1
$\begingroup$

I have climate data for future projection scenarios of the RCP type only (i.e., RCP 4.5 and RCP 8.5) and land cover data for future projection scenarios of the SSP - RCP type (i.e., SSP 2 – RCP 4.5, SSP 5 – RCP 8.5). To build species distribution models, I need to combine the climate and land cover data together to facilitate the extraction of data associated with each SSP-RCP combination.

Since the climate data do not have SSP-RCP combinations like the land cover data, I was wondering if it is correct to associate the climate and land cover data as follows:

Climate RCP 4.5 with land cover SSP 2 – RCP 4.5

Climate RCP 8.5 with land cover SSP 5 – RCP 8.5

Or should I change the climate data to have SSP-RCP combinations like the land cover data to have these associations ?:

Climate SSP 2 – RCP 4.5 with land cover SSP 2 – RCP 4.5

Climate SSP 5 – RCP 8.5 with land cover SSP 5 – RCP 8.5

$\endgroup$

1 Answer 1

1
$\begingroup$

The SSP scenarios were designed to be in accordance with the RCP scenarios. As you illustrate, each SSP scenario is related with one or more RCP scenarios. Fore more details, see e.g. https://www.sciencedirect.com/science/article/pii/S0959378016300681?via%3Dihub and related literature.

The RCP scenarios in essence only contain physical information (energy balance), whereas the SSP scenarios contain different socio-economic scenarios leading to this energy balance. As such, the SSP scenarios (which can influence land cover, depending on socio-economic changes) are relevant for your question.

However, in CMIP6, SSP-RCP results from global climate models should be available. If global climate model data is of a good enough resolution for your question, you should also be able to use these. Yet, given that you're studying species distribution, I would assume that you need a finer resolution. Then, data might not yet be available, and just RCP data from a GCM-RCM (regional climate model) ensemble might be the best choice, as discussed above.

Besides: take care with the climate data and postprocess according to your needs. Inputting raw climate data (with biases for certain variables) might lead to the wrong results.

$\endgroup$
3
  • $\begingroup$ Thank you very much for your answer. I am using NA-CORDEX climate data (regional climate model) with a resolution of 28 km (I couldn't find finer resolutions). These data include only RCP scenarios. On the other hand, WorldClim (global climate models) include combinations of SSP-RCP scenarios at 4 different spatial resolutions: 30s (~1 km), 2 minutes, 5 minutes, and 10 minutes. Even though regional climate models are more precise than global climate models, wouldn't WorldClim data be a better choice since they account for SSP-RCP combinations? $\endgroup$ Commented Jul 11 at 6:10
  • $\begingroup$ Indeed, more recent global climate model data can be a better choice, especially if they have a finer resolution. In that case I would suggest that you choose a global dataset with the resolution depending on your other input data. $\endgroup$
    – Jorn
    Commented Jul 12 at 7:34
  • $\begingroup$ Jorn: Great ! Thank you very much for taking the time to respond to me. $\endgroup$ Commented Jul 12 at 22:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.