Urban Water Security Research Alliance Climate and Water over South - - PowerPoint PPT Presentation

Climate and Water over South East Queensland: Past and Future Wenju Cai

Climate and Water

Science Forum, 19-20 June 2012

Urban Water Security Research Alliance

Considering Climate Change

The current SEQ Water Strategy considers climate change as a change to the maximum yield that can be delivered while meeting the pre-determined Level

• f Service (LOS) performance objectives.

Figure sourced from SEQ Water Strategy

Time sequence of SEQ water storage level (1890-2009)

Research Questions …

• Research question 1: What causes drought in the

SEQ region, and how the properties of drought (intensity, duration and frequency) may change in a warming climate, in addition to forcing by multidecadal variability (in conjunction with QCCCE).

• Research question 2: How best to translate global

projections of climate change (rainfall and other hydrological fields) into future water availability information (in collaboration with DERM).

• Research question 3: How sensitive to the

resolution of climate information is the SEQ stream- flow model (IQQM).

Most models do not produce the observed trends

Cai et al. 2010: Asymmetry in ENSO Teleconnection with Regional Rainfall, Its Multidecadal Variability, and Impact.

• J. Climate, 23, 4944–4955

2011

SEQ summer rainfall

El Niño & La Niña Index

NINO-3.4

NINO3.4 NINO3.4 ≥ ≥ 0.5 0.5° °C for 5 months= El Ni C for 5 months= El Niñ ño

• NINO3.4

NINO3.4 ≤ ≤

• 0.5

0.5° °C for 5 months= La Ni C for 5 months= La Niñ ña a Courtesy of McPdaden

Cai et al. 2010

The nonlinearity oscillates on decadal time scale

El Niño La Niña

El Niño La Niña El Niño La Niña ?

Positive PDO El Nino La Nina Negative PDO El Nino La Nina

Cai et al. 2010

Cai and van Rensch (2012) The 2011 southeast Queensland extreme summer rainfall: a confirmation of a negative Pacific Decadal Oscillation phase?

The flood constitutes a confirmation of the PDO changing to a negative phase The result implies an increased chance of high summer rainfall events over the region during La Niña events in the upcoming decade or so.

95% significance Really becoming negative

Model Details

The hydrologic model used in this study covers the Brisbane River catchment upstream of the tidal limit Historic rainfall and evaporation data is sourced from SILO Urban demands are set to full utilisation of existing entitlements, approximately 280 GL/a, with no restriction rules in place The model simulation period is from 1889 to 2000

ESK OAKEY BOONAH GATTON KILCOY IPSWICH LAIDLEY NAMBOUR BUDERIM NANANGO BRISBANE WOODFORD KINGAROY CALOUNDRA TOOWOOMBA CABOOLTURE BEAUDESERT CROWS NEST PITTSWORTH DECEPTION BAY

STUDY AREA

Wivenhoe Dam Somerset Dam

ESK OAKEY BOONAH GATTON KILCOY IPSWICH LAIDLEY NAMBOUR BUDERIM NANANGO BRISBANE WOODFORD KINGAROY CALOUNDRA TOOWOOMBA CABOOLTURE BEAUDESERT CROWS NEST PITTSWORTH DECEPTION BAY

STUDY AREA

Wivenhoe Dam Somerset Dam

Climate Change Models Analysed

Global Circulation Model (GCM) Cubic Conformal Atmospheric Model (CCAM) downscaling Model Emissions Scenario 1 Model Emissions Scenario 1 CSIRO Mark 3.0 A2 CSIRO Mark 3.0 A2 CSIRO Mark 3.5 A2 CSIRO Mark 3.5 A2 GFDL CM 2.1 A2 GFDL CM 2.1 A2 MIROC 3.2 (medres) A2 MIROC 3.2 (medres) A2 MPI ECHAM 5 A2 MPI ECHAM 5 A2 NCAR CCSM 3.0 A2 NCAR CCSM 3.0 A2 UKMO HADCM3 A2 UKMO HADCM3 A2 UKMO HADGEM1 A2 UKMO HADGEM1 A2 IAP FGOALS 1.0 A1B INMCM 3.0 A2, A1B MIROC 3.2 (hires) A1B CSIRO Mark 3.0 A1B MIUB echo A2 NCAR CCSM 3.0 A1B

8 Directly Comparable 6 Additional Models

1 All model projections were scaled to Emissions Scenario A1FI

Spatial Variation of Climate Projections

• Spatial variation in downscaled projections is more pronounced for

rainfall than for evaporation

Monthly percentage rainfall change for downscaled CSIRO Mk 3.5 (for year 2050 SRES A1FI) (source Kent, D., CSIRO) Monthly percentage evaporation change for downscaled CSIRO Mk 3.5 (for year 2050 SRES A1FI) (source Kent, D., CSIRO)

Downscaling Effects on Rainfall

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Month

40 80 120 160

Mean Monthly Rainfall (mm) Mean Monthly Rainfall - Subcatchment UB1 (07/1889-06/2000)

Global Circulation Model Generation Method (GCM)

• Downscaled models display greater variability in rainfall projections than

GCMs, but generally result in reduced wet season rainfall

Downscaling Effects on Evaporation

• Downscaled models display greater variability in evaporation projections

and are generally higher than the original GCM projection

• Some of this difference may be attributed to the method of calculating the

evaporation change within CCAM

• not all GCMs produced evaporation output (ie. evaporation was derived from

temperature relationships)

Mean Monthly Evaporation (mm) Mean Monthly Evaporation (mm)

Downscaling Effects on Runoff Generation

• The combination of increased evaporation and decreased wet season

rainfall results in significantly reduced runoff generation/inflow to the hydrologic model (IQQM)

Climate Change Effect on Storage Behaviour (GCM)

Date

500000 1000000 1500000 2000000 2500000

Combined Simulated Storage Volume

Somerset and Wivenhoe Dams

Global Circulation Model Generation Method (GCM)

DSV 75% 50% 25%

Several GCMs result in increased storage volume

Downscaling Effects on Storage Behaviour (CCAM)

Storage Volume (ML)

1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

All but one downscaled GCM result in decreased storage volumes

The system does not recover from the federation drought until 1927!

Comparison of Median GCM and CCAM Projections

The combined effect of reduced rainfall and increased evaporation cause the gradual decline in storage volume during these two periods

No Rainfall Change Results (GCM)

No Rainfall Change Results (CCAM)

Urban Water Security Research Alliance THANK YOU www.urbanwateralliance.org.au

Raw Pan evaporation (epan_ave) extracted from CCAM