Recently, modern tools such as measuring carbon in the air, measuring solar activity, reading tree rings, studying air bubbles in antarctic ice,....etc, are used to read clues that were left long ago in order to determine earth's past climates and build a picture of future climates. However, primary parameters for measuring climate change remain bioclimatic variables, mainly temperature and precipitations. Data for past, current, and future of these two parameters (plus others) can be freely downloaded from http://www.worldclim.org/
Please have a look under my profile at our article on Pinus mugo in Majella of this year. We asses meteorological and dendrological data against the regional prediction of climate change at a local spatial scale at a temporal scale of a century. Meteorologic data usually do not go back further in time.
For the same area, a geomorphological assessment of climate change is available on the millennial time scale (referred in our endemics article). further, we assessed the accuracy of interpolated bio-climatic variables (bear article) against independent empirical meteorologic data.
Time series analysis of several climatic variables like- minimum temperature (°C) , maximum temperature (°C), average temperature (°C), precipitation (mm), temperature Seasonality, precipitation Seasonality, solar radiation (kJ m-2 day-1) etc can be used as an indicator of climate change. You can download these variables for different spatial resolutions, from 30 seconds (~1 km2) to 10 minutes (~340 km2). http://worldclim.org/version2. To create these values yourself, you can use the 'biovars' function in the R package dismo.