Glacio-hydrological projections

with downscaled climate data

Megumi Watanabe and Shinjiro Kanae

Tokyo Institute of Technology

2018/02/08

1

Current assessmentsand

our objective

2

Current assessmentsfor impact of climate change

3

(NATIONAL INTEGRATED WATER RESOURCES MANAGEMENT PLAN 2016)

Projected changes in precipitation for RCP4.5

Rainfall

River

discharg

e

Basin outflows and their return periods

To estimate of river discharge taking into

account glacier melt with a regional climate

projection

OBJECTIVE

Research flow

4

5

Hydrological model

Research flow

Energy balance glacier

model with debris effect

Temperature index

glacier model

River discharge

Case1 Case2

Runoff from glaciers (Case1,2)

(Case1,2)

Initial glacier data

Multiple climate data at high elevations

Precipitation, air temperature and etc.

6

Hydrological model

Research flow

Energy balance glacier

model with debris effect

Temperature index

glacier model

River discharge

Case1 Case2

Runoff from glaciers (Case1,2)

(Case1,2)

Initial glacier data

Multiple climate data at high elevations

Precipitation, air temperature and etc.

7

Hydrological model

Research flow

Energy balance glacier

model with debris effect

Temperature index

glacier model

River discharge

Case1 Case2

Runoff from glaciers (Case1,2)

(Case1,2)

Initial glacier data

Multiple climate data at high elevations

Precipitation, air temperature and etc.

8

Hydrological model

Research flow

Energy balance glacier

model with debris effect

Temperature index

glacier model

River discharge

Case1 Case2

Runoff from glaciers (Case1,2)

(Case1,2)

Initial glacier data

Multiple climate data at high elevations

Precipitation, air temperature and etc.

9

Hydrological model

Research flow

Energy balance glacier

model with debris effect

Temperature index

glacier model

River discharge

Case1 Case2

Runoff from glaciers (Case1,2)

(Case1,2)

Initial glacier data

Multiple climate data at high elevations

Precipitation, air temperature and etc.

Past period• Uncertainty

• Data

- Air temperature

- Precipitation

Future period (GCMs)• Uncertainty

• Bias correction

• Multi-model

Uncertainty of climate datafor the past period

The number of gauge (APHRODITE)

The scarcity of in-situ observations

(for temperature and precipitation)

at high elevations

10

Temperature data for the past TA1 TA2

(Hirabayashi et al., 2008)

ERA-Interim

Thermometer

Reanalysis

It could be applied to

sparsely observed regionsRennie et al. (2014)

Sparse

stations

at

elevations

H08

Hybrid of observations and model

(Dee et al., 2012)

https://serc.carleton.edu/eet/envisioningcli

matechange/part_2.html

http://www.ushistory.org/franklin/fun/thermometer.htm

Precipitation data for the pastPR1 PR2

PR3 PR4

©JAXA

APHRODITE Sakai

MSWEP MSWEP+PR

©JAXA

Gauge Gauge

Gauge Satellite Reanalysis Gauge Satellite Reanalysis

(This study)

http://www.stuffintheair.com/rain-gauge.html

Precipitation data for the pastPR1 PR2

PR3 PR4

©JAXA

APHRODITE Sakai

MSWEP MSWEP+PR

©JAXA

©NASA

GaugeGauge

Gauge Satellite Reanalysis Gauge Satellite Reanalysis

(This study)

Inverse

estimation

using

discharge

Inverse

estimation

using

glacier

elevation

Directly

detect rain

drop using

satellite radar

Precipitation data for the pastPR1 PR2

PR3 PR4

©JAXA

APHRODITE Sakai

MSWEP MSWEP+PR

©JAXA

GaugeGauge

Gauge Satellite Reanalysis Gauge Satellite Reanalysis

(This study)

Inverse

estimation

using

discharge

Inverse

estimation

using

glacier

elevation

Directly

detect rain

drop using

satellite radar

15PR2 Sakai et al. (2015)

Gauge

Satellite

derived

glacier

elevation

Correct

(Nuimura et al., 2015)

Inversely estimate

16

𝑷 = 𝑬 +𝑸 + ∆𝑺

P: Precipitation; Q: Observed discharge; E：Evaporation;

∆S: changes in water storage

PR3 MSWEP (Beck et al., 2015)

©JAXA

Multi

CorrectGauged

discharge

http://www.bafg.de/GRDC/E

N/02_srvcs/21_tmsrs/riverdis

charge_node.html

Global Runoff Data Centre

Precipitation data for the pastPR1 PR2

PR3 PR4

©JAXA

APHRODITE Sakai

MSWEP MSWEP+PR

©JAXA

GaugeGauge

Gauge Satellite Reanalysis Gauge Satellite Reanalysis

(This study)

Inverse

estimation

using

discharge

Inverse

estimation

using

glacier

elevation

Directly

detect rain

drop using

satellite radar

18

MicrowaveInfrared Radar (PR)

High emissivity

from the ground

Relationship between

the cloud height and

precipitation

More sensitive

over land as

well as ocean

Yamamoto et al. (2011)

The peak local-time distribution of precipitation showed a

relationship with the topography in the order of precipitation

radar (strongest relationship), microwave radiometer, and

infrared products.

PR4 MSWEP + PR (This study)

©JAXA

19

Hydrological model

Research flow

Energy balance glacier

model with debris effect

Temperature index

glacier model

River discharge

Case1 Case2

Runoff from glaciers (Case1,2)

(Case1,2)

Initial glacier data

Multiple climate data at high elevations

Precipitation, air temperature and etc.

Past period• Uncertainty

• Data

- Air temperature

- Precipitation

Future period (GCMs)• Uncertainty

• Bias correction

• Multi-model

20

Projected change in

temperature for RCP4.5

(NIWRNP 2016)

Projected annual total

precipitation from CMIP5

GCMs (RCP8.5)

Spread

of

climate

models

Uncertainty of climate datafor the future

Climate models

Coarse spatial resolution & bias

Spread among models

Bias correction

Multi-model

21

“Original”

Projected air temperature in 2080-2010 by INM-CM4, RCP8.5

correctionBias

℃“Downscaled”Bias correction

1950 2000 2050 2100Year

Climate

model

Observation

Compare

statics

【Baseline period】 【Projection period】

Correct

statics

!! Can be different

depending on data

for the past !!

22

Eastern Himalaya

Various

climate

models

GHG emission levelScenarios

High

MidLow

ΔTemperature(℃) ΔSnowfall(%)(2070-2099 to 1950-1979)RCP2.6-8.5

CMIP5 GCMs

Multi-scenario, Multi-model

GCM1

GCM3

GCM2

High temperature

Median temperature & snowfall

More snowfall

23

Hydrological model

Research flow

Energy balance glacier

model with debris effect

Temperature index

glacier model

River discharge

Case1 Case2

Runoff from glaciers (Case1,2)

(Case1,2)

Initial glacier data

Multiple climate data at high elevations

Precipitation, air temperature and etc.

Initial glacier data

Initial glacier data

24

The latest glacier inventory“Randolph Glacier Inventory”• A globally complete inventory of

glacier outlines using modern satellite (such as Landsat or ASTER) imagery

• Version 6.0: released July 28, 2017.

• Information • Glacier shape• Location (latitude & longitude)• Glacier area• Altitude• Length

25

…

Glacier shape by RGI 6.0

Legend

glacier_BH

TM_WORLD_BORDERS-0.3

n25e090_dem.tif

-12.17773819 - 1,000

1,000.000001 - 2,000

2,000.000001 - 3,000

3,000.000001 - 3,500

3,500.000001 - 4,000

4,000.000001 - 4,500

4,500.000001 - 5,000

5,000.000001 - 5,500

5,500.000001 - 6,000

6,000.000001 - 7,557.027832

n25e085_dem.tif

-5.14632225 - 1,000

1,000.000001 - 2,000

2,000.000001 - 3,000

3,000.000001 - 3,500

3,500.000001 - 4,000

4,000.000001 - 4,500

4,500.000001 - 5,000

5,000.000001 - 5,500

5,500.000001 - 6,000

6,000.000001 - 8,839.171875

Legend

glacier_BH

TM_WORLD_BORDERS-0.3

n25e090_dem.tif

-12.17773819 - 1,000

1,000.000001 - 2,000

2,000.000001 - 3,000

3,000.000001 - 3,500

3,500.000001 - 4,000

4,000.000001 - 4,500

4,500.000001 - 5,000

5,000.000001 - 5,500

5,500.000001 - 6,000

6,000.000001 - 7,557.027832

n25e085_dem.tif

-5.14632225 - 1,000

1,000.000001 - 2,000

2,000.000001 - 3,000

3,000.000001 - 3,500

3,500.000001 - 4,000

4,000.000001 - 4,500

4,500.000001 - 5,000

5,000.000001 - 5,500

5,500.000001 - 6,000

6,000.000001 - 8,839.171875

Legend

glacier_BH

TM_WORLD_BORDERS-0.3

n25e090_dem.tif

-12.17773819 - 1,000

1,000.000001 - 2,000

2,000.000001 - 3,000

3,000.000001 - 3,500

3,500.000001 - 4,000

4,000.000001 - 4,500

4,500.000001 - 5,000

5,000.000001 - 5,500

5,500.000001 - 6,000

6,000.000001 - 7,557.027832

n25e085_dem.tif

-5.14632225 - 1,000

1,000.000001 - 2,000

2,000.000001 - 3,000

3,000.000001 - 3,500

3,500.000001 - 4,000

4,000.000001 - 4,500

4,500.000001 - 5,000

5,000.000001 - 5,500

5,500.000001 - 6,000

6,000.000001 - 8,839.171875

< 1,000

1,000-2,000

2,000-3,000

3,000-3,500

3,500-4,000

4,000-4,500

4,500-5,000

5,000-5,500

5,500-6,000

> 6,000

Altitude (m)

ⒸNASALandsat

26

Hydrological model

Research flow

Energy balance glacier

model with debris effect

Temperature index

glacier model

River discharge

Case1 Case2

Runoff from glaciers (Case1,2)

(Case1,2)

Initial glacier data

Multiple climate data at high elevations

Precipitation, air temperature and etc.

Temperature index glacier model

27

28

Melt factor

[mm/℃/day]

Surface-air

temperature [℃] 0 [℃]

Snow pack is transformed into glacier ice

MeltingSnowfall

Accumulation Ablation

𝑻𝒊 + 𝒅𝑻 − 𝑻𝟎 𝑫𝑫𝑭

Surface air

temperaturePrecipitation

Glacier model –mass balance-Glacier

Snowfall = Precipitation*Cp

(if 𝑻𝒊 + dT 2)

(Hirabayashi et al., 2013)

Adjustment

factor

Summary

29

Today’s summary

• Multiple climate data for the past period• Air temperature (In-situ / Reanalysis)

• Precipitation (In-situ / Reanalysis / Inverse estimations)

• Climate data for the future period • Bias correction of GCMs

• Multi-GCMs

• Initial glacier data from the inventory

• Temperature index glacier model

• Uncertainty range of climate data

• Uncertainty range of glacier projections

30