Academia.eduAcademia.edu

Figure 7 – uploaded by G. Marshall

See full PDFdownloadDownload figure
Table 14.1 (continued)

Table 14 1 (continued)

Related Figures (43)

Box 14.1, Table 1 | Regional climate impacts of fundamental modes of variability.
Box 14.1, Table 1 | Regional climate impacts of fundamental modes of variability.
Box 14.1, Figure 1| Global distribution of average annual rainfall (in cm/year) from 1979-2010 Global Precipitation Climatology Project (GPCP) database, monsoon precipitation domain (white contours) as defined in Section 14.2.1, and winter storm-tracks in both hemispheres (black arrows). In left (right) column seasonal cor- relation maps of North Atlantic Oscillation (NAO), Southern Oscillation Index (SOI, the atmospheric component of El Nifo-Southern Oscillation (ENSO)) and Southern Annular Mode (SAM) mode indexes vs. monthly temperature (precipitation) anomalies in boreal winter (December, January and February (DJF)), austral winter (June, July and August (JJA)) and austral spring (September, October and November (SON)). Black contours indicate a 99% significance level. The mode indices were taken from National Oceanic and Atmospheric Administration (NOAA, http://www.esrl.noaa.gov/psd/data/climateindices/list/), global temperatures from NASA Goddard Institute of Space Studies Surface Temperature Analysis (GISTEMP, http://data.giss.nasa.gov/gistemp/) and global precipitations from GPCP (http://www.esrl.noaa.gov/psd/data/ gridded/data.gpcp.html). A set of similar states of the climate system that occur more frequently than nearby states due to either more persistence or more often recurrence. In other words, a cluster in climate state space associated with a local maximum in the probability density function.
Box 14.1, Figure 1| Global distribution of average annual rainfall (in cm/year) from 1979-2010 Global Precipitation Climatology Project (GPCP) database, monsoon precipitation domain (white contours) as defined in Section 14.2.1, and winter storm-tracks in both hemispheres (black arrows). In left (right) column seasonal cor- relation maps of North Atlantic Oscillation (NAO), Southern Oscillation Index (SOI, the atmospheric component of El Nifo-Southern Oscillation (ENSO)) and Southern Annular Mode (SAM) mode indexes vs. monthly temperature (precipitation) anomalies in boreal winter (December, January and February (DJF)), austral winter (June, July and August (JJA)) and austral spring (September, October and November (SON)). Black contours indicate a 99% significance level. The mode indices were taken from National Oceanic and Atmospheric Administration (NOAA, http://www.esrl.noaa.gov/psd/data/climateindices/list/), global temperatures from NASA Goddard Institute of Space Studies Surface Temperature Analysis (GISTEMP, http://data.giss.nasa.gov/gistemp/) and global precipitations from GPCP (http://www.esrl.noaa.gov/psd/data/ gridded/data.gpcp.html). A set of similar states of the climate system that occur more frequently than nearby states due to either more persistence or more often recurrence. In other words, a cluster in climate state space associated with a local maximum in the probability density function.
Figure 14.1 | (Upper) Observed (thick contour) and simulated (shading) global monsoon domain, based on the definition of Wang et al. (2011). The observations are based on GPCP v2.2 data (Huffman et al., 2009), and the simulations are based on 26 CMIP5 multi-model mean precipitation with a common 2.5 by 2.5 degree grid in the present day (1986-2005) and the future (2080-2099; RCP8.5 scenario). Orange (dark blue) shading shows monsoon domain only in the present day (future). Light blue shading shows monsoon domain in both periods. (Lower) Projected changes for the future (2080-2099) relative to the present day (1986-2005) in the global monsoon area (GMA) and global monsoon intensity (GMI), global monsoon total precipitation (GMP), standard deviation of interannual variability in seasonal average precipitation (Psd), simple daily precipitatior intensity index (SDII), seasonal maximum 5-day precipitation total (R5d), seasonal maximum consecutive dry days (CDD) and monsoon season duration (DUR), under the RCP2.€ (dark blue; 18 models), RCP4.5 (light blue; 24 models), RCP6.0 (orange; 14 models) and RCP8.5 scenarios (red; 26 models). Units are % except for DUR (days). Box-and-whisker plots show the 10th, 25th, 50th, 75th and 90th percentiles. All of the indices are calculated for the summer season (May to September in the Northern Hemisphere; November tc March in the Southern Hemisphere). The indices of Psd, SDII, Rod and CDD calculated for each model's original grid, and then averaged over the monsoon domains determined by each model at the present-day. The indices of DUR are calculated for seven regional monsoon domains based on the criteria proposed by Wang and LinHo (2002) using regionally averaged climatological cycles of precipitation, and then their changes are averaged with weighting based on their area at the present day. CMIP5 models show a decreasing trend of lower-troposphere wind convergence (dynamical factor) throughout the 20th and 21st centu- ries (Figure 14.2d). With increased moisture (see also Section 12.4), the moisture flux convergence shows an increasing trend from 1980 through the 21st century (Figure 14.2c). Surface evaporation shows a similar trend (Figure 14.2b) associated with warmer SSTs. There- fore, the global monsoon precipitation increases (Figure 14.2a) due to increases in moisture flux convergence and surface evaporation despite In the CMIP5 models the global monsoon area (GMA), the global monsoon total precipitation (GMP) and the global monsoon precipi- tation intensity (GMI) are projected to increase by the end of the 21st century (2081-2100, Hsu et al., 2013; Kitoh et al., 2013; Figure 14.1). See Supplementary Material Section 14.SM.1.2 for the definitions of GMA, GMP and GMI. The CMIP5 model projections show an expan- sion of GMA mainly over the central to eastern tropical Pacific, the southern Indian Ocean and eastern Asia. In all RCP scenarios, GMA is very likely to increase, and GMI is /ikely to increase, resulting in a very likely increase in GMP, by the end of the 21st century (2081-2100). The 100-year median changes in GMP are +5%, +8%, +10%, and +16% in RCP2.6, RCP4.5, RCP6.0, and RCP8.5 scenarios, respectively. Indices of precipitation extremes such as simple daily precipitation intensity index (SDII), defined as the total precipitation divided by the number of days with precipitation greater than or equal to 1 mm, annual max-
Figure 14.1 | (Upper) Observed (thick contour) and simulated (shading) global monsoon domain, based on the definition of Wang et al. (2011). The observations are based on GPCP v2.2 data (Huffman et al., 2009), and the simulations are based on 26 CMIP5 multi-model mean precipitation with a common 2.5 by 2.5 degree grid in the present day (1986-2005) and the future (2080-2099; RCP8.5 scenario). Orange (dark blue) shading shows monsoon domain only in the present day (future). Light blue shading shows monsoon domain in both periods. (Lower) Projected changes for the future (2080-2099) relative to the present day (1986-2005) in the global monsoon area (GMA) and global monsoon intensity (GMI), global monsoon total precipitation (GMP), standard deviation of interannual variability in seasonal average precipitation (Psd), simple daily precipitatior intensity index (SDII), seasonal maximum 5-day precipitation total (R5d), seasonal maximum consecutive dry days (CDD) and monsoon season duration (DUR), under the RCP2.€ (dark blue; 18 models), RCP4.5 (light blue; 24 models), RCP6.0 (orange; 14 models) and RCP8.5 scenarios (red; 26 models). Units are % except for DUR (days). Box-and-whisker plots show the 10th, 25th, 50th, 75th and 90th percentiles. All of the indices are calculated for the summer season (May to September in the Northern Hemisphere; November tc March in the Southern Hemisphere). The indices of Psd, SDII, Rod and CDD calculated for each model's original grid, and then averaged over the monsoon domains determined by each model at the present-day. The indices of DUR are calculated for seven regional monsoon domains based on the criteria proposed by Wang and LinHo (2002) using regionally averaged climatological cycles of precipitation, and then their changes are averaged with weighting based on their area at the present day. CMIP5 models show a decreasing trend of lower-troposphere wind convergence (dynamical factor) throughout the 20th and 21st centu- ries (Figure 14.2d). With increased moisture (see also Section 12.4), the moisture flux convergence shows an increasing trend from 1980 through the 21st century (Figure 14.2c). Surface evaporation shows a similar trend (Figure 14.2b) associated with warmer SSTs. There- fore, the global monsoon precipitation increases (Figure 14.2a) due to increases in moisture flux convergence and surface evaporation despite In the CMIP5 models the global monsoon area (GMA), the global monsoon total precipitation (GMP) and the global monsoon precipi- tation intensity (GMI) are projected to increase by the end of the 21st century (2081-2100, Hsu et al., 2013; Kitoh et al., 2013; Figure 14.1). See Supplementary Material Section 14.SM.1.2 for the definitions of GMA, GMP and GMI. The CMIP5 model projections show an expan- sion of GMA mainly over the central to eastern tropical Pacific, the southern Indian Ocean and eastern Asia. In all RCP scenarios, GMA is very likely to increase, and GMI is /ikely to increase, resulting in a very likely increase in GMP, by the end of the 21st century (2081-2100). The 100-year median changes in GMP are +5%, +8%, +10%, and +16% in RCP2.6, RCP4.5, RCP6.0, and RCP8.5 scenarios, respectively. Indices of precipitation extremes such as simple daily precipitation intensity index (SDII), defined as the total precipitation divided by the number of days with precipitation greater than or equal to 1 mm, annual max-
Figure 14.2 | Time series of simulated anomalies, smoothed with a 20-year running mean over the global land monsoon domain for (a) precipitation (mm day~'), (b) evaporatic (mm day-"), (c) water vapour flux convergence in the lower (below 500 hPa) troposphere (mm day-'), and (d) wind convergence in the lower troposphere (10-3 kg m~? s~’), relati\ to the present-day (1986-2005), based on CMIP5 multi-model monthly outputs. Historical (grey; 29 models), RCP2.6 (dark blue; 20 models), RCP4.5 (light blue; 24 models), RCP6 (orange; 16 models), and RCP8.5 (red; 24 models) simulations are shown in the 10th and 90th percentile (shading), and in all model averages (thick lines).
Figure 14.2 | Time series of simulated anomalies, smoothed with a 20-year running mean over the global land monsoon domain for (a) precipitation (mm day~'), (b) evaporatic (mm day-"), (c) water vapour flux convergence in the lower (below 500 hPa) troposphere (mm day-'), and (d) wind convergence in the lower troposphere (10-3 kg m~? s~’), relati\ to the present-day (1986-2005), based on CMIP5 multi-model monthly outputs. Historical (grey; 29 models), RCP2.6 (dark blue; 20 models), RCP4.5 (light blue; 24 models), RCP6 (orange; 16 models), and RCP8.5 (red; 24 models) simulations are shown in the 10th and 90th percentile (shading), and in all model averages (thick lines).
FAQ 14.1, Figure 1 | Schematic diagram illustrating the main ways that human activity influences monsoon rainfall. As the climate warms, increasing water vapour transport from the ocean into land increases because warmer air contains more water vapour. This also increases the potential for heavy rainfalls. Warming-related changes in large-scale circulation influence the strength and extent of the overall monsoon circulation. Land use change and atmospheric aerosol loading can also affect the amount of solar radiation that is absorbed in the atmosphere and land, potentially moderating the land—sea temperature difference. The land surface warms more rapidly than the ocean surface, so that surface temperature contrast is increasing in most regions. The tropical atmospheric overturning circulation, however, slows down on average as the climate warms due to energy balance constraints in the tropical atmosphere. These changes in the atmospheric circulation lead to regional changes in monsoon intensity, area and timing. There are a number of other effects as to how
FAQ 14.1, Figure 1 | Schematic diagram illustrating the main ways that human activity influences monsoon rainfall. As the climate warms, increasing water vapour transport from the ocean into land increases because warmer air contains more water vapour. This also increases the potential for heavy rainfalls. Warming-related changes in large-scale circulation influence the strength and extent of the overall monsoon circulation. Land use change and atmospheric aerosol loading can also affect the amount of solar radiation that is absorbed in the atmosphere and land, potentially moderating the land—sea temperature difference. The land surface warms more rapidly than the ocean surface, so that surface temperature contrast is increasing in most regions. The tropical atmospheric overturning circulation, however, slows down on average as the climate warms due to energy balance constraints in the tropical atmosphere. These changes in the atmospheric circulation lead to regional changes in monsoon intensity, area and timing. There are a number of other effects as to how
Figure 14.3 | Regional land monsoon domain based on 26 CMIP5 multi-model mean precipitation with a common 2.5° x 2.5° grid in the present-day (1986-2005). For regione divisions, the equator separates the northern monsoon domains (North America Monsoon System (NAMS), North Africa (NAF), Southern Asia (SAS) and East Asian summer (EAS) from the southern monsoon domains (South America Monsoon System (SAMS), South Africa (SAF), and Australian-Maritime Continent (AUSMC)), 60°E separates NAF from SAS and 20°N and 100°E separates SAS from EAS. All the regional domains are within 40°S to 40°N. The Indian summer monsoon is known to have undergone abrupt shifts in the past millennium, giving rise to prolonged and intense droughts (Meehl and Hu, 2006; Sinha et al., 2011; see also Chapter 2). The observed recent weakening tendency in seasonal rainfall and the regional re-distribution has been partially attributed to factors such as changes in black carbon and/or sulphate aerosols (Chung and CMIP3 projections show suppressed rainfall over the equatorial Indian Ocean (Cai et al., 2011¢; Turner and Annamalai, 2012), and an increase in seasonal mean rainfall over India (Ueda et al., 2006; Annamalai
Figure 14.3 | Regional land monsoon domain based on 26 CMIP5 multi-model mean precipitation with a common 2.5° x 2.5° grid in the present-day (1986-2005). For regione divisions, the equator separates the northern monsoon domains (North America Monsoon System (NAMS), North Africa (NAF), Southern Asia (SAS) and East Asian summer (EAS) from the southern monsoon domains (South America Monsoon System (SAMS), South Africa (SAF), and Australian-Maritime Continent (AUSMC)), 60°E separates NAF from SAS and 20°N and 100°E separates SAS from EAS. All the regional domains are within 40°S to 40°N. The Indian summer monsoon is known to have undergone abrupt shifts in the past millennium, giving rise to prolonged and intense droughts (Meehl and Hu, 2006; Sinha et al., 2011; see also Chapter 2). The observed recent weakening tendency in seasonal rainfall and the regional re-distribution has been partially attributed to factors such as changes in black carbon and/or sulphate aerosols (Chung and CMIP3 projections show suppressed rainfall over the equatorial Indian Ocean (Cai et al., 2011¢; Turner and Annamalai, 2012), and an increase in seasonal mean rainfall over India (Ueda et al., 2006; Annamalai
Figure 14.5 | Time series of summer monsoon indices (21-year running mean) relative to the base period average (1986-2005). Historical (gray), RCP4.5 (light blue) and RCP8. (red) simulations by 39 CMIP5 model ensembles are shown in 10th and 90th (shading), and 50th (thick line) percentiles. (a) East Asian summer monsoon (defined as June, July an August (JJA) sea level pressure difference between 160°E and 110°E from 10°N to 50°N), (b) Indian summer monsoon (defined as meridional differences of the JA 850 hPa zon winds averaged over 5°N to 15°N, 40°E to 80°E and 20°N to 30°N, 60°E to 90°E), (c) western North Pacific summer monsoon (defined as meridional differences of the JJA 85 hPa zonal winds averaged over 5°N to 15°N, 100°E to 130°E and 20°N to 30°N, 110°E to 140°), (d) Australian summer monsoon (defined as December, January and Februar (DJF) 850 hPa zonal wind anomalies averaged over 10°S to 0°, 120°E to 150°E). (See Wang et al. (2004) and Zhou et al. (2009c) for indices definitions.)
Figure 14.5 | Time series of summer monsoon indices (21-year running mean) relative to the base period average (1986-2005). Historical (gray), RCP4.5 (light blue) and RCP8. (red) simulations by 39 CMIP5 model ensembles are shown in 10th and 90th (shading), and 50th (thick line) percentiles. (a) East Asian summer monsoon (defined as June, July an August (JJA) sea level pressure difference between 160°E and 110°E from 10°N to 50°N), (b) Indian summer monsoon (defined as meridional differences of the JA 850 hPa zon winds averaged over 5°N to 15°N, 40°E to 80°E and 20°N to 30°N, 60°E to 90°E), (c) western North Pacific summer monsoon (defined as meridional differences of the JJA 85 hPa zonal winds averaged over 5°N to 15°N, 100°E to 130°E and 20°N to 30°N, 110°E to 140°), (d) Australian summer monsoon (defined as December, January and Februar (DJF) 850 hPa zonal wind anomalies averaged over 10°S to 0°, 120°E to 150°E). (See Wang et al. (2004) and Zhou et al. (2009c) for indices definitions.)
Figure 14.6 | As in Figure 14.4, except for (upper) North America Monsoon System (NAMS) and (lower) South America Monsoon System (SAMS,
Figure 14.6 | As in Figure 14.4, except for (upper) North America Monsoon System (NAMS) and (lower) South America Monsoon System (SAMS,
Figure 14.7 | As in Figure 14.4, except for (upper) North Africa (NAF) and (lower) South Africa (SAF).
Figure 14.7 | As in Figure 14.4, except for (upper) North Africa (NAF) and (lower) South Africa (SAF).
Figure 14.8 | (Upper panel) Annual mean precipitation percentage change (SP/P in green/gray shade and white contours at 20% intervals), and relative SST change (colour contours at intervals of 0.2°C; negative dashed) to the tropical (20°S to 20°N) mean warming in RCP8.5 projections, shown as 23 CMIP5 model ensemble mean. (Lower panel) Sea surface temperature (SST) warming pattern indices in the 23-model RCP8.5 ensemble, shown as the 2081-2100 minus 1986-2005 difference. From left: Northern (EQ to 60°N) minus Southern (60°S to EQ) Hemisphere; equatorial (120°E to 60°W, 5°S to 5°N) and Southeast (130°W to 70°W, 30°S to 15°S) Pacific relative to the tropical mean warming; zonal SST gradient in the equatorial Pacific (120°E to 180°E minus 150°W to 90°W, 5°S to 5°N) and Indian (50°E to 70°E, 10°S to 10°N minus 90°E to 110°S, 10°S to EQ) Oceans. (Rightmost) Spatial correlation (r) between relative SST change and precipitation percentage change (8P/P) in the tropics (20°S to 20°N) in each model. (The spatial correlation for the multi-model ensemble mean fields in the upper panel is 0.63). The circle and error bar indicate the ensemble mean and +1 standard deviation, respectively. The upper panel is a CMIP5 update of Ma and Xie (2013), and see text for indices in the lower panel. The Inter-Tropical Convergence Zone (ITCZ) is a zonal band of persis: tent low-level convergence, atmospheric convection, and heavy rainfall Over the Atlantic and eastern half of the Pacific, the ITCZ is displacec north of the equator due to ocean—atmosphere interaction (Xie et al. The South Pacific Convergence Zone (SPCZ, Widlansky et al., 2011) extends southeastward from the tropical western Pacific to French Pol- ynesia and the SH mid-latitudes, contributing most of the yearly rainfall
Figure 14.8 | (Upper panel) Annual mean precipitation percentage change (SP/P in green/gray shade and white contours at 20% intervals), and relative SST change (colour contours at intervals of 0.2°C; negative dashed) to the tropical (20°S to 20°N) mean warming in RCP8.5 projections, shown as 23 CMIP5 model ensemble mean. (Lower panel) Sea surface temperature (SST) warming pattern indices in the 23-model RCP8.5 ensemble, shown as the 2081-2100 minus 1986-2005 difference. From left: Northern (EQ to 60°N) minus Southern (60°S to EQ) Hemisphere; equatorial (120°E to 60°W, 5°S to 5°N) and Southeast (130°W to 70°W, 30°S to 15°S) Pacific relative to the tropical mean warming; zonal SST gradient in the equatorial Pacific (120°E to 180°E minus 150°W to 90°W, 5°S to 5°N) and Indian (50°E to 70°E, 10°S to 10°N minus 90°E to 110°S, 10°S to EQ) Oceans. (Rightmost) Spatial correlation (r) between relative SST change and precipitation percentage change (8P/P) in the tropics (20°S to 20°N) in each model. (The spatial correlation for the multi-model ensemble mean fields in the upper panel is 0.63). The circle and error bar indicate the ensemble mean and +1 standard deviation, respectively. The upper panel is a CMIP5 update of Ma and Xie (2013), and see text for indices in the lower panel. The Inter-Tropical Convergence Zone (ITCZ) is a zonal band of persis: tent low-level convergence, atmospheric convection, and heavy rainfall Over the Atlantic and eastern half of the Pacific, the ITCZ is displacec north of the equator due to ocean—atmosphere interaction (Xie et al. The South Pacific Convergence Zone (SPCZ, Widlansky et al., 2011) extends southeastward from the tropical western Pacific to French Pol- ynesia and the SH mid-latitudes, contributing most of the yearly rainfall
Figure 14.9 | Seasonal cycle of zonal mean tropical precipitation change (2081-2100 in RCP8.5 minus 1986-2005) in CMIP5 multi-model ensemble (MME) 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 climato- logical rainfall. (Adapted from Huang et al., 2013.)
Figure 14.9 | Seasonal cycle of zonal mean tropical precipitation change (2081-2100 in RCP8.5 minus 1986-2005) in CMIP5 multi-model ensemble (MME) 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 climato- logical rainfall. (Adapted from Huang et al., 2013.)
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) Sea surface temperature (SST colour contours at 0.1°C intervals) relative to the tropical mean (20°S to 20°N), and precipitation (shading and white contours at 20 mm per month intervals). (b) Surface wind velocity (ms~'), and sea surface height deviation from the global mean (contours, centimetres). Over the equatorial Indian Ocean, ocean—atmospheric changes form Bjerknes feedback, with the reduced SST warming and suppressed convection in the east. (Updated with CMIP5 from Xie et al., 2010b.) Basin-mean SST has risen steadily for much of the 20th century, a trend captured by CMIP3 20th century simulations (Alory et al., 2007). The SST increase over the North Indian Ocean since about 1930 is notice- ably weaker than for the rest of the basin. This spatial pattern is sug- gestive of the effects of reduced surface solar radiation due to Asian brown clouds (Chung and Ramanathan, 2006) and it affects Arabian The IOD variability in SST remains nearly unchanged in future projec- tions of CMIP3 and CMIP5 (Ihara et al., 2009; Figure 14.11a) despite the easterly wind change that lifts the thermocline (Figure 14.10b) and intensifies thermocline feedback on SST in the eastern equatorial Indian Ocean. The global increase in atmospheric dry static stability weakens the atmospheric response to zonal SST gradient changes,
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) Sea surface temperature (SST colour contours at 0.1°C intervals) relative to the tropical mean (20°S to 20°N), and precipitation (shading and white contours at 20 mm per month intervals). (b) Surface wind velocity (ms~'), and sea surface height deviation from the global mean (contours, centimetres). Over the equatorial Indian Ocean, ocean—atmospheric changes form Bjerknes feedback, with the reduced SST warming and suppressed convection in the east. (Updated with CMIP5 from Xie et al., 2010b.) Basin-mean SST has risen steadily for much of the 20th century, a trend captured by CMIP3 20th century simulations (Alory et al., 2007). The SST increase over the North Indian Ocean since about 1930 is notice- ably weaker than for the rest of the basin. This spatial pattern is sug- gestive of the effects of reduced surface solar radiation due to Asian brown clouds (Chung and Ramanathan, 2006) and it affects Arabian The IOD variability in SST remains nearly unchanged in future projec- tions of CMIP3 and CMIP5 (Ihara et al., 2009; Figure 14.11a) despite the easterly wind change that lifts the thermocline (Figure 14.10b) and intensifies thermocline feedback on SST in the eastern equatorial Indian Ocean. The global increase in atmospheric dry static stability weakens the atmospheric response to zonal SST gradient changes,
Figure 14.11 | CMIP5 multi-model ensemble mean standard deviations of interannual variability for September to November in pre-industrial (PiControl; blue bars) and RCP8.5 (red) runs: (a) the Indian Ocean dipole index defined as the western (50°E to 70°E, 10°S to 10°N) minus eastern (90°E to 110°E, 10°S to 0°) SST difference; (b) zonal wind in the central equatorial Indian Ocean (70°E to 90°E, 5°S to 5°N); and (c) sea surface height in the eastern equatorial Indian Ocean (90°E to 110°E, 10°S to 0°). The standard deviation is normalized by the pre-industrial (PiControl) value for each model before ensemble average. Blue box-and-whisker plots show the 10th, 25th, 50th, 75th and 90th percentiles of 51-year windows for PiControl, representing natural variability. Red box-and-whisker plots represent inter-model variability for RCP8.5, based on the nearest rank. (Adapted from Zheng et al., 2013.)
Figure 14.11 | CMIP5 multi-model ensemble mean standard deviations of interannual variability for September to November in pre-industrial (PiControl; blue bars) and RCP8.5 (red) runs: (a) the Indian Ocean dipole index defined as the western (50°E to 70°E, 10°S to 10°N) minus eastern (90°E to 110°E, 10°S to 0°) SST difference; (b) zonal wind in the central equatorial Indian Ocean (70°E to 90°E, 5°S to 5°N); and (c) sea surface height in the eastern equatorial Indian Ocean (90°E to 110°E, 10°S to 0°). The standard deviation is normalized by the pre-industrial (PiControl) value for each model before ensemble average. Blue box-and-whisker plots show the 10th, 25th, 50th, 75th and 90th percentiles of 51-year windows for PiControl, representing natural variability. Red box-and-whisker plots represent inter-model variability for RCP8.5, based on the nearest rank. (Adapted from Zheng et al., 2013.)
Figure 14.12 | Idealized schematic showing atmospheric and oceanic conditions of the tropical Pacific region and their interactions during normal conditions, El Nifo condi- tions, and in a warmer world. (a) Mean climate conditions in the tropical Pacific, indicating sea surface temperatures (SSTs), surface wind stress and associated Walker Circulation, the mean position of convection and the mean upwelling and position of the thermocline. (b) Typical conditions during an EI Nifio event. SSTs are anomalously warm in the east; convection moves into the central Pacific; the trade winds weaken in the east and the Walker Circulation is disrupted; the thermocline flattens and the upwelling is reduced. (c) The likely mean conditions under climate change derived from observations, theory and coupled General Circulation Models (GCMs). The trade winds weaken; the thermocline flattens and shoals; the upwelling is reduced although the mean vertical temperature gradient is increased; and SSTs (shown as anomalies with respect to the mean tropical-wide warm- ing) increase more on the equator than off. Diagrams with absolute SST fields are shown on the left, diagrams with SST anomalies are shown on the right. For the climate change fields, anomalies are expressed with respect to the basin average temperature change so that blue colours indicate a warming smaller than the basin mean, not a cooling (Collins et al., 2010).
Figure 14.12 | Idealized schematic showing atmospheric and oceanic conditions of the tropical Pacific region and their interactions during normal conditions, El Nifo condi- tions, and in a warmer world. (a) Mean climate conditions in the tropical Pacific, indicating sea surface temperatures (SSTs), surface wind stress and associated Walker Circulation, the mean position of convection and the mean upwelling and position of the thermocline. (b) Typical conditions during an EI Nifio event. SSTs are anomalously warm in the east; convection moves into the central Pacific; the trade winds weaken in the east and the Walker Circulation is disrupted; the thermocline flattens and the upwelling is reduced. (c) The likely mean conditions under climate change derived from observations, theory and coupled General Circulation Models (GCMs). The trade winds weaken; the thermocline flattens and shoals; the upwelling is reduced although the mean vertical temperature gradient is increased; and SSTs (shown as anomalies with respect to the mean tropical-wide warm- ing) increase more on the equator than off. Diagrams with absolute SST fields are shown on the left, diagrams with SST anomalies are shown on the right. For the climate change fields, anomalies are expressed with respect to the basin average temperature change so that blue colours indicate a warming smaller than the basin mean, not a cooling (Collins et al., 2010).
Figure 14.13 | Intensities of El Nifio and La Nifa events for the last 60 years in the eastern equatorial Pacific (Nifio3 region) and in the central equatorial Pacific (Nifio4 region), and the estimated linear trends, obtained from Extended Reconstructed Sea Surface Temperature v3 (ERSSTv3).
Figure 14.13 | Intensities of El Nifio and La Nifa events for the last 60 years in the eastern equatorial Pacific (Nifio3 region) and in the central equatorial Pacific (Nifio4 region), and the estimated linear trends, obtained from Extended Reconstructed Sea Surface Temperature v3 (ERSSTv3).
Figure 14.14 | Standard deviation in CMIP5 multi-model ensembles of sea surface temperature variability over the eastern equatorial Pacific Ocean (Nino3 region: 5°S- 5°N, 150°W-90°W), a measure of El Nino amplitude, for the pre-industrial (PI) control and 20th century (20C) simulations, and 21st century projections using RCP4.5 and RCP8.5. Thirty-one models are used for the ensemble average. Open circles indicate multi-model ensemble means, and the red cross symbol is the observed standard devia- tion for January 1870 — December 2011 obtained from HadlSSTv1. The linear trend and climatological mean of seasonal cycle have been removed. Box-whisker plots show the 16th, 25th, 50th, 75th, and 84th percentiles. not significantly distinguished from natural modulations (Stevenson, 2012; Figure 14.14). Because the change in tropical mean conditions (especially the zonal gradient) in a warming climate is model depend- ent (Section 14.4.1), changes in ENSO intensity for the 21st century (Solomon and Newman, 2011; Hu et al., 2012a) are uncertain (Figure 14.14). Future changes in ENSO depend on competing changes in cou- pled ocean—atmospheric feedback (Philip and Van Oldenborgh, 2006; Collins et al., 2010; Vecchi and Wittenberg, 2010), and on the dynami- cal regime a given model is in. There is high confidence, however, that ENSO will remain the dominant mode of natural climate variability in the 21st century (Collins et al., 2010; Guilyardi et al. 2012; Kim and Yu 2012; Stevenson 2012).
Figure 14.14 | Standard deviation in CMIP5 multi-model ensembles of sea surface temperature variability over the eastern equatorial Pacific Ocean (Nino3 region: 5°S- 5°N, 150°W-90°W), a measure of El Nino amplitude, for the pre-industrial (PI) control and 20th century (20C) simulations, and 21st century projections using RCP4.5 and RCP8.5. Thirty-one models are used for the ensemble average. Open circles indicate multi-model ensemble means, and the red cross symbol is the observed standard devia- tion for January 1870 — December 2011 obtained from HadlSSTv1. The linear trend and climatological mean of seasonal cycle have been removed. Box-whisker plots show the 16th, 25th, 50th, 75th, and 84th percentiles. not significantly distinguished from natural modulations (Stevenson, 2012; Figure 14.14). Because the change in tropical mean conditions (especially the zonal gradient) in a warming climate is model depend- ent (Section 14.4.1), changes in ENSO intensity for the 21st century (Solomon and Newman, 2011; Hu et al., 2012a) are uncertain (Figure 14.14). Future changes in ENSO depend on competing changes in cou- pled ocean—atmospheric feedback (Philip and Van Oldenborgh, 2006; Collins et al., 2010; Vecchi and Wittenberg, 2010), and on the dynami- cal regime a given model is in. There is high confidence, however, that ENSO will remain the dominant mode of natural climate variability in the 21st century (Collins et al., 2010; Guilyardi et al. 2012; Kim and Yu 2012; Stevenson 2012).
Figure 14.15 | Changes to sea level pressure (SLP) teleconnections during December, January and February (DJF) in the CMIP5 models. (a) SLP anomalies for El Nifo during the 20th century. (b) SLP anomalies for La Nina during the 20th century. (c) SLP anomalies for El Nifio during RCP4.5. (d) SLP anomalies for La Nina during RCP4.5. Maps in (a){d) are stippled where more than two thirds 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 Nifo (La Nifia) 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 ensemblemember detrended Nino3.4 SST. (e) Change in the ‘centre of mass’ of the Aleutian Low SLP anomaly, RCP4.5—20th century. The Aleutian Low SLP centre of mass is a vector with two elements (lat, lon), and is defined as the sum of (lat, lon) weighted by the SLP anomaly, over all points in the region 180°E to 120°E, 40°N to 60°N having a negative SLP anomaly during El Nifo. In a warmer climate, the increase in atmospheric moisture intensifies temporal variability of precipitation even if atmospheric circulation The North Atlantic Oscillation (NAO), the North Pacific Oscillation (NPO) and the Northern and Southern Annular Modes (NAM and SAM) are dominant modes of variability in the extratropics. These modes are the focus of much research attention, especially in impact studies, where they are often used as aggregate descriptors of past regional climate
Figure 14.15 | Changes to sea level pressure (SLP) teleconnections during December, January and February (DJF) in the CMIP5 models. (a) SLP anomalies for El Nifo during the 20th century. (b) SLP anomalies for La Nina during the 20th century. (c) SLP anomalies for El Nifio during RCP4.5. (d) SLP anomalies for La Nina during RCP4.5. Maps in (a){d) are stippled where more than two thirds 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 Nifo (La Nifia) 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 ensemblemember detrended Nino3.4 SST. (e) Change in the ‘centre of mass’ of the Aleutian Low SLP anomaly, RCP4.5—20th century. The Aleutian Low SLP centre of mass is a vector with two elements (lat, lon), and is defined as the sum of (lat, lon) weighted by the SLP anomaly, over all points in the region 180°E to 120°E, 40°N to 60°N having a negative SLP anomaly during El Nifo. In a warmer climate, the increase in atmospheric moisture intensifies temporal variability of precipitation even if atmospheric circulation The North Atlantic Oscillation (NAO), the North Pacific Oscillation (NPO) and the Northern and Southern Annular Modes (NAM and SAM) are dominant modes of variability in the extratropics. These modes are the focus of much research attention, especially in impact studies, where they are often used as aggregate descriptors of past regional climate
Figure 14.16 | Summary of multi-model ensemble simulations of wintertime (December to February) mean North Atlantic Oscillation (NAO), Northern Annular Mode (NAM) anc Southern Annular Mode (SAM) sea level pressure (SLP) indices for historical and RCP4.5 scenarios produced by 39 climate models participating in CMIP5. Panels (a)—(c) show tim series of the ensemble mean (black line) and inter-quartile range (grey shading) of the mean index for each model. Panels (d)-(f) show scatter plots of individual model 2081-2101 time means versus 1986-2005 time means (black crosses) together with (-2,+2) standard error bars. The NAO index is defined here as the difference of regional averages: (90°V to 60°E, 20°N to 55°N) minus (90°W to 60°E, 55°N to 90°N) (see Stephenson et al., 2006). The NAM and SAM are defined as zonal indices: NAM as the difference in zonal meal SLP at 35°N and 65°N (Li and Wang, 2003) and SAM as the difference in zonal mean SLP at 40°S and 65°S (Gong and Wang, 1999). All indices have been centred to have zer time mean from 1861-1900. Comparison of simulated and observed trends from 1961-2011 is shown in Figure 10.13. Figures 14.16a, b summarize the wintertime NAO and NAM indices simulated by models participating in the CMIP5 experiment (Gillett and Fyfe, 2013). The multi-model mean of the NAO and NAM indices are similar and exhibit small linear trends in agreement with those shown for the NAM index in AR4 (AR4, Figure 10.17a). The multi-model mean projected increase of around 1 to 2 hPa from 1850 to 2100 is smaller than the spread of around 2 to 4 hPa between model simula- tions (Figure 14.16).
Figure 14.16 | Summary of multi-model ensemble simulations of wintertime (December to February) mean North Atlantic Oscillation (NAO), Northern Annular Mode (NAM) anc Southern Annular Mode (SAM) sea level pressure (SLP) indices for historical and RCP4.5 scenarios produced by 39 climate models participating in CMIP5. Panels (a)—(c) show tim series of the ensemble mean (black line) and inter-quartile range (grey shading) of the mean index for each model. Panels (d)-(f) show scatter plots of individual model 2081-2101 time means versus 1986-2005 time means (black crosses) together with (-2,+2) standard error bars. The NAO index is defined here as the difference of regional averages: (90°V to 60°E, 20°N to 55°N) minus (90°W to 60°E, 55°N to 90°N) (see Stephenson et al., 2006). The NAM and SAM are defined as zonal indices: NAM as the difference in zonal meal SLP at 35°N and 65°N (Li and Wang, 2003) and SAM as the difference in zonal mean SLP at 40°S and 65°S (Gong and Wang, 1999). All indices have been centred to have zer time mean from 1861-1900. Comparison of simulated and observed trends from 1961-2011 is shown in Figure 10.13. Figures 14.16a, b summarize the wintertime NAO and NAM indices simulated by models participating in the CMIP5 experiment (Gillett and Fyfe, 2013). The multi-model mean of the NAO and NAM indices are similar and exhibit small linear trends in agreement with those shown for the NAM index in AR4 (AR4, Figure 10.17a). The multi-model mean projected increase of around 1 to 2 hPa from 1850 to 2100 is smaller than the spread of around 2 to 4 hPa between model simula- tions (Figure 14.16).
In summary, the increased ability in simulating blocking in some models indicate that there is medium confidence that the frequency of NH and SH blocking will not increase, while trends in blocking intensity and persistence remain uncertain. The implications for blocking related regional changes in North America, Europe and Mediterranean and Central and North Asia are therefore also uncertain [Box 14.2 and 14.8.3, 14.8.6, 14.8.8] Box 14.2, Figure 1 | Annual mean blocking frequency in the NH (expressed in % of time, that is, 1% means about 4 days per year) as simulated by a set of CMIP! models (colour lines) for the 1961-1990 period of one run of the historical simulation. Grey shading shows the mean model result plus/minus one standard deviation Black thick line indicates the observed blocking frequency derived from the National Centers for Environmental Prediction/National Center for Atmospheric Researct (NCEP/NCAR) reanalysis. Only CMIP5 models with available 500 hPa geopotential height daily data at http://pcmdi3.|Inl.gov/esgcet/home.htm have been used. Block ing is defined as in Barriopedro et al. (2006), which uses a modified version of the(Tibaldi and Molteni, 1990) index. Daily data was interpolated to a common regula 2.5° x 2.5° longitude—latitude grid before detecting blocking.
In summary, the increased ability in simulating blocking in some models indicate that there is medium confidence that the frequency of NH and SH blocking will not increase, while trends in blocking intensity and persistence remain uncertain. The implications for blocking related regional changes in North America, Europe and Mediterranean and Central and North Asia are therefore also uncertain [Box 14.2 and 14.8.3, 14.8.6, 14.8.8] Box 14.2, Figure 1 | Annual mean blocking frequency in the NH (expressed in % of time, that is, 1% means about 4 days per year) as simulated by a set of CMIP! models (colour lines) for the 1961-1990 period of one run of the historical simulation. Grey shading shows the mean model result plus/minus one standard deviation Black thick line indicates the observed blocking frequency derived from the National Centers for Environmental Prediction/National Center for Atmospheric Researct (NCEP/NCAR) reanalysis. Only CMIP5 models with available 500 hPa geopotential height daily data at http://pcmdi3.|Inl.gov/esgcet/home.htm have been used. Block ing is defined as in Barriopedro et al. (2006), which uses a modified version of the(Tibaldi and Molteni, 1990) index. Daily data was interpolated to a common regula 2.5° x 2.5° longitude—latitude grid before detecting blocking.
Figure 14.17 | General consensus assessment of the numerical experiments described in Supplementary Material Tables 14.SM.1 to 14.SM.4. 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 judgement after subjective normalization of the model projections. Four metrics were considered: the percent change in (I) the total annual frequency of tropical storms, (Il) the annual frequency of Category 4 and 5 storms, (Ill) 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 are 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. A number of studies since the AR4 have attempted to project future changes in tropical cyclone tracks and genesis at inter- or intra-ba- sin scale (Leslie et al., 2007; Vecchi and Soden, 2007b; Emanuel et al., 2008; Yokoi and Takayabu, 2009; Zhao et al., 2009; Li et al., 2010b; Murakami and Wang, 2010; Lavender and Walsh, 2011; Murakami et al., 2011a, 2013). These studies suggest that projected changes in tropical cyclone activity are strongly correlated with projected changes in the spatial pattern of tropical SST (Sugi et al., 2009; Chauvin and Royer, 2010; Murakami et al., 2011b; Zhao and Held, 2012) and asso- ciated weakening of the Pacific Walker Circulation (Vecchi and Soden, 2007a), indicating that reliable projections of regional tropical cyclone activity depend critically on the reliability of the projected pattern of SST changes. However, assessing changes in regional tropical cyclone frequency is still limited because confidence in projections critically depend on the performance of control simulations (Murakami and Sugi, 2010), and current climate models still fail to simulate observed temporal and spatial variations in tropical cyclone frequency (Walsh et al., 2012). As noted above, tropical cyclone genesis and track variability The reduction in signal-to-noise ratio that accompanies changing focus from global to regional scales also lengthens the emergence time scale (i.e., the time required for a trend signal to rise above the nat- ural variability in a statistically significant way). Based on changes in tropical cyclone intensity predicted by idealized numerical simulations with carbon dioxide (CO,)-induced tropical SST warming, Knutson and Tuleya (2004) suggested that clearly detectable increases may not be manifest for decades to come. The more recent high-resolution dynam- ical downscaling study of Bender et al. (2010) supports this argument and suggests that the predicted increases in the frequency of the strongest Atlantic storms may not emerge as a statistically significant signal until the latter half of the 21st century under the SRES A1B emis- sionscenario. However, regional forcing by agents other than GHGs,
Figure 14.17 | General consensus assessment of the numerical experiments described in Supplementary Material Tables 14.SM.1 to 14.SM.4. 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 judgement after subjective normalization of the model projections. Four metrics were considered: the percent change in (I) the total annual frequency of tropical storms, (Il) the annual frequency of Category 4 and 5 storms, (Ill) 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 are 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. A number of studies since the AR4 have attempted to project future changes in tropical cyclone tracks and genesis at inter- or intra-ba- sin scale (Leslie et al., 2007; Vecchi and Soden, 2007b; Emanuel et al., 2008; Yokoi and Takayabu, 2009; Zhao et al., 2009; Li et al., 2010b; Murakami and Wang, 2010; Lavender and Walsh, 2011; Murakami et al., 2011a, 2013). These studies suggest that projected changes in tropical cyclone activity are strongly correlated with projected changes in the spatial pattern of tropical SST (Sugi et al., 2009; Chauvin and Royer, 2010; Murakami et al., 2011b; Zhao and Held, 2012) and asso- ciated weakening of the Pacific Walker Circulation (Vecchi and Soden, 2007a), indicating that reliable projections of regional tropical cyclone activity depend critically on the reliability of the projected pattern of SST changes. However, assessing changes in regional tropical cyclone frequency is still limited because confidence in projections critically depend on the performance of control simulations (Murakami and Sugi, 2010), and current climate models still fail to simulate observed temporal and spatial variations in tropical cyclone frequency (Walsh et al., 2012). As noted above, tropical cyclone genesis and track variability The reduction in signal-to-noise ratio that accompanies changing focus from global to regional scales also lengthens the emergence time scale (i.e., the time required for a trend signal to rise above the nat- ural variability in a statistically significant way). Based on changes in tropical cyclone intensity predicted by idealized numerical simulations with carbon dioxide (CO,)-induced tropical SST warming, Knutson and Tuleya (2004) suggested that clearly detectable increases may not be manifest for decades to come. The more recent high-resolution dynam- ical downscaling study of Bender et al. (2010) supports this argument and suggests that the predicted increases in the frequency of the strongest Atlantic storms may not emerge as a statistically significant signal until the latter half of the 21st century under the SRES A1B emis- sionscenario. However, regional forcing by agents other than GHGs,
FAQ 14.2, Figure 1 | Projected 21st century changes in annual mean and annual extremes (over land) of surface air temperature and precipitation: (a) mean surface temperature per °C of global mean change, (b) 90th percentile of daily maximum temperature per °C of global average maximum temperature, (c) mean precipitation (ir % per °C of global mean temperature change), and (d) fraction of days with precipitation exceeding the 95th percentile. Sources: Panels (a) and (c) projected change: in means between 1986-2005 and 2081-2100 from CMIP5 simulations under RCP4.5 scenario (see Chapter 12, Figure 12.41); Panels (b) and (d) projected change: in extremes over land between 1980-1999 and 2081-2100 (adapted from Figures 7 and 12 of Orlowsky and Seneviratne, 2012).
FAQ 14.2, Figure 1 | Projected 21st century changes in annual mean and annual extremes (over land) of surface air temperature and precipitation: (a) mean surface temperature per °C of global mean change, (b) 90th percentile of daily maximum temperature per °C of global average maximum temperature, (c) mean precipitation (ir % per °C of global mean temperature change), and (d) fraction of days with precipitation exceeding the 95th percentile. Sources: Panels (a) and (c) projected change: in means between 1986-2005 and 2081-2100 from CMIP5 simulations under RCP4.5 scenario (see Chapter 12, Figure 12.41); Panels (b) and (d) projected change: in extremes over land between 1980-1999 and 2081-2100 (adapted from Figures 7 and 12 of Orlowsky and Seneviratne, 2012).
Figure 14.18 | Maps of precipitation changes for North America in 2080-2099 with respect to 1986-2005 in June, July and August (above) and December to February (belov in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes in 2075-2099 wit respect to 1979-2003 in the SRES A1B scenario with the 12 member 60 km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulation Model 3.2 (AGCM3.. multi-physics, multi-SST ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface air temperature changes in each scenario. Lig! hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% | models (or members) have the same sign with the ensemble mean changes. Warm-season precipitation also exhibits significant increases in Alaska, northern Canada, and Eastern NA by century's end (Figures 14.18, Al.19, Al.22). However, CMIP5 models disagree on the sign of the precipitation change over the rest of NA (Figures Al.26 and Al.27), consistent with CMIP3 results (Figure 14.18; Neelin et al., 2006; Rauscher et al., 2008; Seth et al., 2010). One set of high resolution simulatons (Endo et al., 2012) shows a tendency towards more precipi- tation than either CMIP3 or CMIP5 models (Figure 14.18), suggesting the simulated warm-season precipitation change in the region may be resolution-dependent. Future precipitation changes associated with In summary, it is very likely that by mid-century the anthropogenic warming signal will be large compared to natural variability such as that stemming from the NAO, ENSO, PNA, PDO, and the NAMS in all
Figure 14.18 | Maps of precipitation changes for North America in 2080-2099 with respect to 1986-2005 in June, July and August (above) and December to February (belov in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes in 2075-2099 wit respect to 1979-2003 in the SRES A1B scenario with the 12 member 60 km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulation Model 3.2 (AGCM3.. multi-physics, multi-SST ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface air temperature changes in each scenario. Lig! hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% | models (or members) have the same sign with the ensemble mean changes. Warm-season precipitation also exhibits significant increases in Alaska, northern Canada, and Eastern NA by century's end (Figures 14.18, Al.19, Al.22). However, CMIP5 models disagree on the sign of the precipitation change over the rest of NA (Figures Al.26 and Al.27), consistent with CMIP3 results (Figure 14.18; Neelin et al., 2006; Rauscher et al., 2008; Seth et al., 2010). One set of high resolution simulatons (Endo et al., 2012) shows a tendency towards more precipi- tation than either CMIP3 or CMIP5 models (Figure 14.18), suggesting the simulated warm-season precipitation change in the region may be resolution-dependent. Future precipitation changes associated with In summary, it is very likely that by mid-century the anthropogenic warming signal will be large compared to natural variability such as that stemming from the NAO, ENSO, PNA, PDO, and the NAMS in all
Figure 14.19 | Maps of precipitation changes for Central America and Caribbean in 2080-2099 with respect to 1986-2005 in June to September (above) and December tc March (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes ir 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12 member 60 km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulation Mode 3.2 (AGCM3.2) multi-physics, multi-SST ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface air temperature changes in eact scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes.
Figure 14.19 | Maps of precipitation changes for Central America and Caribbean in 2080-2099 with respect to 1986-2005 in June to September (above) and December tc March (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes ir 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12 member 60 km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulation Mode 3.2 (AGCM3.2) multi-physics, multi-SST ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface air temperature changes in eact scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes.
Figure 14.20 Maps of precipitation changes for South America in 2080-2099 with respect to 1986-2005 in June to September (above) and December to March (below) in th SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes in 2075-2099 with respec to 1979-2003 in the SRES A1B scenario with the 12-member 60-km mesh Meteorological Research Institute (MRI)- Atmospheric General Circulation Model 3.2 (AGCM3.2) multi physics, multi-SST ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface air temperature changes in each scenario. Light hatchin denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% of models (0 members) have the same sign with the ensemble mean changes. Regarding future projections, CMIP5 models indicate higher temper- atures over all of SA, with the largest changes in southeastern Ama- zonia by century's end. (Figures Al.28, Al.29, Table 14.1). Temperature changes projected by RCMs forced by a suite of CMIP3 models agree that the largest warming occurs over the southern Amazon during aus- tral winter. Regional models project a greater frequency of warm nights CMIP5 results confirm precipitation changes projected by CMIP3 models in the majority of SA regions, with increased confidence, as more models agree in the changes (AI.30, Al.34, Al31, Al35). Inter-mod- el spread in precipitation also decreased in some SA regions from CMIP3 to CMIP5 (Blazquez and Nufez, 2012). CMIP5 precipitation
Figure 14.20 Maps of precipitation changes for South America in 2080-2099 with respect to 1986-2005 in June to September (above) and December to March (below) in th SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes in 2075-2099 with respec to 1979-2003 in the SRES A1B scenario with the 12-member 60-km mesh Meteorological Research Institute (MRI)- Atmospheric General Circulation Model 3.2 (AGCM3.2) multi physics, multi-SST ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface air temperature changes in each scenario. Light hatchin denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% of models (0 members) have the same sign with the ensemble mean changes. Regarding future projections, CMIP5 models indicate higher temper- atures over all of SA, with the largest changes in southeastern Ama- zonia by century's end. (Figures Al.28, Al.29, Table 14.1). Temperature changes projected by RCMs forced by a suite of CMIP3 models agree that the largest warming occurs over the southern Amazon during aus- tral winter. Regional models project a greater frequency of warm nights CMIP5 results confirm precipitation changes projected by CMIP3 models in the majority of SA regions, with increased confidence, as more models agree in the changes (AI.30, Al.34, Al31, Al35). Inter-mod- el spread in precipitation also decreased in some SA regions from CMIP3 to CMIP5 (Blazquez and Nufez, 2012). CMIP5 precipitation
Figure 14.21 | (a) December, January and February (DJF) and (b) June, July and August (JJA) relative precipitation change in 2071-2100 with respect to 1961-1990 in the A1B scenario from an teary of 10 Regional Climate Models (RCMs) participating in the Europe—South America Network for Climate Change Assessment and Impact Studies-La Plata Basin (CLARIS-LPB) Project. Hatching denotes areas where 8 out of 10 RCMs agree in the sign of the relative change. (c) DJF and (d) JJA dispersion among regional model projections of precipitation changes averaged over land grid points in Southeastern South America (SESA, 35°S to 25°S, 60°W to 50°W), South Atlantic Convergence Zone (SACZ, 25°S to 15°S, 45°W to 40°W) and southern Amazonia (15°S to 10°S, 65°W to 55°W), indicated by the boxes in (a).
Figure 14.21 | (a) December, January and February (DJF) and (b) June, July and August (JJA) relative precipitation change in 2071-2100 with respect to 1961-1990 in the A1B scenario from an teary of 10 Regional Climate Models (RCMs) participating in the Europe—South America Network for Climate Change Assessment and Impact Studies-La Plata Basin (CLARIS-LPB) Project. Hatching denotes areas where 8 out of 10 RCMs agree in the sign of the relative change. (c) DJF and (d) JJA dispersion among regional model projections of precipitation changes averaged over land grid points in Southeastern South America (SESA, 35°S to 25°S, 60°W to 50°W), South Atlantic Convergence Zone (SACZ, 25°S to 15°S, 45°W to 40°W) and southern Amazonia (15°S to 10°S, 65°W to 55°W), indicated by the boxes in (a).
Figure 14.22 | Maps of precipitation changes for Europe and Mediterranean in 2080-2099 with respect to 1986-2005 in June to August (above) and December to Febrt ary (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes | 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12 member 60 km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulatio Model 3.2 (AGCM3.2) multi-physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface a temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while den: hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes. Recent studies have clearly identified a possible amplification of tem- perature extremes by changes in soil moisture (Jaeger and Seneviratne, Some new investigations have focussed on the uncertainties associat- ed with model projections. A large ensemble of RCM-GCM shows that the temperature response is robust in spite of a considerable uncer- tainty related to choice of model combination (GCM/RCM) and sam-
Figure 14.22 | Maps of precipitation changes for Europe and Mediterranean in 2080-2099 with respect to 1986-2005 in June to August (above) and December to Febrt ary (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes | 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12 member 60 km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulatio Model 3.2 (AGCM3.2) multi-physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface a temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while den: hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes. Recent studies have clearly identified a possible amplification of tem- perature extremes by changes in soil moisture (Jaeger and Seneviratne, Some new investigations have focussed on the uncertainties associat- ed with model projections. A large ensemble of RCM-GCM shows that the temperature response is robust in spite of a considerable uncer- tainty related to choice of model combination (GCM/RCM) and sam-
Figure 14.23 | Maps of precipitation changes for Africa in 2080-2099 with respect to 1986-2005 in June to September (above) and December to March (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes in 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12-member 60-km mesh Meteorological Research Institute (MRI)-Atmospheric Generl Circulation Model 3.2 (AGCM3.2) multi- physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface air temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes. In projections of the 21st century, the CMIP3 models produced both significant drying and significant moistening (Held et al., 2005; Biasutti and Giannini, 2006; Cook and Vizy, 2006; Lau et al., 2006), and the mechanisms by which a model dries or wets the Sahel are not fully understood (Cook, 2008). At least qualitatively, the CMIP3 ensem- ble simulates a more robust response during the pre-onset and the demise portion of the rainy season (Biasutti and Sobel, 2009; Seth et al., 2011). Rainfall is projected to decrease in the early phase of the seasons—implying a small delay in the main rainy season, but is pro- jected to increase at the end of the season—implying an intensifica- tion of late-season rains (d'Orgeval et al., 2006), although this appears to be less robust in the CMIP5 models (Section 14.2.4). Projections of a change in the timing of the rains is common to other monsoon regions (Li et al., 2006; Biasutti and Sobel, 2009; Seth et al., 2011), including southern Africa. An evaluation of six GCMs over East Africa by Conway et al. (2007) reveals no clear multi-model trend in mean annual rainfall by the 2080s, but some indications of increased SON and decreased March,
Figure 14.23 | Maps of precipitation changes for Africa in 2080-2099 with respect to 1986-2005 in June to September (above) and December to March (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation changes in 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12-member 60-km mesh Meteorological Research Institute (MRI)-Atmospheric Generl Circulation Model 3.2 (AGCM3.2) multi- physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface air temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes. In projections of the 21st century, the CMIP3 models produced both significant drying and significant moistening (Held et al., 2005; Biasutti and Giannini, 2006; Cook and Vizy, 2006; Lau et al., 2006), and the mechanisms by which a model dries or wets the Sahel are not fully understood (Cook, 2008). At least qualitatively, the CMIP3 ensem- ble simulates a more robust response during the pre-onset and the demise portion of the rainy season (Biasutti and Sobel, 2009; Seth et al., 2011). Rainfall is projected to decrease in the early phase of the seasons—implying a small delay in the main rainy season, but is pro- jected to increase at the end of the season—implying an intensifica- tion of late-season rains (d'Orgeval et al., 2006), although this appears to be less robust in the CMIP5 models (Section 14.2.4). Projections of a change in the timing of the rains is common to other monsoon regions (Li et al., 2006; Biasutti and Sobel, 2009; Seth et al., 2011), including southern Africa. An evaluation of six GCMs over East Africa by Conway et al. (2007) reveals no clear multi-model trend in mean annual rainfall by the 2080s, but some indications of increased SON and decreased March,
Figure 14.24 | Maps of precipitation changes for Central, North, East and South Asia in 2080-2099 with respect to 1986-2005 in June to September (above) and Decembe to March (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation change: in 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12-member 60-km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulatior Model 3.2 (AGCM3.2) multi-physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface ai temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dens hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes.
Figure 14.24 | Maps of precipitation changes for Central, North, East and South Asia in 2080-2099 with respect to 1986-2005 in June to September (above) and Decembe to March (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation change: in 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12-member 60-km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulatior Model 3.2 (AGCM3.2) multi-physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface ai temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dens hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes.
Figure 14.25 | Linear trend for local summer (a) total precipitation and (b) R95 (summer total precipitation when PR >95th percentile) during 1961-2006. The unit is % per 50-years. The trends statistically significant at the 5% level are dotted. The daily precipitation data over Australia and China are produced by the Australian Water Availability Project (AWAP, Jones et al., 2009a) and National Climate Centre China of China Meteorological Administration (Wu and Gao, 2013), respectively, while that over the other area is compilec by the Highly Resolved Observational Data Integration Towards the Evaluation of Water Resources (APHRODITE) project (Yatagai et al., 2012). The resolution of precipitation date set is 0.5° x 0.5°. Local summer is defined as June, July and August in the Northern Hemisphere, and December, January and February in the Southern Hemisphere.
Figure 14.25 | Linear trend for local summer (a) total precipitation and (b) R95 (summer total precipitation when PR >95th percentile) during 1961-2006. The unit is % per 50-years. The trends statistically significant at the 5% level are dotted. The daily precipitation data over Australia and China are produced by the Australian Water Availability Project (AWAP, Jones et al., 2009a) and National Climate Centre China of China Meteorological Administration (Wu and Gao, 2013), respectively, while that over the other area is compilec by the Highly Resolved Observational Data Integration Towards the Evaluation of Water Resources (APHRODITE) project (Yatagai et al., 2012). The resolution of precipitation date set is 0.5° x 0.5°. Local summer is defined as June, July and August in the Northern Hemisphere, and December, January and February in the Southern Hemisphere.
Figure 14.26 | Maps of precipitation changes for West Asia in 2080-2099 with respect to 1986-2005 in June, July and August (above) and December, January and Februar (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). The figures on the right are the precipitation change: in 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12-member 60-km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulatior Model 3.2 (AGCM3.2) multi-physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface ai temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes.
Figure 14.26 | Maps of precipitation changes for West Asia in 2080-2099 with respect to 1986-2005 in June, July and August (above) and December, January and Februar (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). The figures on the right are the precipitation change: in 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12-member 60-km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulatior Model 3.2 (AGCM3.2) multi-physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface ai temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while dense hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes.
Figure 14.27 | Maps of precipitation changes for Southeast Asia, Australia and New Zealand in 2080-2099 with respect to 1986-2005 in June to September (above) and Decem ber to March (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation change: in 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12-member 60- km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulatior Model 3.2 (AGCM3.2) multi-physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface ai temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while densé hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes.
Figure 14.27 | Maps of precipitation changes for Southeast Asia, Australia and New Zealand in 2080-2099 with respect to 1986-2005 in June to September (above) and Decem ber to March (below) in the SRES A1B scenario with 24 CMIP3 models (left), and in the RCP4.5 scenario with 39 CMIP5 models (middle). Right figures are the precipitation change: in 2075-2099 with respect to 1979-2003 in the SRES A1B scenario with the 12-member 60- km mesh Meteorological Research Institute (MRI)-Atmospheric General Circulatior Model 3.2 (AGCM3.2) multi-physics, multi-sea surface temperature (SST) ensembles (Endo et al., 2012). Precipitation changes are normalized by the global annual mean surface ai temperature changes in each scenario. Light hatching denotes where more than 66% of models (or members) have the same sign with the ensemble mean changes, while densé hatching denotes where more than 90% of models (or members) have the same sign with the ensemble mean changes.
Table 14.1 (continued)
Table 14.1 (continued)
(continued on next page)
(continued on next page)
Table 14.1 (continued)
Table 14.1 (continued)
Table 14.2 | Assessed confidence (high, medium, low) in climate projections of regional temperature and precipitation change from the multi-model ensemble of CMIP5 model. for the RCP4.5 scenario. Column 1 refers to the SREX regions (cf. Seneviratne et al, 2012, page 12. The region’s coordinates can be found from their online Appendix 3.A) and si additional regions including the two polar regions, the Caribbean, Indian Ocean and Pacific Island States (see Annex | for further details). Columns 2 to 4 show confidence in models ability to simulate present-day mean temperature and precipitation as well as the most important phenomena for that region based on Figures 9.39, 9.40, and 9.45. In column 4 the individual phenomena are listed, with associated confidence levels shown below, in the same order as the phenomena. Note that only phenomena assessed in Figure 9.45 are listed. Column 5 is an interpretation of the relevance of the main climate phenomena for future regional climate change, based on Table 14.3. Note that the SREX regions are smalle than the regions listed in Table 14.3. Columns 6 and 7 express confidence in projected temperature and precipitation changes, based solely on model agreement for 2080-209: vs. 1985-2005, as listed in Table 14.1 and in the maps shown in Annex |. The confidence is assessed for two periods for temperature (DJF and JJA) and two-half year periods fo precipitation (October to March and April to September). When the projections are consistent with no significant change, it is marked by an asterisk (*) and the assigned confidence is medium. Further details on how confidence levels have been assigned are provided in the Supplementary Material (Section 14.SM.6.1).
Table 14.2 | Assessed confidence (high, medium, low) in climate projections of regional temperature and precipitation change from the multi-model ensemble of CMIP5 model. for the RCP4.5 scenario. Column 1 refers to the SREX regions (cf. Seneviratne et al, 2012, page 12. The region’s coordinates can be found from their online Appendix 3.A) and si additional regions including the two polar regions, the Caribbean, Indian Ocean and Pacific Island States (see Annex | for further details). Columns 2 to 4 show confidence in models ability to simulate present-day mean temperature and precipitation as well as the most important phenomena for that region based on Figures 9.39, 9.40, and 9.45. In column 4 the individual phenomena are listed, with associated confidence levels shown below, in the same order as the phenomena. Note that only phenomena assessed in Figure 9.45 are listed. Column 5 is an interpretation of the relevance of the main climate phenomena for future regional climate change, based on Table 14.3. Note that the SREX regions are smalle than the regions listed in Table 14.3. Columns 6 and 7 express confidence in projected temperature and precipitation changes, based solely on model agreement for 2080-209: vs. 1985-2005, as listed in Table 14.1 and in the maps shown in Annex |. The confidence is assessed for two periods for temperature (DJF and JJA) and two-half year periods fo precipitation (October to March and April to September). When the projections are consistent with no significant change, it is marked by an asterisk (*) and the assigned confidence is medium. Further details on how confidence levels have been assigned are provided in the Supplementary Material (Section 14.SM.6.1).
OOUr Table 14.3 | Summary of the relevance of projected changes in major phenomena for mean change in future regional climate. The relevance is classified into high (red), medium (yellow), low (cyan), and ‘no obvious relevance’ (grey), based on confidence that there will be a change in the phenomena (‘HP’ for high, ‘MP’ for medium, ‘LP’ for low), and confidence in the impact of the phenomena on each region (‘HI’ for high, ‘MI’ for medium, ‘LI’ for low). More information on how these assessments have been constructed is given in the Supplementary Material (Section 14.SM.6.1).
OOUr Table 14.3 | Summary of the relevance of projected changes in major phenomena for mean change in future regional climate. The relevance is classified into high (red), medium (yellow), low (cyan), and ‘no obvious relevance’ (grey), based on confidence that there will be a change in the phenomena (‘HP’ for high, ‘MP’ for medium, ‘LP’ for low), and confidence in the impact of the phenomena on each region (‘HI’ for high, ‘MI’ for medium, ‘LI’ for low). More information on how these assessments have been constructed is given in the Supplementary Material (Section 14.SM.6.1).
Related topics:
GeographyEnvironmental ScienceAtmospheric CirculationClimate PatternsExtratropical Cyclones
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved