What does the band's value signify of Landsat-8 L1C, Sentinel-2 L1C and MODIS MCD43A4

Data Sources
1

I was doing some processing on the Landsat-8 L1C bands downloaded from sentinel-hub. All the pixel values were between 0-1. Where as when i downloaded the same Landsat-8 LIC (level 1) scene from USGS EarthExplorer. Values of all the bands were between 0-65535. So, I know scenes downloaded from USGC gives you the DN values in level-1 product. So, what does the pixel values denote when downloaded from sentinel-hub for different sensors i.e. Landsat-8 L1C, Sentinel-2 L1C and MODIS MCD43A4

2

This FAQ will hopefully answer your question:
https://www.sentinel-hub.com/faq/how-are-values-calculated-within-sentinel-hub-and-how-are-they-returned-output

We do suggest you check our latest API, which gives you control over what you are getting - reflectance or DN
https://docs.sentinel-hub.com/api/latest/#/data/Landsat-8

3

@gmilcinski I wanted to compute Land surface temperature (LST) as mentioned here from DN. So, how do i get the MLT.txt or MTL.json to convert the DN to radiance also for further calculation. I don’t see any documentation in the above link.

4

You are probably referring to MTL, right? You can get these directly on AWS, e.g. this link

So what you can do is:
-use WFS service to get relevant information about the landsat path
-use DN data from Sentinel Hub API
-compile a path to MTL file
-calculate

5

@gmilcinski Can you please elaborate how do i get landsat path and row for a date range for my area of interest as you mentioned using WFS service?

I have a suggestion. Isn’t it defying the purpose of eo-learn. If i have to iterate over all the landsat-tiles(path and row) to get their respective MTL file. Where as the purpose of eo-learn it to save from this hussle. Can we add the MTL in the existing platform itself the way QA band has been added?

6

By using WFS request, you can actually get information about the AWS location, e.g.
https://services-uswest2.sentinel-hub.com/ogc/wfs/INSTANCE_ID?REQUEST=GetFeature&SERVICE=WFS&VERSION=1.1.1&SRS=EPSG:4326&BBOX=-102.86243167867883,32.12394382391051,-103.3169910048507,31.899492878755403&TYPENAMES=DSS6&TIME=2019-03-31/2019-06-30/P1D&MAXCC=30&OUTPUTFORMAT=application/json

you will find:
http://landsat-pds.s3.amazonaws.com/c1/L8/031/038/LC08_L1TP_031038_20190625_20190625_01_RT/LC08_L1TP_031038_20190625_20190625_01_RT

For your suggestion - you are most welcome to add this functionality to the eo-learn. eo-learn is an open-source package and community contributions are more than welcome.
Adding MTL to the core Sentinel Hub service would be probably possible, but currently not high on our priority list.

7

@gmilcinski https://docs.sentinel-hub.com/api/latest/#/data/Landsat-8 does this gives you surface reflectance or TOA reflectance? Reason being i am asking this because on USGS you have to first request for surface reflectance product of landsat-8.

8

These are Top of the atmospher data (more info here https://registry.opendata.aws/landsat-8/)

Collection 2 data are being prepared by the USGS at this moment and should become available on AWS in a couple of months. Once they are, we will integrate them as well.