International Conference on Flood Resilience: Experiences in Asia and Europe 5-7 September 2013, Exeter, United Kingdom Seith Mugume Dr. Diego Gomez Professor David Butler 1 Statistical downscaling methods for climate change impact assessment on urban rainfall extremes for cities in tropical developing countries – A review
15
Embed
Statistical downscaling methods for climate change …icfr2013.ex.ac.uk/papers/D1_Mugume.pdfStatistical downscaling methods for climate change impact ... Example of historical 30 year
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
International Conference on Flood Resilience:
Experiences in Asia and Europe
5-7 September 2013, Exeter, United Kingdom
Seith Mugume
Dr. Diego Gomez
Professor David Butler
1
Statistical downscaling methods for climate change impact
assessment on urban rainfall extremes for cities in tropical
developing countries – A review
Presentation outline
Background
Climate change impact assessment on urban hydrology
Overview of downscaling methods
Methodological suitability in tropical developing country cities
Conclusions
2
‘Top-Down’ climate impact assessment framework
3
Increasing
envelope of
uncertainty
Based on Wilby & Dessai (2010); Onof et al 2009; Sunyer et al 2012, Kendon et al. 2012
150 - 400 km monthly
(1.5) – 50 km daily
1 – 5 km 5 – 15 minutes
Energy & Land use
Driving forces
(population, income, lifestyle, technology)
Impact models (e.g.
urban flood models)
Regional Climate
Models (RCM)
Response options
Emission scenarios
General circulation
Models (GCM)
Spatial scale Temporal scale
Statistical downscaling
Climate change impact assessment at an urban scale
4
Willems et al. 2012
Overview of downscaling techniques
5
Statistical downscaling
• Use empirical-based relationships to convert course scale climate model
outputs to finer urban scales
Temporal downscaling
Spatial downscaling
• Key assumptions:
Local scale climate variables = f (large scale atmospheric variables)
Function can be deterministic or stochastic
Ratio remains unchanged under climate change
6
Delta change (Change Factor) methods
• Used to quantify changes in rainfall frequencies and intensities between a control and a future period for specified return periods
• Computed as a ratio of future to control rainfall intensity statistics
𝐶𝐹 =𝑆𝑡𝑅𝐶𝑀𝑓𝑢𝑡
𝑆𝑡𝑅𝐶𝑀𝑐𝑜𝑛 (1)
𝑆𝑡𝐹𝑢𝑡= 𝑆𝑡𝑂𝑏𝑠 ∙ 𝐶𝐹 (2)
Where StRCMfut RCM results for future period
StRCMcon RCM results for control period
StObs Observed statistics
7
Continuous versus event based change factors
8
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1 2 3 4 5 6 7 8 9 10 11 12
Ch
an
ge
F
ac
tor
Month 2030 Ensemble Average 2050 Ensemble Average
2080 Ensemble AverageFigure 2: Example of historical 30 year 1-hour EULER II design
storm for Wuppertal (OBS) and downscaled version based on
future climate model projections (ECHAM5 and HADCM3
denoted as ECH and HAD respectively) (Olsson et al., 2012)
(Event based case)
Figure 1: Example of monthly change factors computed from an
ensemble of regional climate models for Kampala for future periods
2001-2030, 2041-2070 and 2071-2100 against a control period of
1961-1990 (Continuous case )
Merits
Easy and quick to apply
Preserves characteristics of observed data
Only relative changes transferred from climate
model data to observed time series
Demerits
Deterministic
Dependent on GCM/RCM model reliability
Requires equivalent climate model and
observed data
Uncertainties
Range of computed change factors
Uncertainty in CFs for February: 0.94 - 2.52
9
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0 1 2 3 4 5 6 7 8 9 10 11 12
Ch
an
ge
F
ac
tor
Month
Monthly rainfall change factors for Kampala Control period (1961-1990): Future Period (2071-2080), Scenario RCP 4.5