Dear all,
I'm doing my time-series research using MODIS snow products' snow albedo (version 6) band to calculate daily snow water equivalent. However, my study area is heavily cloud-contaminated. Therefore, the majority of the pixels have been masked out due to the MODIS algorithm. The pixels are not enough to analyze for seasonal behavior of snow water equivalent. I found MCD43A1.006 MODIS BRDF-Albedo Model Parameters Daily 500m. But it is wavelength-specified albedos. Will it be possible to apply the bands corresponding to the snow reflectance curve? Thank you in advance.
My study area is actually a head-water basin, and really big in terms of area. If I go for either Sentinel or Landsat, it will be extremely and unnecessarily detailed for my research area, in addition to more preprocessing steps such as mosaic, merging and such. Thus, I went for MODIS. Another confusion is to find out the snow albedo, and I'm sure snow albedo and albedo are totally different definitions. Hope I answered that make sense. And many appreciations for your question.
To Analysis The Albedo Changes, Use The Data Of The MODIS Sensor Terra Satellite In The Form Of 8-Day Composite Images.
The Advantage Of 8-Day Data Over Daily Data Is That These Images Are Less Cloudy.
With The least Amount Of Information Lost Per Frame, The Average Albedo Can Be Obtained.
Finally, My Suggestion is To Use The Google Earth Engine System for Accurate and Practical Processing With The Lowest Error Rate, Because it Uses Ready-Made Corrected Products.
The presence of clouds in images can make it difficult to separate classes. These clouds are the cause of error, especially in spectral indices. Therefore, it is recommended that you mask the cloud as much as possible.
Is it possible to reconstruct the image information by masking the cloud?
One misconception is that by masking the cloud, it can be fabricated. Keep in mind that the cloud causes the loss of surface information. That is, the sun's waves do not reach the earth's surface. As a result, we do not have surface waves and energy does not reach the sensor.
By masking the cloud, the DN values and values that the cloud has are removed from the image and are not used in calculations and statistical analysis. Note that if your goal is to use statistical information, the pixel values of the cloud should be converted to No Data.
Thank you for your answer, sir. I totally understand about the disadvantages of the cloud contamination as well. I also used GEE for my research as well. And yet, I still have to test it. Again, many appreciations for your suggestion.
ESA Surface albedo 10-daily gridded data from 1981 to present may help. The website is https://cds.climate.copernicus.eu/cdsapp#!/dataset/satellite-albedo?tab=form
Haiqing Soong , is there any way I can incorporate that dataset with google earth engine? I don't find any corresponding image collections. And many appreciations for your suggestion.
Swun Wunna Htet this data is in netcdf format. You can transfer it to tiff format.That may help.
- Badges
- Science topic
- Similar topics
- Mathematics