 because it is the fraction of the light the is reflected to the sensor. Negative values are therefore not possible and could be due to artefacts (e.g. some atmospheric corrections). You should always make sure that your data is well calibrated before using those indices. As mentioned by @whuber, in your case valid values should be truncated in 0-10000 which is the interval used to store 0-1 values in integers.
Best Answer
Yes, you can directly apply the normalized difference between Green and SWIR without any issues. You can even apply any kinds of calculation between Green band of Sentinel and SWIR band of Landsat or whatever data source if you would like to.
In Google Earth Engine, the resolution of the resulting MNDWI (or Landsat/Sentinel/MODIS image) is dynamic, i.e. undetermined until you ask GEE to do something that requires a determined resolution.
For example, when you add the resulting MNDWI image to
Map(usingMap.addLayer(mndwi)), GEE will determine the resolution of this image on the fly based on your current zoom level. If you want to fix a single resolution, says 10m, for your image, simply addmndwi = mndwi.reproject(mndwi.projection(), null, 10)before adding your image toMap.When exporting an image, GEE requires you to specify a resolution for your image, usually via the
scaleargument. Yes you can pick any value for the resolution.More on image resolution in GEE from here.