Holiday Inn, Arlington, VA (Dogwood room), 25-26 June 2009

Li Li and Peter Gaiser Naval Research Laboratory

Washington, DC, USA

With contributions fromDrs. S. Seneviratne, G. Nedoluha, and S. Steele-Dunne

Joint AMSR Science Team meeting

Observations of the 2003 European Heat Waves Using Soil Moisture and Vegetation Data

2

2003 European Heat Waves

Heat waves are the most lethal type of weather phenomenon

• One of the hottest summers on record in Europe. • 35,000 people died as a result of the heat wave. • Uninsured crop loss totaling $12.3 Billions

Extreme Heat Wave MechanismsRoles of Soil Moisture, Vegetation, and LST

3

Energy Balance.

Rn - G = Le + H = Qt

Rn : net radiation available at surfaceG : heat flux into the soilLe : latent heat fluxH : sensible heat fluxQt : total turbulent energy flux into the PBL

Bowen Ratio (H/Le)

Negative ForcingOn Precipitation

Drying of Soil Moisture

Soil Moisture – Precipitation Feedback(By Dr. B.F. Zaithik)

• Reduction in Precipitation Recycling

• Deepening of daytime of PBL (Planetary boundary layer) with increased entrainment of hot/dry air

• Reduction in net radiation (Rn) and

Qt

By P. Houser/GMU

2003 European Heat wave:Land-Atmosphere Interaction Modeling

• Initiated by anticyclonic atmospheric circulations

• Low soil moisture induced a positive feedback mechanism between surface conditions and atmospheric circulation.

– Strong positive radiative anomalies and a large precipitation deficit in the months preceding the extreme event contributed to a rapid loss of soil moisture

– Low soil moisture resulted in strongly reduced evapotranspiration and latent cooling

4

Climate High-Resolution

Model (CHRM) Simulations

Temperature anomaly due to soil water perturbation.

Left panel: Dry run vs. control run.

Right panel: Wet run vs. control run.Courtesy of Drs. E. Fischer and S. SeneviratneInstitute of Atmospheric and Climate Science ETHZurich, Switzerland

5

WindSat Land Algorithm & Validation

j j

xyy

jj

sjmjxLH 2

2

2

)}({221 )2log()(

Physical Retrieval Algorithm:

• Six channel at 10, 18 and 37 GHz• Simultaneous retrievals of soil moisture,

vegetation water content, and LST

Validation Strategy: Comparisons of multiple data sources

• Soil Moisture Climatology– Climate Regime– Global precipitation climatology

• Ground In-Situ– Micronet/SMEX – JAXA, NAFE

• Precipitation NEXRAD, AVHRR

• Vegetation Dynamics– GVF/AVHRR– NDVI/NDII/MODIS

Ancillary data

Resample to Earth-fixed Grid

Soil Moisture and Vegetation Retrieval

No

Land EDR

Land QC

Water body/OceanSnow/Ice, Frozen SoilRain, Mountain, RFI?

PRT Simulations

No

PRT Converge?

Yes

WidSat SDR

Soil Moisture

Vegetation

Land SurfaceTemperature

6

Soil Moisture and Precipitation Climatology

• Global soil moisture patterns are consistent with dry/wet patterns of climate regimes.

– All the major deserts– US continental climate– high moisture in the subarctic regions– consistent with major Monsoon systems

• Progression of ITCZ in West Africa– Consistent with West Africa monsoon– Shows asymmetrical wave form

Northward progression of ITCZ

Soil Moisture In-Situ Validations

USDA Validation Sites:SMEX03: rangeland and winter wheat SMEX04: sparse shrubland, mountainsSMEX05: agricultural – corn, soybean

Other Validation Sites:SMOSREX, France: » Grassland, wheat, corn CEOP, Mongolia: » Arid rangeland

France Mongolia

United States

8

Vegetation Water Content

The WindSat vegetation water content is consistent with the AVHRR NDVI data.

GVF

Vegetation Data Validation

The Normalized Difference Vegetation Index (NDVI): N = (R0.85-R0.66)/(R0.85+R0.66)

Green Vegetation Fraction: GVF = (N-Ns)/(Nv-Ns)

SMEX05 Vegetation Validation

European Summer Climate

9

• Potentially influenced by SST and SM, when the large-scale westerly flow weakens in the summer

• Drought propagation play an important role in maintaining the hot summer

• Using precipitation data as proxy; no direct SM evidence published By Vautard et al, 2007

Soil Moisture

WindSat retrievals and anomalies associated with 2003 European heat wave

10

During early Spring•Warmer spring temperature•Normal soil moisture level •Rapid vegetation green-up

During Spring•Soil moisture was below average •VWC was above the average until around day 150-160, depleting soil moisture

During the Summer•When the heat wave hits near day 220, both soil moisture and VWC were low.

The 2003 European Extreme Heat Wave

11

WindSat Data Captured three reported LST anomalies, and revealed their differences in surface conditions

Day 60 – 110• Warmer spring temperature• Normal soil moisture level • Rapid vegetation green-up

Day 110 - 160• Soil moisture was below average • VWC was above the average until day 150-160,

depleting soil moisture

Late Summer• When the heat wave hits near day 220, both soil

moisture and VWC were low.

LST Soil Moisture

Week 232003

Week 235 year mean

SMOSREX Site

Spatial & Temporal Variations of the Anomalies

12

13

Conclusions

• The passive microwave land observations can:– Capture anomalies of soil moisture and vegetation water

content – provide a detailed description of the evolution of 2003

European heat waves.– corroborate the climate model simulations suggesting an

important role of drought in the deadly 2003 heat wave.

14Summer Rains

Dry Winter

Rainy Season

Vegetation Water Content

Green Vegetation Fraction

The WindSat vegetation water content is consistent with the Green Vegetation Fraction derived from AVHRR data.

WindSat Vegetation Data Validation

Brazil Cerrado

Dr. Liang/NOAA/NESDIS

SMEX05 Vegetation Validation

The Normalized Difference Infrared Index (NDII) is

linearly related to canopy water (EWT):NDII = (R0.85-R1.65)/(R0.85+R1.65) Assuming linear allometric relationships, the VWC is then

linearly related to EWT and NDII

The first comparison of large scale Vis/IR and MW Vegetation data