1. © Yann Arthus-Bertrand / Altitude Observed climate trends and future projections for Africa Observed climate trends and future projections for Africa Dr. Joseph K. Kanyanga IPCC AR5 WG 1 Lead Author

2. © Yann Arthus-Bertrand / Altitude Summary: for Africa KEY MESSAGE 1 on Observed Climate trends over Africa It is evident that Africa’s climate has changed and physical impacts already felt: Temperature increase both over land and ocean surfaces – Each of the last three decades has been successively warmer at the Earth’s surface than any preceding decade since 1850. Increase in frequency and severity of extreme weather events especially during the last 30 to 60 years: floods and droughts. - At local level, this is manifest through sustained changes in onset, cessation and intensities.

3. © Yann Arthus-Bertrand / Altitude Summary: for Africa KEY MESSAGE 2 on Observed Climate trends over Africa It is extremely likely that human activities have been the dominant cause of the observed changes in climate (warming). - The climate changes corresponds to period of increase in GHG emissions (enhanced CO2 concentrations).

4. © Yann Arthus-Bertrand / Altitude Summary: for Africa KEY MESSAGE 3 on Future Climate Projections over Africa Further climate change is inevitable in the coming decades.

5. © Yann Arthus-Bertrand / Altitude Chapter 14: Climate Phenomena and their Relevance for Future Regional Climate Change Chapter 14: Climate Phenomena and their Relevance for Future Regional Climate Change

6. Future regional climate change Future changes in phenomena ASSESSMENTS Confidence level of CMIP5 models to represent the key processes in these climate phenomena Relevance of such phenomena for future change in regional climates CMIP5 regional climate projections Monsoon Systems Tropical Phenomena (CZ’s, MJO, IOD) Annular and Dipolar Modes Large-scale Storm Systems El Niño-Southern Oscillation This chapter assesses the scientific literature on projected changes in major climate phenomena contingent on global mean temperatures continue to rise and more specifically their relevance for future change in regional climates. Other Relevant Phenomena

7. Monsoons There is growing evidence of improved skill of climate models in reproducing climatological features of the global monsoon. Taken together with identified model agreement on future changes, the global monsoon is likely to strengthen in the 21st century with increases in its area and intensity, while the monsoon circulation weakens.

8. Monsoons Model skill in representing regional monsoons is lower compared to the global monsoon and varies across different monsoon systems.

9. Tropical phenomena: Convergence Zones Rainfall Change (medium confidence) “wet-get-wetter” over CZ regions “warmer-get-wetter” over oceans Figure 14.8: Upper panel: Annual-mean precipitation percentage change (  P/P in green/gray shade and white contours at 20% intervals), and relative SST change (colour contours at intervals of 0.2°C; negative shaded) to the tropical (20S–20N) mean warming in RCP8.5 projections, shown as 23 CMIP5 model ensemble mean. Figure 14.9: Seasonal cycle of zonal-mean tropical precipitation change (2081– 2100 in RCP8.5 minus 1986–2005) in CMIP5 multimodel ensemble mean. Eighteen CMIP5 models were used. Stippling indicates that more than 90% models agree on the sign of MME change. The red curve represents the meridional maximum of the climatological rainfall. Adapted from Huang et al. (2013). The seasonal-mean rainfall is projected to increase on the ITCZ equatorward flank More warming and rainfall at north of the equator. Less zonal SST gradient across the equatorial Pacific that contribute to the weakened Walker cells.

10. Tropical phenomena Madden-Julian Oscillation Low confidence in assessing future change due to the poor skill of models and its high sensitivity to SST warming pattern Atlantic Ocean Modes Low confidence in projected changes over the tropical Atlantic - both for the mean and interannual modes Indian Ocean Modes High confidence in projected changes: reduced (enhanced) warming and decreased (increased) rainfall in the east (west) Figure 14.10: September to November changes in a 22-model CMIP5 ensemble (2081–2100 in RCP8.5 minus 1986–2005 in historical run). (a) SST (colour contours at 0.1°C intervals) relative to the tropical mean (20°S–20°N), and precipitation (shading and white contours at 20 mm per month intervals). (b) Surface wind velocity (m s –1 ), and sea surface height deviation from the global mean (contours, cm). Over the equatorial Indian Ocean, ocean- atmospheric changes form Bjerknes feedback, with the reduced SST warming and suppressed convection in the east.

11. El Niño-Southern Oscillation Figure 14.14: Standard deviation of Niño3 SST anomalies from CMIP5 model experiments. PI, 20C, RCP4.5, RCP8.5 indicate pre-industrial control experiments, 20th century experiments, and 21st century experiment from the RCP4.5 and RCP8.5. Open dot and solid black line indicate multi-model ensemble mean and median, respectively, and the cross mark is 20th observation, respectively. Thick bar and thin outer bar indicate 50% and 75% percentile ranges, respectively. Figure 14.15: Changes to sea level pressure teleconnections during DJF in the CMIP5 models. (a) SLP anomalies for El Niño during the 20th century. (b) SLP anomalies for La Nina during the 20th century. (c) SLP anomalies for El Niño during RCP4.5. (d) SLP anomalies for La Niña during RCP4.5. Maps in (a)-(d) are stippled where more than 2/3 of models agree on the sign of the SLP anomaly ((a),( b): 18 models; (c),(d): 12 models), and hatched where differences between the RCP4.5 multi-model mean SLP anomaly exceed the 60th percentile (red- bordered regions) or are less than the 40th percentile (blue-bordered regions) of the distribution of 20th century ensemble means. In all panels, El Niño (La Niña) periods are defined as years having DJF NINO3.4 SST above (below) one standard deviation relative to the mean of the detrended time series. For ensemble-mean calculations, all SLP anomalies have been normalized to the standard deviation of the ensemble-member detrended NINO3.4 SST. Future changes in EN intensity in CMIP5 models are model-dependent, however ENSO will very likely remain the dominant mode of natural climate variability Medium confidence that ENSO-induced teleconnection patterns will shift eastward over the North Pacific and North America, and low confidence in changes in the intensity and spatial pattern of El Niño in a warmer climate.

12. Annular and Dipolar Modes Figure 14.16: Summary of multi-model ensemble simulations of wintertime (Dec-Feb) mean NAO, NAM and SAM sea-level pressure indices for historical and RCP4.5 scenarios produced by 39 climate models participating in CMIP5. Panels a-c) show time series of the ensemble mean (black line) and inter-quartile range (grey shading) of the mean index for each model. NAO/NAM is likely to become slightly more positive (on average) SAM positive trend is likely to weaken as ozone depletion recovers Medium confidence that projected changes in NAO and SAM are sensitive to boundary processes (stratosphere-troposphere interaction, ozone chemistry, response to Arctic sea ice loss), which are not yet well represented in many climate models

13. Large-Scale Storm Systems : Tropical Cyclones Figure 14.17: General consensus assessment of the numerical experiments described in Supplementary Material Tables 14.SM.1–14.SM.4 in the. All values represent expected percent change in the average over period 2081–2100 relative to 2000–2019, under an A1B-like scenario, based on expert judgment after subjective normalisation of the model projections. Four metrics were considered: the percent change in I) the total annual frequency of tropical storms, II) the annual frequency of Category 4 and 5 storms, III) the mean Lifetime Maximum Intensity (LMI; the maximum intensity achieved during a storm’s lifetime), and IV) the precipitation rate within 200 km of storm centre at the time of LMI. For each metric plotted, the solid blue line is the best guess of the expected percent change, and the coloured bar provides the 67% (likely) confidence interval for this value (note that this interval ranges across – 100% to +200% for the annual frequency of Category 4 and 5 storms in the North Atlantic). Where a metric is not plotted, there is insufficient data (denoted "insf. d.") available to complete an assessment. A randomly drawn (and coloured) selection of historical storm tracks are underlain to identify regions of tropical cyclone activity.  The global frequency will likely either decrease or remain essentially unchanged  Both global mean tropical cyclone maximum wind speed and rainfall rates will likely increase, but there is low confidence in region-specific projections  The frequency of the most intense will more likely than not increase substantially in some basins

14. Large-Scale Storm Systems : Extra-tropical Cyclones High confidence that the global number of extra-tropical cyclones will be unlikely to decrease by more than a few percent High confidence that a small poleward shift will be likely in the Southern Hemisphere storm-track, but the magnitude is model-dependent Medium confidence that a poleward shift in the North Pacific storm-track will be more likely than not Medium confidence that it is unlikely that the response of the North Atlantic storm-track will be a simple poleward shift Low confidence in the magnitude of regional storm-track changes and their impact on regional surface climate Despite systematic biases in simulating storm-tracks, most models agree on the future changes in the number of extra-tropical cyclones.

15. Relevance of projected changes in major phenomena for mean change in future regional climate

16. Explanations on the entries of relevance table The level of relevance is assigned onto a combination of the confidence (high, medium, low) that there will be a non-zero change in the phenomena induced by anthropogenic forcing, and the confidence (high, medium, low) that any phenomenon impacts on temperature and precipitation in each region. The relevance is then classified into high (red), medium (yellow), low (blue), and no obvious relevance (grey) following the criteria: None of the statements on confidence levels include anything about the magnitude of phenomena changes or regional climate impacts, but merely the assessment of the literature (both model projections and observations) on whether or not there will be an effect.

21. KEY MESSAGE 4 on Observed Trends & Future Climate Projections over Africa Versus Development Climate change poses a challenge for growth and development in Africa.

22. © Yann Arthus-Bertrand / Altitude www.climatechange2013.org Further Information