Dear Ankur,

Firstly, you can make plots showing the relationships between the two variables for each station. Depending on the characteristics of the area and spatial variability of the variables and their relationships, you might cluster these relationships and make plots with relations between the two variables for specific regions.

Secondly, you can test different functions varying from linear relations, linear relations between for instance log-transformed variables, exponential relations and higher-order functional forms. You might also have a look at the literature on relations between these variables found in other areas.

These analyses can be done in e.g. Excel combined with a GIS package or codes such as R.

Good luck,

Martijn

Dear Ankur,

Since you have a huge amount of data , I suggest that Interpolate them onto a regular grid.

You can use the following guide to do this:

http://www.ncl.ucar.edu/Document/Functions/Built-in/idsfft.shtml

When you have data with regular grid you can use one of these statistics functions, to develop a relationship between your parameters:

https://www.ncl.ucar.edu/Document/Functions/statistics.shtml

For instance, the fifth example at the link below, computes the (Pearson) sample linear cross-correlations at each latitude and longitude of two kinds of data (w(time,lat,lon) and q(time,lat,lon)):

https://www.ncl.ucar.edu/Document/Functions/Built-in/escorc.shtml

To use the above codes, you need to install NCAR Command Language (NCL)

Good luck,

Hi Ankur,

there are several methods to find the relation between two variables, the most known is to use regressions of type: linear, power, logarithmic, ...etc, otherwise you find also the new techniques of artificial intelligence to link the two variables, Artificial Neural Networks ANN, ...etc

good luck