Dynamical downscaling outputs

The conformal-cubic atmospheric model (CCAM) was used as the regional climate model, a variable-resolution global climate model (GCM) developed by the Common-wealth Scientific and Industrial Research Organisation (CSIRO) was used, Six GCM simulations of the Coupled Model Intercomparison Project Phase Five (CMIP5) and Assessment Report Five (AR5) of the Intergovernmental Panel on Climate Change (IPCC) for the emission scenarios described by Representative Concentration Pathway 8.5 (RCP 8.5) were used at a 8 km resolution. Future climate projections obtained at the 10th, 50th and 90th percentile. To compute the indices, we used policy-relevant climate projections from six GCMs that reflect levels of mitigation action from the Paris Agreement targets of 1.5 and 2°C and 3°C. The baseline period of 1661-1991 is used to calculate anomalies for each of the climate indices.

• Key baseline period message: Annual temperature for the baseline period (1961-1990) shows a spatial gradient with the annual average temperature being the highest in the Northeastern parts of the Lowveld and lowest Highveld escarpment and the Southwestern parts of the province.

• Future projection key messages: The multi-model projections of change in average annual temperature over the province are indicative of a warming over the Mpumalanga province. The general spatial temperature patterns, shared among a majority of the models suggest more warming relative to the baseline period in the southwestern and central parts of the province particularly in Gert Sibande and Nkangala Districts. This is in comparison to the eastern parts of the province which span the Ehlanzeni District for all the global warming levels under RCP8.5.

• Key baseline period message: The temperature swing between day and night during the baseline period relatively higher within the Gert Sibande and Nkangala regions compared to Ehlanzeni districts.

• Future projection key messages: The models projections suggest an increase in DTR during the 1.5, 2.0 and 3.0 °C global warming periods. The widening of the swing between night and day temperatures, under RCP8.5, needs further analysis in conjunction with driving factors such as relative humidity to further explain spatial patterns. Notably, the increase in DTR is consistent with that of RX5day index suggesting a potential link between the increase in DTR with moisture/humidity which is known to drive some of the precipitation extremes spatial patterns under climate change.

• Key baseline period message: The six model ensemble is consistent in reflecting annual precipitation as being highest over the Highveld escarpment area relative to the rest of the country during the baseline period (1961-1990).

• Future projection key messages: At different Global Warming Levels (GWLs), the model ensemble projections are suggestive of a decline in annual total precipitation (rnd24 index) in most parts of the Ehlanzeni district as well as the northern part of Gert Sibande district as global temperature reach 2°C, under RCP8.5. Comparing the 10th and 90th model ensemble percentiles, we note that there is a pronounced uncertainty in six model projections as indicated by the conflicting direction of change, relative to the baseline, in annual precipitation totals for the rest of the Mpumalanga province at the 1.5 and 3.0 °C global warming levels.

• Key baseline period message: The annual average of the number of CDD are relatively higher over the western part of the Mpumalanga Province (especially over the Northwestern part) during the baseline period (1961 -1990). The Highveld escarpment is reflected as having experienced the least CDD over the period.

• Future projection key messages: The models agree on a projected increase in CDD index, relative to the baseline, over a larger part of Nkangala District (Northwestern region) and Gert Sibande (Southwestern region) as global temperature reach 2 and 3°C Under RCP8.5. The highest count of CDDs is projected to occur over Nkangala district under the 3.0o global temperature warming scenario. The signal of change is less developed over Ehlanzeni as can be seen by conflicting signs of change between the 10 and 90th ensemble percentiles.

• Key baseline period message: The highest annual count of days with precipitation ≥ 20mm occurs over the Mpumalanga escarpment areas (i.e., central to north-eastern parts).

• Future projection key messages: Over the 1.5°C the model signal of change in R20mm index is less developed among different models and there is no model consensus in the direction of change between the 10th and 90th ensemble percentiles. As global temperature approach 2.0 °C the Highveld escarpment areas a projected to experience a decline in the count of days with very heavy precipitation. Towards the end of the century (i.e., 3.0 °C GWL) the model are consistent in suggesting an increase in the number of days with heavy precipitation for the central to Southwestern parts of the Gert Sibande district.

• Key baseline period message: The highest annual count of maximum number of CWDs occurs over the Mpumalanga escarpment areas (i.e., central to north-eastern parts).

• Future projection key messages: The models agree on a projected increase in the rx5day index over the northern parts of the Enhlanzeni District during the GWLs (1.5, 2.0 and 3.0o warming). As the projected gobal temperature reach a 3.0o warming, the models agree on an increase in rx5day over most of the Nkangala and Gert Sibande region as well. The 1.5 and 2.0 °C warming scenario projections for the province reflect a pronounced uncertainty among the ensemble members for most of the province.

• Key baseline period message: The highest annual count of maximum number of CWDs occurs over the Mpumalanga escarpment areas (i.e., central to north-eastern parts).

• Future projection key messages: The six model-ensemble suggest a decline in the annual maximum number of Consecutive Wet Days (CWD), relative to the baseline period, among the models over the high lying escarpment areas during the 1.5, 2.0 and 3.0°C global warming periods. For the rest of the province the change in CWD is inconclusive as there is conflicting signs in the direction of change .