Dear Sharma,

I asked something similar (I tried to merge DEM and Temperature data) and several colleagues sent me very interesting answers.  Maybe these answers can help you.

https://www.researchgate.net/post/Is_there_any_method_to_build_a_map_through_combination_of_topographic_data_DEM_and_temperature_data_points_weather_stations

Dear Sharma,

the most straightforward way to do this is, in my opinion, combining the two rasters as bands in a multiband raster dataset. This can be done using the Composite bands (Data management) tool, paying attention to the fact that the output raster dataset will take the cell size from the first raster band you put in the list.

For instructions on how to use this function: http://resources.arcgis.com/en/help/main/10.1/index.html#//00170000009p000000

I really hope this helps.

Best regards.

http://resources.arcgis.com/en/help/main/10.1/index.html#//00170000009p000000

 Thanks Luca and Alexandre. I will try out the suggestions. @Alexandre Ribó which method actually worked well for you?

Dear Sharma,

I find Matlab or R the easiest softwares to calculate the relationships between two rasters. Particularly if you have many time steps and want to do sofisticated statistical analyses. 

Best,

Matti

@Matti. Can you suggest some link or something which can guide me how to go about it?

Dear Sharma,

here is some example code

http://www.gisarea.com/topic/3292-spatial-analysis-regression-and-correlation-between-2-rasters/

You can read in to matlab e.g. GeoTIFF, but you need to make sure that the rasters have same extent and resolution. This you can fix easiest in ArcGIS. After that you can use the example code and just google the needed commands to matlab. 

Best,

Matti

Thanks!

You cannot expect reasonable results. Unfortunately, the average elevation of a TRMM pixel (app. 28 x 28 km2 at the equator) on which you could regress the precipitation totals over a given time span, are a poor measure of the actual altitude.

How about comparing the two with swath profiles? Saw some articles where people have compared the two by drawing swath (in ArcGIS).

A swath profile allows you to have an idea on the variability of elevation in a strip o a certain width containing the swath line. Unfortunately the average elevation over a pixel is averaged in two dimensions. I cannot see how you could use this information. Be also aware of the fact that the precipitation gradients with altitude generally grow non-linearly up to a certain elevation above which they tend to disappear. 

Dear Sharma,

If you are familiar with ILWIS open source GIS and RS tool. You can use cross operation performs an overlay of two raster maps. Pixels on the same positions in both maps will be compared and it will give you an output cross map and a cross table.

Abeyou W.

 @Ezio Would you please elaborate why swath would be unreasonable? Even if it is averaged in two dimension, wouldn't that be average elevation over that area plotted against the precipitation? And that is what we want to see. How precipitation is varying with elevation across a traverse.

And regarding the growth of precipitation, wouldn't it show that relation when it is plotted against elevation?

Also, regarding your earlier answer about pixel resolution of precipitation and elevation. What if I increase the resolution of precipitation instead of regressing elevation? Thereby preventing loss of elevation data?Would that be reasonable?

you can using mapcalculator with this formula: (TRMM*X)+DEM. The value X is depend on the number digit of TRMM, for example the value raster of TRMM=40 and DEM=40 then X=100. The result of raster valur is 4040.

Dear Shubhra Sharma , were you able resolve this issue? If yes, please can you give a guide? I am facing a similar issue where I need to integrate GPM rainfall data into DEM for stream power analysis in mountainous region. Thank you.