Dear all,
I am looking for high spatial and temporal resolution NDVI (or other vegetation indicators) satellite data. So if you have experiences and verification researches, I would like to have your suggestion on which NDVI is high quality that can be used as reference estimate ?
Some of the online NDVI products are of moderate resolution. If you can have access to a high resolution, try to develop your NDVI image in any RS software package.
the Sentinel 2 provides 10m multispectral+pan images from the earth and it is for free. you can calculate NDVI for each pixel. Though there are NDVI product, please Google: GIOVANNI-TOVAS. It is a visualization tools that provides different satellite product.
If you need high resolution images (quickbird 2 or worldview 3 etc.), you have to pay around 20 USD for each square kilometer.
Dear Wuletawu,
Here are some further questions and comments on your request:
- How high is "high enough" in terms of spatial resolution? What is your application, and what resolution would be adequate to address it?
- Does it matter where or when those data were acquired? What is your purpose?
- Please remember that, due to technological constraints, the higher the spatial resolution the lower the revisit time. Hence, in general, if you insist on "very high" spatial resolution (say, sub-meter), you will be able to acquire such images very infrequently. On the other hand, if you really need a very high temporal resolution (i.e., a high frequency of revisit), then you will have to settle for much lower spatial resolutions.
- Also be advised that high spatial resolution images are generally expensive, as Farid mentioned earlier: they are acquired by private companies and are suitable for commercial applications, so you will have a very limited access to these data unless you also have a big budget.
- If and when you do get hold of satellite data, please consider using modern, state of the art algorithms to take advantage of the cutting edge technology: there is no point in using an obsolete algorithm like NDVI, which was invented in 1974 at a time when Earth Observation satellites only had two broad, crude, uncalibrated spectral channels, to analyze multispectral data today. Technology and the science of remote sensing have improved drastically over the last few decades, so you should look into current approaches, such as retrieving the Fraction of Absorbed Photosynthetically Active Radiation (FAPAR), or the Leaf Area Index (LAI), or the albedo, or some other quantitative property of plants rather than relying on a simplistic and antiquated formula. You will find more information on this topic here:
http://fapar.jrc.ec.europa.eu/Home.php
- Note also that there is no such thing as a "best" satellite data set: this is like asking "which tool is the best to build a house?". Each instrument is designed to address a particular problem, and you may need multiple tools to be applied to multiple datasets to address your problem.
- In this regard, it is also critical to keep in mind the level of accuracy you will need to achieve in your application: unless you constrain your approach in this way, any answer will be acceptable. Hence, you should select your data input and algorithms to match or exceed your accuracy expectations.
- Lastly, if you clarify what you are trying to achieve and your resources and timing constraints, it may be easier to provide you with more specific suggestions.
Best regards, Michel.
Dear Michel,
Thank you so much for the detail discussion of the issue.
I exactly know what you mentioned as i have some experience on the rainfall products. I also know that thee is no one "best" product. However, from the researchers point of view, some researchers did already verification of the products and I am planning to build my research based on those results. I will clarify my need as follows:
1. High spatula resolution: like few kilometre e.g. from 10 km to 25 km resolution.
2. Temporal resolution of daily to weekly (or even monthly). I know modIs and it is relatively good products, but in terms of data length, they are short.
3. My applications is to use these products to study resource/land degradation at regional scale in africa (particularly east africa)
4. I need long term data like from 1980 or so until present
5. Regards to the indicators, yes sure I will consider to use what ever (e.g. Fraction of Absorbed Photosynthetically Active Radiation (FAPAR), or the Leaf Area Index (LAI)) is more accurate for my application.
Thank you, and looking forward for you suggestions based on this comments.
best regards,
wuletawu
Hello Wuletawu,
Based on your requirements of NDVI it seems NOAA's AVHRR would be something to look at. Daily records of NDVI data on a 0.05 degree by 0.05 degree global grid from 1981 to 2013 are freely available. These NDVI products are derived from Surface Reflectance data acquired by AVHRR. You can download the data from EarthExplorer under the category of Vegetation Monitoring - NOAA CDR NDVI.
Having said that, I completely agree with Prof. Michel's insightful answers. NDVI is becoming outdated with the current set of advancements in remote sensing technologies and you may want to consider using some more sophisticated indices if it helps justify your research and provided there is availability of data.
For regular data, you can use modis data,
and better information Sentinel 2, there is a RED Edge band, so get vegetation for spectral signature.
- Badges
- Science topic