Climate induced challenges on the freshwater availability and climate extremes for the Ganges-Brahmaputra-Meghna rivers
in Bangladesh at 1.5°C and 2°C global warming
Professor A.K.M. Saiful Islam
Talk organized by the Oxford Martin Visiting Fellow Programme
Environmental Change Institute, University of Oxford, 21 May 2018
Bangladesh University of Engineering and Technology
Photo Credit: The Daily Ittefaq, 11th July 2017
Outline of the presentation
• Geographical Context of Bangladesh
• Recently Observed Extreme Climate Events in Bangladesh
• Multi-model Ensemble Regional Climate Projections for Impactstudies – HAPPI and CORDEX simulations
• Changes in Rainfall Extremes in the future
• Changes in Metrological Droughts in the future
• Changes in Mean, Low and High Flows in the major Transboundaryrivers: Ganges Brahmaputra Meghna (GBM) River Systems
• Changes in Coastal Permanent Inundation due to Sea Level Rise
Geographical Context
• Bangladesh is a delta of 3 majorriver systems: Ganges,Brahmaputra and Meghna.
• The rivers in Meghna Basin inthe northeast region are flashyin character.
• There is a strong modulation oftide from Bay of Bengal on theriver discharge.
• Mostly flat topography (Between0-5m above MSL)
• Most of the country is veryfertile, can grow crops and havea high agriculture-dependenteconomy.
• However, Bangladesh is alsoknown for it resilient to naturaldisasters like Floods, Cyclone,Storm Surges, Flash Floods, Landslides, River Bank Erosion andThunderstorm & Lightning
Extreme Climate Events in Bangladesh during 2016-2018
Heatwave Flash Flood Lightening Cyclone Roanu Floods
Flash Flood Cyclone Mora Landslide Lightening Floods
6-30 April 2016 12-13 May 201617-21 April 2016 17-23 May 2016 August 2016
13 June 20171-10 April 2017 20 June 201728-31 May 2017 August 2017
2016
2017
Lightning Blast Diseases in Rice
2018 29 April -11 May 2018 1-10 April 2017
67 People Killed
90% Crop damaged 135 people died 168 peopled died 22 people died 3.9 M affected
57 People Killed 30% Crop damaged 57 people killed 24 People Killed 3.7M affected
Global temperature constantly rising since 1880
August came in 0.98˚C above the average from 1951-1980, and 0.16˚C above August 2014, the previous record holder. The record keeps 2016 on track to as the hottest year in the books.
Comparing El Nino impact under global warming
The magnitude of the surface warming is strongly different in the RCP
scenarios, showing the potential impact of mitigation policies.
Time series of global annual mean surface air temperature anomalies (relative to 1986–
2005) from an ensemble of model simulations performed in the framework of CMIP5.
Figure from Collins et al. (2013).
Changes in global mean surface temperature
Global warming will exceed 1.5C by 2025 and 2C by 2040Near surface global annual mean warmingsince pre-industrial for simulations fromCMIP5, CMIP3 and by a HadCM3 perturbedparameter experiments of SRES A1B andthe RCPs. Both concentration and emissionsdriven simulations.
Betts et al. (2011)
50% models
HAPPI Atmospheric simulation
• HAPPI project are atmosphere-only climate simulations, forced by sea-surface temperatures (SSTs), sea-ice concentration (SIC) and green-house gas concentrations.
• Current day simulations (2006-2015), and SSTs from CMIP5 modeloutput are used to estimate the future scenarios corresponding to1.5◦C and 2◦C global warming above pre-industrial conditions, at theend of the 21st century.
• Four models were used for this analysis, and each model had around20 simulations which are bias corrected using ISIMIP2b bias-corrections methods-• CAM4-2degree
• MIROC5
• ECHAM6.3-LR
• NorESM1-HAPPI
Regional Climate Modeling (RCM) for Bangladesh over CORDEX: South Asia
• GCM provides output more than 150kmresolution which is not enough to capturemesoscale processes.
• RCM daily output with horizontalresolution 50km are available for SouthAsia CORDEX domain.
• Predictions are considered for extremeemission scenarios, RCP 8.5
RCM Projections using CIMP5 data
GCM Ensemble RCMRCP8.5 (Year of Crossing)
SWL 1.5 SWL 2 SWL 4
ACCESS1-0 r1i1p1 CSIRO-CCAM-1391M 2034 2046 2085
CCSM4 r1i1p1 CSIRO-CCAM-1391M 2016 2031 2079
CNRM-CM5 r1i1p1 SMHI-RCA4 2032 2046 2088
CNRM-CM5 r1i1p1 CSIRO-CCAM-1391M 2032 2046 2088
EC-EARTH r12i1p1 SMHI-RCA4 2019 2035 2083
CM5A-MR r1i1p1 SMHI-RCA4 2020 2034 2069
MIROC5 r1i1p1 SMHI-RCA4 2038 2052 -
MPI-ESM-LR r1i1p1 CSIRO-CCAM-1391M 2021 2040 2083
MPI-ESM-LR r1i1p1 MPI-CSC-REMO2009 2021 2040 2083
MPI-ESM-LR r1i1p1 SMHI-RCA4 2021 2040 2083
GFDL-ESM2M r1i1p1 SMHI-RCA4 2040 2055 -
Specific Warming Levels (SWLs)It is the mean annual global temperature increase by the end of the century related topreindustrial period (1880). Paris Agreement in 2015 emphasis on reducing GHGs to keep theincrease of global mean temperature below 2C and effort should be made to reduce further to1.5C with respect to pre-industrial period.
Changes of mean temperature and rainfall over Bangladesh at SWLs
Changes of mean seasonal rainfall anomaly at SWLs
Changes of Mean monthly temperature at SWLs
Temperature is rising and chances of having extremes are increasing
TXx- maximum of daily maximum temperature
TNn- minimum of daily minimum temperature
TX90P – Percentage of days when maximum temperature is higher than 90th percentile value.
TX10P- Percentage of days when maximum temperature is lower than 10th percentile value
Changes of Extreme Rainfall
Rx1- maximum 1-day rainfall
Rx50- number of days when rainfall > 50mm
Rx1- maximum 1-day rainfall
Rx50- number of days when rainfall > 50mm
Changes of Rainfall Indices at SWLs
Index Unit SWL 1.5 SWL 2 SWL 4
CDD days 4.16 8.88 13.49
CWD days -6.46 -5.97 -10.4
R10mm days 0.01 1.21 1.98
R20mm days 3.35 6.65 8.91
R100mm days 28.51 26.24 59.77
RX1day mm 9.33 12.39 16.13
RX5day mm 5.73 7.99 10.33
Rainfall Indices• Dry days shows increasing trend. So does the
heavy precipitation in both severity and intensity.With increased number of days with very heavyprecipitation, the chance of floods, flash floods,urban floods and landslides is projected toincrease.
* Baseline period (1995 – 2005)
Changes of Meteorological Droughts
Drought Category SPI/SPEI Values
No drought 0 ≤ Index
Near normal/Mild drought
-1.0 < Index < 0
Moderate drought -1.5 < Index ≤ -1.0
Severe drought -2.0 < Index ≤ -1.5
Extreme drought Index ≤ -2.0
Ensemble time series of SPI (left panel) and SPEI (right panel) at 12-month timescale averaged over the country. The red patch is the ensemble spread of the index values.
Both SPI and SPEI are calculated relative to a baseline period of 1981 to 2010.
Density distribution of SPI and SPEI values at 3 Month time-scale.
Standard Precipitation Index (SPI) and Standard Precipitation and Evapotranspiration Index (SPEI)
Droughts will be increased in the future
Projected future median SPEI values over Bangladesh from 11 ensemble of Regional Climate Model (RCM).
Changes of Meteorological Droughts
• Inclusion of evapotranspiration in the evaluation of drought is important in the context of global warming.
• The country is expected see more and more deviation from the climatic mean condition.
• At the end of the century, the climate of the country may settle to a condition which may be considered “moderate drought” compared to current climate.
• Long meteorological drought will impact the agriculture and socio-economic condition.
Percentage of area affected with time
Water Resources Impact Assessment:SWAT Modeling for the GBM basins
• The GBM basins are located over India (64%), China (18%), Nepal (9%), Bangladesh (7%) and Bhutan (3%), and the elevation of the basins range from about 0 to above 8000 m above mean sea level (amsl).
• The Ganges basin has the largest area out of the three (907,000 km2), the Brahmaputra basin is the second largest (583,000 km2) and the Meghna basin is the smallest (65,000 km2)
• About 60-70% of the total annual precipitation falls during the monsoon (June-September) and another 20-25% falls during the pre-monsoon (March-May)
• The Brahmaputra is a major transboundary river which has an average discharge of approximately 20,000 m3/s and the highest flow was of 102,000 m3/s.
• Soil and Water Assessment Tool (SWAT) is presently one of the most commonly used hydrological models is semi-distributed, process based.
• Model developed using SRTM 90m DEM, GlobCover landuse map, FAO Digital Soil Map of the World and WFDEI weather data.
Ganges-Brahmaputra-Megha Basins
SWAT Model Evaluation
Performance of the SWAT models during calibration and validation periods
Statistical Indicators NSE R2 PBIAS RSR
Ganges
Calibration 0.80 0.81 12.47 0.44
Validation 0.77 0.79 15.52 0.49
Brahmaputra
Calibration 0.81 0.84 15.23 0.43
Validation 0.79 0.83 17.51 0.47
Meghna
Calibration 0.85 0.86 5.33 0.38
Validation 0.83 0.85 5.84 0.41
Changes in Flow of Brahmaputra - HAPPI Runs
Changes in Flow of Ganges -HAPPI Runs
Changes in Flow of Meghna - HAPPI Runs
Changes in Flows of Brahmaputra – CORDEX Runs
Changes in Flows of Meghna – CORDEX Runs
Changes in High Flows – CORDEX Runs
Median
Brahmaputra Meghna
1.5°C 2°C 4°C 1.5°C 2°C 4°C
4% 5% 22% 9% 12% 42%
Median
Brahmaputra Meghna
1.5°C 2°C 4°C 1.5°C 2°C 4°C
26% 36% 137% 22% 43% 107%
Median
Brahmaputra Meghna
1.5°C 2°C 4°C 1.5°C 2°C 4°C
3% 10% 22% 8% 11% 32%
Median
Brahmaputra Meghna
1.5°C 2°C 4°C 1.5°C 2°C 4°C
-8% -16% -50% -16% -21% -60%
Changes in Low Flows – CORDEX Runs
Coastal modeling using Delft3D
• DELFT3D- FLOW is a multi-dimensional (2D or 3D) hydrodynamic(and transport) simulation program which calculates unsteadyflow and transport phenomena that result from tidal andmeteorological forcing on a rectilinear or a curvilinear, boundaryfitted grid.
Tidal Validation: Complex Error of tidal constituents
0
10
20
30
40
50
M2 S2 K1 O1 total
Co
mp
lex
Erro
rs(c
m)
Harmonic Constituents
Hiron Point
FES2012 FES2014 DELFT3D
0
10
20
30
40
50
60
M2 S2 K1 O1 total
Co
mp
lex
Erro
r(cm
)
Harmonics Constituents
Char Changa
FES2012 FES2014 DELFT3D
0
10
20
30
40
50
M2 S2 K1 O1 total
Co
mp
lex
Erro
r(cm
)
Harmonic Constituents
Cox's Bazar
FES2012 FES2014 DELFT3D
Hiron Point
FES2014: 41.9 cm
DELFT3D: 17.5 cm
Char Changa
FES2014: 52.6 cm
DELFT3D: 44.8 cm
Cox’s Bazar
FES2014: 37.6 cm
DELFT3D: 40.2 cm
Changes of Inundation due to SLR
SLR(m)
Inundated Area ( square km)
Percent of Bangladesh
Affected Population
(million)
0.5 2000 1.6 2.5
1 3930 3.8 6
1.5 5300 5.1 8
SLR
(m)
Inundated Area
(km2)
% of inundation
Area
0.5m 491 11.37
1m 1847 42.78
1.5m 2635 61.04
Inundation of the Sundarbans
0.5m SLR
1.0m SLR
Key Messages• Extreme Rainfall over Bangladesh will be increased in the future.
Chances of flash flood and land slides will be increased.• Projected changes in mean annual precipitation and temperature
over the basins are larger in 4°C than in 1.5°C or 2°C. Changes are greater over Meghna basin than Brahmaputra basin.
• Annual discharges of the Meghna basin change almost linearly with changes in the annual basin-averaged precipitation, while changes in discharges of the Brahmaputra basin are less sensitive to changes in precipitation.
• Mean monthly flows are projected to increase the most in July (for Brahmaputra and Ganges) and in June (for Meghna).
• Floods are likely to increase in both rivers as well as flood durations. • However, Hydrological droughts are likely to decrease in both rivers
along with drought durations.• SLR rise will also cause permanent inundations in some parts of the
coastal areas of Bangladesh.
Thank you!
