What to be aware of
Spatially Coherent Projections: what to be aware of
There are a few things to be aware of when using the Spatially Coherent Projections (SCPs).
1. The SCPs are not available in a user-friendly format
The SCP data is available in .csv format from the UKCP09 .csv archive. You will not be able to access the SCP information or create related graphical output directly via the UKCP09 User Interface .
2. The SCPs are not a replacement for UKCP09
The SCPs provide eleven projections that are spatially coherent across grid squares. They are designed to complement the probabilistic projections, which are not spatially coherent in nature. The SCPs are derived from the 11 regional climate model simulations used in UKCP09, and are scaled to achieve a better representation of the wider set of outcomes for future global temperature rise included in the UKCP09 probabilistic projections. However, bear in mind that the SCPs should not be treated as a replacement for the probabilistic projections. This is because the SCPs still under-sample the full set of possible outcomes available through the probabilistic results, because the latter consider a wider range of possible future patterns of climate change. Nevertheless, the SCPs may produce specific examples of changes in some impact indicators lying outside the 10-90% range of the probabilistic projections.
3. The SCP output cannot be combined with the UKCP09 probabilistic output.
You should not attempt to create customised climate projections by using climate variables from the SCP projections in combination with other climate variables from the UKCP09 probabilistic projections. The SCPs also cannot be used to drive the UKCP09 weather generator. Choose whether you want to use the SCP projections or the UKCP09 projections for your analysis, but do not combine them.
This is because the two types of projections have been created in different ways. The SCPs have been generated from the 11-member RCM projections to achieve spatial coherence, whereas the probabilistic projections were generated by combining a broader set of climate model projections, and therefore provide a more comprehensive representation of uncertainty.
However, the probabilistic projections can be used to provide context for the SCPs, as explained in point 4 below.
4. Each of the 11 model runs that underpin the SCPs is equally plausible.
In addition to providing a less complete sampling of uncertainties, the SCP differ from the UKCP09 probabilistic projections in that no weighting has been placed on individual model projections in order to account for variations in the reliability of their results.
For these reasons, we recommend that SCP data be checked against the UKCP09 projections to see where each ensemble member lies within the probabilistic distribution. This can provide an assessment of the proportion of the full probabilistic range that is explored by the SCP data.
5. The SCPs do not represent a smooth trajectory of change through the 21st century.
The SCPs are provided for seven overlapping thirty year periods. In any specific SCP, the trajectory between consecutive periods will not be completely smooth, partly because climate change will not necessarily progress in a completely linear fashion through the 21st century, and partly because the climate change signal for any given period and location is modulated by the unpredictable effects of natural variability on decadal or longer time scales. This is also likely to apply to the evolution of 21st century in the real world. While realistic in this respect, this lack of smoothness may make the SCP trajectories difficult to use for some adaptation decisions requiring timing and investment choices. When doing this it may be more appropriate to use a set of smoothed trajectories approximately characterising the secular climate changes underlying the time series of 30 year average values provided.