I want to study the trend of cloud cover over an area of interest. I want to do monthly, annual and seasonal analysis for the same. I want to know about the generation of annual cloud cover data from monthly data
The simplest is to calculate the average annual values from the monthly values. After all, the same is true for other climatic and meteorological values, such as temperature, pressure, wind, etc.
As Milivoj wrote, you can obtain the annual cloud cover by adding the twelve average monthly cloud cover values and then dividing by twelve to give the average annual cloud cover.
On the other hand, if you want the total time there was cloud cover, you will have to add the monthly values but not divide by twelve.
You can do the average of monthly or weekly data to get annual cloud cover over an area. And if you want to get the maximum cloud cover over a area for a particular time interval then you can also use Maximum Value Composite(MVC) of the given multi-temporal datasets.
MVC is pixel-by-pixel process in which each value is examined, and only the highest value is retained for each pixel location. After all pixels have been evaluated, the result is known as an MVC image.
Because the months are not all equal in length, strictly speaking each month should be weighted by the number of days as a fraction of the year, e.g., January is weighted by 31/365 but April by 30/365. However, I doubt that those slight changes would have a significant effect.
It is rather easy task to perform, you can take each month cloud cover and multiply it by the days in the months divided by the total days in the year. Then add that to the statistics. For seasonal, it is the same, but you have less numbers of days within each season.
Your question seems very simple. As suggested above by several researchers, you just need to sum up the monthly mean and then divide by 12.
However, if you want to use a certain software, such as ArcGIS, SPSS, Excel etc. and you dont know how to use it to do this, then this is another question.