| Resumo: | Climate changes refer to any substantial changes in measures of climate lasting for an extended period. One the important ways to track and investigate the causes and effects of climate change is through the use of indicators, like sea surface temperature (SST). Thus, the main aim of this dissertation is to understand the impacts of climate changes in future SST variability with analysis of SST data from global climate models (GCMs) of CMIP5 project. The methodology adopted comprised two fundamental steps: 1) worldwide SST data division in regions with K-means cluster and validation of CMIP5 models with a comparative analysis between SST data of CGMs of CMIP5 and SST data of Era Interim reanalysis; 2) calculate differences between future (2070-2100 and 2020-2050) and historical (1975-2005) SST data for RCP4.5 and RCP 8.5 climatic scenarios, and compute trends between 1975 and 2100 for both RCP scenarios. SST data of CGMs of CMIP5 are divided into eight regions. Most of GCMs of CMIP5 have a good SST reproducibility. North Hemisphere regions present a higher thermal amplitude when compared with equivalent regions of South Hemisphere. The differences between long-term future and historical regime are larger than differences between near-term future and historical regime. For long-term future, worldwide SST has mean increments of 2.46 oC and 1.35 oC for RCP 8.5 and RCP 4.5 scenarios, respectively. For near-term future, SST has mean increments of 0.86 oC and 0.73 oC for RCP 8.5 and RCP 4.5 scenarios, respectively. Polar Regions present higher percentual variability, being possible to conclude that Polar Regions will have larger changes, reflecting an overheating on these regions. For RCP 8.5 scenario, SST trends are 4.34 oC/dec (maximum) and 1.64 oC/dec (minimum) for north sub-tropical region (STRN) and south polar region (PRS), respectively. Regarding RCP 4.5 scenario, SST trends are 2.64 oC/dec (maximum) and 0.94 oC/dec (minimum) for STRN and PRS, respectively. From this work it can be concluded that is not possible to select only one global model, since each has its advantages. North Hemisphere will present higher SST trends. Overall, this thesis analysis has revealed that SST will increase worldwide, however, this depends on the different globe locations
|
|---|