Hydrometeorological ensemble forecasts for the 28 September 2012 (IOP8) extreme flash-flood in Murcia, Spain A. Amengual and V. Homar Grup de Meteorologia, Departament de Física, Universitat de les Illes Balears, Palma, Mallorca, Spain e-mail: [email protected]Universitat de les Illes Balears Reunión PREDIMED 2014A 5-6 Junio
Universitat de les Illes Balears. Reunión PREDIMED 2014A 5-6 Junio. Hydrometeorological ensemble forecasts for the 28 September 2012 (IOP8) extreme flash-flood in Murcia, Spain. A. Amengual and V. Homar. Grup de Meteorologia, Departament de Física, - PowerPoint PPT Presentation
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
Hydrometeorological ensemble forecasts for the 28 September 2012 (IOP8) extreme flash-
flood in Murcia, Spain
A. Amengual and V. Homar
Grup de Meteorologia, Departament de Física, Universitat de les Illes Balears,
Hydrometeorological ensemble forecasts for the 28 September 2012 (IOP8) extreme flash-flood in Murcia,
Spain
1. The Guadalentín flash-flood event
2. Hydrological and meteorological tools
3. Probabilistic versus deterministic QPFs
4. Probabilistic versus deterministic QDFs
5. Conclusions and further remarks
1. The Guadalentín flash-flood event: synoptic situation
• Entrance of a deep upper-level closed trough • Generation of a surface mesoscale cyclone • Advection of warm and moist air toward Almería and Murcia from the Mediterranean
Convergence zone between easterly advection and westerly low-level flow+ orographic enhancement
quasi-stationary mesoscale convective system
H500+T500+PV250
T850+SLP
27 September 2012 12 UTC 28 September 2012 12 UTC
• Torrential precipitation took place on 27, 28 and 29 September 2012
• Daily precipitation amounts: 214 mm in Andalucía, 240 mm in Murcia and 230 mm in Valencia
• The Guadalentín catchment is a medium size basin with an area of 3343 km2 and a length close to 121 km
• Accumulated rainfall in 8 h up to 214 mm inside the basin
• Peak discharges: - 616.3 m3s-1 in Lorca- 1081.2 m3s-1 in Paretón de Totana
1. The Guadalentín flash-flood event: observations
m m
40
50
60
75
100
125
150
175
200
225
(a)
1. The Guadalentín flash-flood event
• 10 casualties. Material losses estimated at about 120 M€
• Schemes: Microphysics ─ WSM6; Long-wave radiation ─ RRTM ; Short wave radiation ─ Dudhia; surface model ─ NOAH; time-step ─ 30 s
• The experiments consider a 48 h period simulation (27/09/2012 - 29/09/2012 00 UTC)
HEC-HMS model set-up
• Loss rate: Soil Conservation Service Curve Number (SCS-CN) model
• Transform: SCS Unit Hydrograph model
• Flow routing: Muskingum method
• Reservoirs: elevation-storage-outflow relationship + initial elevation of the water level
• The experiments consider a 72 h period simulation (27/09/2012-01/10/2012 00 UTC)
2. Hydrological and meteorological tools
0
250
500
750
1000
1250
1500
1750
2000
2500
3000
m eters (asl)
MEDITERRANEAN SEA
SPAIN
FRANCE
ITALY
ALGERIA
Pyrenees
Baetic system
Iberian system
M urciaCHS
Alps
Atlas
Alm eria
Balearic Islands
C atalonia
Valencia
Andalusia
Alboran sea
Gulf of Lyon
3. Probabilistic versus deterministic QPFs
Difficulties to correctly forecast precise location and timing of convectively-driven rainfall system affecting a medium size basin
-88.4
-5.7
EV (%)
-89.90.12control
-2.60.91rain-gauges
EP (%)NSEGuadalentín
0
10
25
50
75
100
125
150
175
200
225
(b) m m Paretón (2384.7 km2)
0
100
200
300
400
500
600
700
800
900
1000
1100
1200
27/09/201200:00
27/09/201212:00
28/09/201200:00
28/09/201212:00
29/09/201200:00
29/09/201212:00
30/09/201200:00
30/09/201212:00
01/10/201200:00
date
Q (m
3 s-1)
Q observed
Q rain-gauges
Q WRF control
(b) Paretón (2384.7 km2)
0
100
200
300
400
500
600
700
800
900
1000
1100
1200
27/09/201200:00
27/09/201212:00
28/09/201200:00
28/09/201212:00
29/09/201200:00
29/09/201212:00
30/09/201200:00
30/09/201212:00
01/10/201200:00
date
Q (m
3 s-1)
Q observed
Q rain-gauges
Q WRF control
(b)
Flow observations only available for this study case: perfect-model assumption. Optimal estimation of the initial conditions and dynamical formulation after calibration.
3. Probabilistic versus control QPFs
Mesoscale EPS (WRF)
• Diversity source only from IC/BC (dynamical downscaling)
• Obtained from ECMWF-EPS forecast (Global Singular Vectors)
• 50 equally-likely members
Study of the spatial and temporal uncertainties of QPFs into a medium-sized catchment
(d) Probability-matched ensemble mean
• WRF ensemble comprises 51 elements (control + 50 perturbed)
• Important spread on rainfall values
• Essential role of atmospheric dynamical forcing
0
10
25
50
75
100
125
150
175
200
225
(c) m m
0
10
25
50
75
100
125
150
175
200
225
(d) m m
(c) Ensemble mean ( in mm, shaded) and standard deviation (in mm, continuous line starting at 10 mm interval)
3. Probabilistic versus control QPFs
4. Probabilistic versus deterministic QDFs
• Elements of the HEPS are considered equally-like
• Cumulative distribution functions (CDFs) of driven runoff peak flows
observedcontro lensem bleensem ble m eanprobability-m atchedQ p(T=25yrs)Q p(T=35yrs)
Paretón (2384.7 km 2)(b)
5. Conclusions and further remarks
• WRF control simulation is deficient for the Guadalentín event: maximum precipitation amounts are obtained quite far away from the basin
• EPS reduce biases obtained for the control forecast
• For civil protection purposes, a hypothetical first warning for a peak flow exceeding Qp (T = 25 yrs) would have produced a probability of exceedence of 0.4 and 0.3 at Lorca and Paretón. This fact points out the benefits of a HEPS versus a deterministic prediction system • The performance of the hydrometeorological simulations strongly depends on the
initial conditions of the databases and on the case under study
• References:
Amengual et al. (2014): Hydrometeorological ensemble forecasts for the 28 September 2012 (IOP8) extreme flash-flood in Murcia,Spain. Quart. J. R. Meteorol. Soc [submitted]