The entire dataset is 12TB because it contains _daily_ global 25x25km maps for all years between 1950-2099 probably for a bunch of different variables.
You don't need to download the entire thing to get the gist of it, although it probably would have helped if they also delivered a digest version with for example yearly or 5-yearly averages.
Edit: There is also a THREDDS data catalog present which allows you to extract slices or subsets of the entire data. Other than that it's missing a nice visualization tool I'd say its a pretty reasonable way of making such a huge amount of data available.
Would be great to see a nice visualisation tool built on top of OpenStreetMaps or something like that which allows you to see the predicted temperatures for a given location though time. Sadly this is way over my capabilities, also a 11 TB dataset is not easy to work with :(
It would literally take me 546 Days 19 Hours 30 Minutes 40.26 Seconds to download this dataset. I'm sure it reads like horror though.
I recall reading about an investing company making a pretty penny by using data like this to forecast the growth of certain crops.
If the earth is 509 million square kilometers and the granularity of the data is 25sq km this means we have about 20.3m data points. It implies then on average each point is 540mb. The dataset seems to be broken down by year rather than location though, a UX issue I'd say...
For some helpful visualizations of these data, check out http://climateinternational.org
This article is about creating the projections and that they are available. Is there an article describing what the projections have projected?
For some helpful visualizations of these data, check out: climateinternational.org
Business idea for anyone with some time on their hands. Set up a site, and sell harddives with this data already on it.
NASA has released detailed global climate change projections. If you would like to know what these projections are please drink from this data firehose.
Let's see if these fare any better than the projections used by the IPCC ...
http://esr.ibiblio.org/?p=5297
"The most recent climate model simulations used in the AR5 indicate that the warming stagnation since 1998 is no longer consistent with model projections even at the 2% confidence level."
Kudos to them for publishing these, though. If they are rotten, it ought to be easy to tell.
Will it rain next weekend?
Several comments in this thread mention visualization. These data can be easily visualized with the Unidata IDV [1] (same ppl that make THREDDS). For those interested, go to the IDV dashboard, "Data Choosers" tab and enter
http://dataserver3.nccs.nasa.gov/thredds/catalog/bypass/NEX-...
for the catalog. At that point you can browse the dataset, add the data as a data source, subset it (serverside), and visualize it. I and others have made a bunch of videos of how to use the IDV[2].
[1] https://www.unidata.ucar.edu/software/IDV/
[2] http://goo.gl/n0Frpb