As climate variability increases, regions are becoming more vulnerable to its negative impacts. The agricultural sector will be hit hardest by climate change, not only suffering from direct influences but also from multiplier effects. In order to adapt, farmers need to know how weather will change in their area – requiring data that goes beyond usual national forecasts.
The CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) data portal recently released downscaled, ready-to-be-used data corresponding to the IPCC’s 5th Assessment Report (AR5). The data include around 25 Global Climate Models (GCMs) for four Representative Concentration Pathways (RCP) and four time periods (2030s, 2050s, 2070s and 2080s).
CCAFS-Climate, created in 2008, houses thousands of gigabytes worth of global datasets for climate change projections that can be used in climate change impact assessments. The difference between these data and, say, those provided by GCMs at raw resolution is that they have been downscaled to make them useful on smaller scales and in countries for which complicated Regional Climate Models are not available.
Learn more: Perspectives on downscaling of climate data
Even though GCMs can be very precise, their results come on scales on the order of 100 km – too large to be of use when trying to calculate climate risks for a municipality, region or watershed. Downscaling techniques allow researchers to obtain regional rather than global predictions of climatic changes, and in fact, statistical methods (such as the delta method used with these CMIP5 data) can reach resolutions as fine as 1 km or less.
Without these high-resolution data inputs, the production of precise and accessible assessment tools for conservation planning, niche modeling, crop modeling or biodiversity monitoring would be greatly hindered, if not impossible.
CCAFS-Climate launches new user interface
To accompany the release of the highly anticipated IPCC 5th Assessment Report data, CCAFS-Climate is also rolling out a new and improved user interface that enhances the functionality and visual appeal of the portal. The changes include a fancy new search engine for the easy location of datasets, the option of downloading files by geographic tile rather than the entire globe, and a new version of the MarkSim weather generator for CMIP5 data.
The storage capacity of the CCAFS-Climate portal has also been increased, now weighing in at a hefty 14 terabytes of data.
In 2013, more than 14,000 people visited the CCAFS-Climate data portal. Twenty peer-reviewed publications cited use of it data, bringing the total sum of citations since 2009 to almost 90. Users have accessed the portal from 164 different countries – almost 85% of all the countries in the world.
Future plans for CCAFS-Climate include the possibility of housing data from crop models in addition to climate models, and the extraction of daily GCM data at its original resolution.
Visit CCAFS-Climate and get the data while it’s hot!
CAUTION: Technical content below
Changes in the CCAFS-Climate portal as of 24 January, 2014 are as follows:
- New user friendly search engine
- Option of downloading files by geographic tile
- Full set of downscaled IPCC 5th Assessment Report GCM data
- New version of MarkSim weather generator for CMIP5 data
- More than 20 TB storage capacity
Technical statistics on new files and portal capabilities:
- 17,808 new files for CMIP5 data
- 2.7 TB of new data
- 4 RCPs (rcp2.6, rcp4.5, rcp6.5 and rcp8.5)
- 106 GCMs (about 25 models per RCP)
- 4 future time periods (2030s, 2050s, 2070s and 2080s)
- 5 climatological variables
- 4 resolutions, the highest at 1 km2
- More than 100,000 files in the database for diverse downscaling methods
- 14 TB of total files hosted
READ MORE: New updates to Marksimgcm weather generator
Watch presentation: Downscaling of GCM for its use in Agriculture and NRM Research
The CCAFS-Climate team thanks Jaime Tarapues, Julián Ramirez and Carlos Navarro, all of whom have contributed enormously to this new release.
Blog by Caity Peterson, with input from Carlos Navarro.