A CMIP5 Model Selection Specific to South Africa's Winter Rainfall Zone
| dc.contributor.advisor | Jack, Christopher | |
| dc.contributor.author | Marsh, Peter | |
| dc.date.accessioned | 2023-03-31T08:27:38Z | |
| dc.date.available | 2023-03-31T08:27:38Z | |
| dc.date.issued | 2022 | |
| dc.date.updated | 2023-03-29T08:51:52Z | |
| dc.description.abstract | This study undertakes a CMIP5 model selection specific to the Winter Rainfall Zone (WRZ) of South Africa, seeking to reduce the range of future climate projections through identifying a subset of models with increased realism and independence. In order to navigate the subjectivity in identifying relevant circulation metrics to assess models against, the ‘Day Zero' drought is used as a characteristic episode. Here initially the extensive literature produced subsequent to the drought has been drawn on to identify and evaluate relevant regional process metrics, before utilising the anomalous conditions during the drought to validate various assessment methods. The dynamics subsequently identified as being most influential to rainfall supply in the WRZ include the South Atlantic subtropical jet stream responsible for steering of mid-latitude storm systems, the South Atlantic subtropical high, and the presence, or preferably absence, of precipitation blocking subsidence, and the prevalence of mid-latitude storm systems, critical for transport and upliftment of moisture to the region. Models were subsequently assessed against these metrics and scored following the technique of McSweeney et al. (2015). Unrealistic models were removed from the ensemble while significantly biased models were also excluded as their absence did not significantly reduce the range of future projections. The same scoring methods were then utilised to create a genealogy of models attaining similar results to that of Knutti, Masson & Gettelman (2013). A subset of 6 CMIP5 models which are more independent and historically more realistic than that of the full ensemble were subsequently identified. While the range of future temperature projections of the final ensemble are somewhat constrained in comparison to the full ensemble, the primary utility is argued to be the reduced number of models where a future researcher may consider each model's projected future climate pathway individually before selecting a model, or models, which best informs their use case, whilst being assured that this model performs suitably well in the region and that the initial ensemble considered adequately represents model uncertainty, while strong similarity between two or more models within the ensemble will not be unduly biasing results. | |
| dc.identifier.apacitation | Marsh, P. (2022). <i>A CMIP5 Model Selection Specific to South Africa's Winter Rainfall Zone</i>. (). ,Faculty of Science ,Department of Environmental and Geographical Science. Retrieved from http://hdl.handle.net/11427/37614 | en_ZA |
| dc.identifier.chicagocitation | Marsh, Peter. <i>"A CMIP5 Model Selection Specific to South Africa's Winter Rainfall Zone."</i> ., ,Faculty of Science ,Department of Environmental and Geographical Science, 2022. http://hdl.handle.net/11427/37614 | en_ZA |
| dc.identifier.citation | Marsh, P. 2022. A CMIP5 Model Selection Specific to South Africa's Winter Rainfall Zone. . ,Faculty of Science ,Department of Environmental and Geographical Science. http://hdl.handle.net/11427/37614 | en_ZA |
| dc.identifier.ris | TY - Master Thesis AU - Marsh, Peter AB - This study undertakes a CMIP5 model selection specific to the Winter Rainfall Zone (WRZ) of South Africa, seeking to reduce the range of future climate projections through identifying a subset of models with increased realism and independence. In order to navigate the subjectivity in identifying relevant circulation metrics to assess models against, the ‘Day Zero' drought is used as a characteristic episode. Here initially the extensive literature produced subsequent to the drought has been drawn on to identify and evaluate relevant regional process metrics, before utilising the anomalous conditions during the drought to validate various assessment methods. The dynamics subsequently identified as being most influential to rainfall supply in the WRZ include the South Atlantic subtropical jet stream responsible for steering of mid-latitude storm systems, the South Atlantic subtropical high, and the presence, or preferably absence, of precipitation blocking subsidence, and the prevalence of mid-latitude storm systems, critical for transport and upliftment of moisture to the region. Models were subsequently assessed against these metrics and scored following the technique of McSweeney et al. (2015). Unrealistic models were removed from the ensemble while significantly biased models were also excluded as their absence did not significantly reduce the range of future projections. The same scoring methods were then utilised to create a genealogy of models attaining similar results to that of Knutti, Masson & Gettelman (2013). A subset of 6 CMIP5 models which are more independent and historically more realistic than that of the full ensemble were subsequently identified. While the range of future temperature projections of the final ensemble are somewhat constrained in comparison to the full ensemble, the primary utility is argued to be the reduced number of models where a future researcher may consider each model's projected future climate pathway individually before selecting a model, or models, which best informs their use case, whilst being assured that this model performs suitably well in the region and that the initial ensemble considered adequately represents model uncertainty, while strong similarity between two or more models within the ensemble will not be unduly biasing results. DA - 2022_ DB - OpenUCT DP - University of Cape Town KW - Geographical Sciences LK - https://open.uct.ac.za PY - 2022 T1 - A CMIP5 Model Selection Specific to South Africa's Winter Rainfall Zone TI - A CMIP5 Model Selection Specific to South Africa's Winter Rainfall Zone UR - http://hdl.handle.net/11427/37614 ER - | en_ZA |
| dc.identifier.uri | http://hdl.handle.net/11427/37614 | |
| dc.identifier.vancouvercitation | Marsh P. A CMIP5 Model Selection Specific to South Africa's Winter Rainfall Zone. []. ,Faculty of Science ,Department of Environmental and Geographical Science, 2022 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/37614 | en_ZA |
| dc.language.rfc3066 | eng | |
| dc.publisher.department | Department of Environmental and Geographical Science | |
| dc.publisher.faculty | Faculty of Science | |
| dc.subject | Geographical Sciences | |
| dc.title | A CMIP5 Model Selection Specific to South Africa's Winter Rainfall Zone | |
| dc.type | Master Thesis | |
| dc.type.qualificationlevel | Masters | |
| dc.type.qualificationlevel | MSc |