You can get MODIS images from this site:

http://ladsweb.nascom.nasa.gov/data/search.html

Info about different MODIS products are in the following site:

https://lpdaac.usgs.gov/

and you can use HEG for conversion of hdf format to TIFF

http://newsroom.gsfc.nasa.gov/sdptoolkit/HEG/HEGDownload.html

Also, if you have a number of images, then it is the best to work on the software packages that have good scripting capabilities (e.g. ENVI). I prefer Matlab :)

I am still downloading MODIS 16 day VI data and land surface from 2000 to present but the analysis and data processing are my problem.

I am interested in using matlab and more especially wavelets but how do i go about it. My programming skills is not all that as a result I was looking at the Wavelet Toolbox but I still have difficulties of importing the images and doing my analysis. I wish to see the wavelets of the MODIS data (2D) and the climate data (1D) and then make my comparisons

matlab has built-in reader for importing hdf files. I'm sure you can find more if you just make a google search on it

Okay Thank you! What possibly can I do next? How do i use the matlab wavelet. Do I have to do something to the MODIS data before?

I think before going to the data analysis you must have a clear view of what you are going to do. I recommend to have a literature review on the papers that used MODIS for drought analysis. Especially, find a review paper to be able to compare different methods.

Drought stress mean water stress hence, every plant including grass has limited ability to this stress so, there are storing relationship between drought and grassland degradation

First you need to establish the grass cover in the targeted area using the Remote sensing data and later relate it to the rainfall, humidity and temperature regimes and see if a close relationship can be established. Very often this may not be related as the grazing pressure can intervene. But at higher rainfall levels this certainly would show up. There could be other pressures such as soil erosion and excessive wind velocity. The model can incorporate all these parameters. This need to be incorporated in GIS and the number of observations that need to be considered depends upon the frequency of drought or rainfall.

Hi Jonathan, you can do the NDVI by MODIS bands and if the indice will low, maybe the grassland will be afected. You must be attempt to caracterize the spectral behavior of the grass in the drought and in other time of no drought to distinguish grass of others uses.

Do anyone know which is be best quality image with high-resolution and how much cost?

before going to interpret the area by images and establishing any kind of relationship u need to know some facts , 1st of all u need to an acquaintance with the area , like the climate of area , its geology, vegetation etc. as vegetation response to weather/climate change fast and quickly than any other kind of factors and among weather elements rainfall/precipitation is having vital effects. so just study the rainfall data of last 10 year or more . u will find that lesser the rainfall in the area different will be the response of vegetation (due to lesser formation of chlorophyll into the leaves) as compare to the vegetation during the higher rainfall season of years. now calculate NDVI of critical year i.e. the year of lesser rainfall/ the year of mild rainfall / the year of heavy rainfall with the help of images of that relevant times. now comes to the drought/ lesser rainfall times, the change in rainfall surely bring change in vegetation which lead to the change in soil structure of the area . if the drought is prolonged the soil will slowly and steadily start losing its strength in the starting years which will be fasten in adventing years, and start degrading and eroding . as soil also differently response to incident energy in different weather conditions which can be identify by image of different seasons. grassland areas are found in the areas where adverse climatic condition are prevalent so a slight change in weather have vital effect on these areas

Now comes to the point-

1 just go for rainfall data of past years of the area

2 classify the critical year of rainfall

3 study the vegetation of the critical years with imagery of that year and derive NDVI,

4 analyse the changes occurred due to change in weather

5 study the soil response also during different critical years

6 u will come to conclusion that scanty rainfall years have more effect on soil and vegetation , lead to slower growth of vegetation and further degradation of land. such changes appear on images directly or after few classification of images

Virender, thank you very much. you have really given me a lot of insight. please if there is more please add more. its like you have given all the answer. thank you once again. please if you have more documents on it could you kindly sent me. i wish we could even collaborate in this research i am doing

This link may help u a bit

https://water.tallyfox.com/hot-topic/soil-moisture-and-hot-days-examined-globally

but this is related to soil moisture and heat . Which indirectly may be having the solution of ur problem for some extent

Hi Virender, I have been trying hard to get information on how to determine the critical years. I have rainfall data of rain gauge stations from 1954-2004. Please could you hint me on how I can use this to determine the critical years?

We used MODIS NPP (Net primary productivity) and GIMMS NDVI data to evaluate spatio-temporal trends of rangeland degradation over the dryland portion of US (17 western states basically). MODIS NPP data are available at 1km resolution, for 2000 to 2009 globally. GIMMS NDVI are available from 1982 to 2000. Depending on how big your study area is (or the scale of the study) , you may be able to use them. If you need to use MODIS reflection data, you will have to pick a suitable vegetation index (i.e. NDVI, SAVI, etc) to match with the vegetation characteristics. If you decide to use vegetation indices, you also have to consider the temporal patterns of vegetation changes (i.e. growing season etc) and then pick the period of the year for your study period. Else, you may end up using time series of data and combine them in a way to reflect the entire year. For example if you integrate all monthly or daily NDVIs over the growing season it will fairly represent the vegetation condition over a given year. This integrated NDVI data product for your study area (i.e. grasslands) could be then compared with the climate data to see their relationships. As climate variables, I would recommend precipitation, temperature, PET (potential evapo-transpiration). Depending on the data availability, you may even be able to derive Aridity Index data product, which is the ratio between mean annual precipitation and mean annual PET. This Index is a very good indicator of drought severity.

Ofcourese, if you search literature on NDVI, SAVI, Drought, NPP, etc with the key words you will find a lot of papers. Most of them are on the African region (Sahel), but there are some on Mongolia as well.

This is very interesting, but requires lot of remote sensing knowledge and skill. I wish you Good luck!!

Hi Ranjani! Thank you so much for the insights. However, is there any possibility of having any practical insight to your work, for example a step-wise procedure?

I would advice the best study areacwould be a Savannah grassland landcover. It's response to drought is usually significantly evident.

Also, in terms of remote sensing technique, try to see the relationship between suitable moisture and vegetation indices using time series of images.

You can also use NDVI to assess health status and try to link it with phenology i.e the the growth stage of the vegetation. This will assist you to compare same growth stages over certain years. Also bear in mind that rainfall is not the only factor that causes rangeland degradation, another important variable is grazing. You can also calculate SAVI as it maximises soil effect especially in heavely degraded areas.. All the best CHARLIE!!!

you can try latest version of IDRISI software. May be a trial version of IDRISI Selva is available at its homepage that will be helpful for the time series analysis