My question is based on the paper Using NOAA AVHRR and SPOT VGT data to estimate surface parameters: application to a mesoscale meteorological model.

The authors have calculated albedo, surface emissivity and thermal inertia for a small region in Europe using the AVHRR data.

In a similar way I have land cover information at 30 meter resolution derived from Landsat images.

I need to calculate albedo for a small region and a climatological period(say Boreal summer). I used the Semi Automatic Classifer Plugin from QGIS to download Landsat images for the period June 1st through August 31st(2014 and 2013) and I got very few images over my region. I need at least 50 to 70 images(more would be better).

I believe MODIS satellite images provide an alternative to calculate the albedo. It is at a lower resolution but I believe a downscaling can be done. My question is can it provide sufficient number of passes in order to come up with a climatological albedo ? The presence of clouds is not a problem as cloud masking can be done.

My ultimate purpose is to use these parameters such as albedo in a meteorological context such as with a weather model that predicts the weather.

  • Is there a reason for you not using the MODIS Albedo product? Commented Jun 14, 2015 at 11:07
  • The MCD43 product series is on a 16-day periodicity with a spatial resolution of 500m. If this is not enough for you, you should atleast read up on their methodology, to give you a starting point for making your own data - see umb.edu/spectralmass/terra_aqua_modis/… for information on the methodology and data access. Commented Jun 14, 2015 at 11:36
  • @MikkelLydholmRasmussen - If you can summarize all the MODIS options that is relevant for my question I will happily upvote your answer. All I need is 500 m resolution and a daily pass.
    – user36959
    Commented Jun 14, 2015 at 11:42
  • All you need is a tall order. It is not produced as a standard product. You will have to work on creating it yourself. You may be able to produce it using the BRDF/Albedo model parameters - MCD43A1. Commented Jun 14, 2015 at 13:04
  • @MikkelLydholmRasmussen - when you say creating it myself you mean use Band Math to calculate albedo right ?
    – user36959
    Commented Jun 14, 2015 at 13:20

1 Answer 1


I would strongly recommend you use the MCD43 product, instead of calculating the albedo product by yourself, for a couple of reasons:

1) Albedo products are not that easy to calculate. Based on your previous question asking if that could be achieved by using Band Math, i'd assume you, at least at that time, didn't fully understand the algorithm that goes into it. Therefore it would be a pain to come up with a convincing albedo data by yourself.

2) Whenever a new product is being used in a scientific paper, you'll have to provide information regarding its accuracy. On the other hand, MODIS albedo product is a widely accepted and commonly used albedo product. If you are only interested in the science question as you described in your question, it is not worth taking the great pains to derive a product by yourself and try to defend it.

3) In reality, most of the time, you really don't need a daily albedo product. Lacking such a need is actually one of the reason why they don't provide daily albedo product readily. Boreal forest is my focus too and from my experience, in a 16-day window the forest albedo usually stays quite stable (unless it is during the transition periods between snow-on and snow-off). So in most cases you can use the 16-day albedo without compromising accuracy.

  • what you are saying is that the climatological product and the 16 day product are not likely to be very different. Am I right ?
    – user36959
    Commented Jan 2, 2016 at 1:33
  • could you also explain why the authors of the paper that I have cited calculate the climatological albedo?
    – user36959
    Commented Jan 2, 2016 at 1:34
  • @gansub It appears to me that the paper was talking about a weather prediction model. Atmospheric science is not my expertise but i'd assume they were in need of high temporal resolution albedo inputs because weather shifts quickly. I don't know your specific needs are and how you are going to use the albedo product, but chances are you don't have to have a daily albedo product in order to address your question. And to your first question, yes, as long as you are not looking at something that changes quickly like weather patterns, they are not that different.
    – TonyC
    Commented Jan 2, 2016 at 2:01
  • understood and I appreciate your response. Would this apply in complex topography as well ? I mean inside a mountainous valley during say a monsoon period ? or doing periods of snowfall some days and others clear weather ?
    – user36959
    Commented Jan 2, 2016 at 2:57
  • @gansub Topography will have some impacts but since the modis product you are getting are essentially mean values both spatially and temporally, as long as you are aware of such a factor and acknowledge it in your paper, i doubt it will impose significant negative impacts on your results.
    – TonyC
    Commented Jan 2, 2016 at 5:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.