Snow Cover Analysis Using NDSI

The differentiation between clouds and snow can be difficult because of the similarity in spectral characteristics in the visible wavelength range. It is important to differentiate between cloud and snow to improve the accuracy of cloud detection. The calculation of snow cover is also critical in a wide range of applications. Snow cover affects the climate change and global atmospheric circulation. Snowmelt is  a water resource for nearby agricultural lands and groundwater storage in many parts of the world.

Snow and ice hydroclimotology can be a great threat to life by causing monsoon flooding, droughts, landsliding and other complex emergencies.

It therefore makes it necessary to differentiate between cloud and snow to maintain proper analysis and data.

In the following section, we will be discussing a technique on differentiating cloud and snow in satellite imagery and the related theory.

THEORY

The Normalized Difference Snow Index (NDSI) is designed to detect snow over the NDSI range from 0.0 to 1.0 which is supposed to be the theoretical snow range.  Ice and snow have very bright appearance which indicates they have very high visible reflectance (VIS). Snow and ice absorb shortwave infrared wavelength and hence have a very low reflectance in SWIR. This is a characteristic used to distinguish between snow and most cloud types. In Landsat bands, the NDSI ratio is given by-

NDSI= BAND2-BAND5/BAND2+BAND5

METHOD

We considered two images of Band 11 and Band 8A from farmonaut.com which has similar wavelength as Band 2 and Band 5 respectively.  The idea here is to treat the image as an array of pixels. Keeping that in mind, finding NDSI value becomes straight forward. We then need to find the difference between two grayscale images and addition of two grayscale images and find the ratio.

NOTE: It is important to consider two images of equal resolution which here means same height and width. Before performing the following pseudo codes, make sure to check the dimensions of both the images.

 

PSEUDO CODES

 

1. The program will allow the user to add and subtract two images.

 #import the libraries

 #read the two grayscale images

 #perform matrix addition and subtraction of the two images

 #save the two resultant images

 

2. The program will allow the user to find ratio of two images and ndsi values

 #import the libraries

 #read the two grayscale images

 #perform matrix division of the two images

 #print out the values of the division

 #display the resultant image.

 

STUDY

We considered the satellite imagery of Reykjavik, Iceland on 21st May ,2018 of Band 8A and Band 11.

 

Following are the results .

B11-B8A
B11+B8A
NDSI

OBSERVATION

The final image signifies snow as completely black and everything else such as clouds, vegetation, landmass,water as white or gray which means all these ends up having similar pixel value in this image. This helps in accurate calculation of snow cover percentage.

The opposite result can be obtained by changing the sign of the equation. Following is the result obtained.

 

 

NDSI Inverted

In the above image, inverted result is obtained. The bright pixels indicates snow cover and everything else is relatively black.

Either of the two images can be used for snow cover analysis.

 

 

We considered the mountain Tirich Mir in Hindu Kush and performed similar analysis.  Following are the satellite imagery used for the analysis.

 

 

After performing arithmetic operation on the image, we obtained the NDSI images.

 

 

We calculated the the snow cover percentage in both the NDSI images and got the same result that is 41.38%. This makes it certain that both the images are equally accurate for snow cover analysis.

 

 

LIMITATIONS

This approach is completely reliable on the reflectance property of snow and sometimes water, shiny rock surfaces and even vegetation covers can also have similar reflectance properties as snow which can lead to inaccurate results.

 

 

We will keep posting about any such informative information on to our blogs, to help as many people as possible. Farmonaut is built upon a vision to bridge the technological gap between farmers and strives to bring state-of-the-art technologies in the hands of each and every farmer. For any queries/suggestions, please contact us at support@farmonaut.com.

We have some more interesting articles coming up soon. Stay tuned!

Wait!!

Before that…

Follow us at:

FARMONAUT

Satellite Based Crop Health Monitoring, Crop Issue Identification System, Farmers’ Social Network, Govt. Approved Farming Database, Satellite Imagery Access For Research and Much More

Close Menu