It is possible to observe changes in the weather patterns from an analysis of the data. The success of this depends on the size of the change compared with the natural variability. Examples include the change in the rainfall patterns in the Perth region in Western Australia. (e.g. http://www.water.wa.gov.au/PublicationStore/first/88706.pdf ). I have been involved in a study looking at the change in the spatial distribution of rainfall through time across the Australian continent. This is currently under review - once it is accepted, I can send you a copy.

Obviously, it is possible. But you should consider some points. Firstly, i should say that the climate change scenarios vary greatly in different location, highly dependent on distance from the sea and atmosphere cyclones. Secondly, depending on the site you must have at least 50 to 90 annual and better monthly data. The more is extreme the climate, the lower you need data. I said monthly, because climate change may not impact on the location at annual scale and you should consider more precisely.

For extracting the change trend in this data I offer to use smooth spline.

Of course there are 'local impacts' due to climate change. As Ehsan pointed out, since climate change scenarios vary greatly with different location, 'region specific' or micro level studies have the significance. But when stating 'there is a change', it will be always better and useful to find out the attributing factor/s also. Simple example, rapid urbanization or land use change can change the temperature and hence the hydrologic cycle. This is also a 'climate change' by definition, but different type.

My dear, thanks to all. I would like to repeat last part of my question.

" Is it better to search for the answer by analyzing local data or to use some statistical downscaling methodologies applied to regional climate change parameters?"

I'd go with a debiased dataset such as HCN. Local data (first option) will have errors including those from changing equipment, relocating, and similar. I don't really understand the second option - are you talking about a PRISM-like approach to fill in the gaps between the observation points?

Dear Muhammad

If I get your meaning right as I mentioned before, if the climate condition of your location has some extreme conditions, according to the type of the extreme you should downscale or upscale your results. As an example, if the humidity of your local is very low you should downscale the results and vice versa. Another example is polar conditions which has the higher temperature increase which if we extrapolate that to whole globe, we may overestimate the impacts. So, this one should be downscaled too.

Dear Ehsan

Thank you much for the explaination, now I fully get your reply.

Dear Muhammad

the answer is yes, we can assess climate change from local data,

I have about three publications about that,

1- https://www.researchgate.net/publication/49590057_Temperature_Trends_and_Distribution_in_the_Arabian_Peninsula

2- https://www.researchgate.net/publication/237020126_Spatio-Temporal_Variability_of_Evapotranspiration_over_the_Kingdom_of_Saudi_Arabia?ev=prf_pub

3- https://www.researchgate.net/publication/237020125_Analysis_of_Evapotranspiration_Variability_and_Trends_in_the_Arabian_Peninsula?ev=prf_pub

Best regards

Article Temperature Trends and Distribution in the Arabian Peninsula

Article Spatio-Temporal Variability of Evapotranspiration over the K...

Article Analysis of Evapotranspiration Variability and Trends in the...

Dear Muhammad

Thanks a lot for sharing the publications