Data and methods Acknowledgments 1. Mr. ABM Sertajur Rahmann, and Mr. Md. Irfanul Islam for their helps in field survey 2. JSPS Grant-in-Aid for Young Scientist (B) 3. GELs Education Program in Hiroshima University (MEXT Special Coordination Fund) Results Conclusions 1. Relative SLR trend (1993~2007) 4.46 mm/yr at Hiron Point 2. Relative SLR to 2050 0.34 m 3. 1.91 m < 100-yr RL < 2.48 m 4. The residual non-linear trend can be considered as a relative SLR trend due to long-term local effects such as seismic movements and local subsidence. 5. Ensemble EMD is a useful tool for determining a non-linear trend, detrending and filtering. Further improvements 1. Semi-empirical approach (Rahmstorf, 2007; Vermeer and Rahmstorf, 2009; Rahmstorf et al., 2012) can be applied for further improvement of regional projection rather than simple extension of polynomial. 2. Return levels of storm surge depend on past condition without considering regional future climates. It also has to be improved. Han Soo Lee* (http://home.hiroshima-u.ac.jp/hslee) Literature cited Rahmstorf, S., (2007). Science, 315(5810), 368-370. Vermeer, M. and Rahmstorf, S., (2009). PNAS, 106(51), 21527-21532. Rahmstorf, S., Perrette, M. and Vermeer, M., (2012). Climate Dynamics, 39(3-4), 861-875. Further reading Lee, H.S., (2013). Estimation of extreme sea levels along the Bangladesh coast due to storm surge and sea level rise using EEMD and EVA. JGR: Oceans, 118(9), 4273-4285. * Graduate School for International Development and Cooperation, Hiroshima University, 1-5-1 Kagamiyama, Higashi-Hiroshima 739-8529, Japan ([email protected]) Bangladesh need more accurate regional relative sea level rise scenarios! modeling extreme sea level relative sea level rise (SLR) extreme storm surge Hiron Point Objective Time scale: Short-term Coastal structure: Design factor Time scale: long-term Coastal structure: Fatigue failure + 1 2 3 4 5 6 7 8 EEMD: Ensemble empirical mode decomposition => detrending and detiding tool EVA: Extreme value analysis IMF: Intrinsic mode function Data: 32-yr hourly sea level records at Hiron Point 5 relative SLR trend : +4.46 mm/yr 6 Detided storm surge data Source data 2 3 Composites of semi-diurnal tides Composites of diurnal tides Non-linear trend 4 4 7 Parameter Estimate Std. error Location (μ) 0.920 0.049 Scale (σ) 0.258 0.033 Shape (ξ) -0.170 0.086 Table 1. Distribution parameters estimated by maximum likelihood method GEV distribution 1.57 m < 100yr RL< 2.14 m 95% 1. Reconstructed source data is decomposed into 18 IMFs and the residue (non-linear trend) 2. Following the definition of storm surge, the tidal variations are detided 3. Then, annual maxima of storm surges is applied to EVAto obtain extreme storm surge 4. Regional projection of relative SLR is conducted by polynomial extension 8 Regional SLR projection: 0.34 m to 2050 relative SLR trend storm surge Return Level
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Data and methods
Acknowledgments1. Mr. ABM Sertajur Rahmann, and Mr. Md. Irfanul
Islam for their helps in field survey
2. JSPS Grant-in-Aid for Young Scientist (B)
3. GELs Education Program in Hiroshima University
(MEXT Special Coordination Fund)
Results
Conclusions1. Relative SLR trend (1993~2007)
4.46 mm/yr at Hiron Point
2. Relative SLR to 2050 0.34 m
3. 1.91 m < 100-yr RL < 2.48 m
4. The residual non-linear trend can be
considered as a relative SLR trend due
to long-term local effects such as
seismic movements and local
subsidence.
5. Ensemble EMD is a useful tool for
determining a non-linear trend,
detrending and filtering.
Further improvements1. Semi-empirical approach (Rahmstorf,
2007; Vermeer and Rahmstorf, 2009;
Rahmstorf et al., 2012) can be applied
for further improvement of regional
projection rather than simple extension
of polynomial.
2. Return levels of storm surge depend
on past condition without considering
regional future climates. It also has to
be improved.
Han Soo Lee* (http://home.hiroshima-u.ac.jp/hslee)
Literature citedRahmstorf, S., (2007). Science, 315(5810), 368-370.
Vermeer, M. and Rahmstorf, S., (2009). PNAS, 106(51),
21527-21532.
Rahmstorf, S., Perrette, M. and Vermeer, M., (2012).
Climate Dynamics, 39(3-4), 861-875.
Further readingLee, H.S., (2013). Estimation of extreme sea levels along
the Bangladesh coast due to storm surge and sea level rise
using EEMD and EVA. JGR: Oceans, 118(9), 4273-4285.
* Graduate School for International Development and Cooperation, Hiroshima University, 1-5-1 Kagamiyama, Higashi-Hiroshima 739-8529, Japan([email protected])
Bangladeshneed more accurate
regional relative sea level rise
scenarios!
modeling
extreme sea level
relative
sea level rise
(SLR)
extreme
storm surge
Hiron Point
Objective
Time scale:
Short-term
Coastal structure:
Design factor
Time scale:
long-term
Coastal structure:
Fatigue failure
+
1
2
3
4
5 6
7
8
EEMD: Ensemble
empirical mode
decomposition
=>
detrending and
detiding tool
EVA:Extreme value
analysis
IMF:Intrinsic mode
function
Data:32-yr hourly sea
level records at
Hiron Point
5
relative SLR trend : +4.46 mm/yr6
Detided storm surge dataSource data2
3
Composites of semi-diurnal tides
Composites of diurnal tides
Non-linear trend
4
4
7
Parameter Estimate Std. error
Location (μ) 0.920 0.049
Scale (σ) 0.258 0.033
Shape (ξ) -0.170 0.086
Table 1. Distribution
parameters estimated
by maximum
likelihood method
GEV
distribution
1.57 m < 100yr RL< 2.14 m
95%
1. Reconstructed source data is decomposed into 18 IMFs and the residue (non-linear trend)
2. Following the definition of storm surge, the tidal variations are detided
3. Then, annual maxima of storm surges is applied to EVA to obtain extreme storm surge
4. Regional projection of relative SLR is conducted by polynomial extension
8
Regional SLR projection:
0.34 m to 2050
relative
SLR trend
storm surge Return Level
Related Documents
