william crosson, ashutosh limaye, charles laymon national space science and technology center...

24
William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using Soil Moisture Retrievals Using C- and C- and X-band Radiometer Observations X-band Radiometer Observations in SMEX02: Successes and in SMEX02: Successes and Challenges Challenges

Upload: gregory-heath

Post on 13-Jan-2016

221 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

William Crosson, Ashutosh Limaye, Charles Laymon

National Space Science and Technology CenterHuntsville, Alabama, USA

Soil Moisture Retrievals Using C- andSoil Moisture Retrievals Using C- andX-band Radiometer Observations in X-band Radiometer Observations in

SMEX02: Successes and ChallengesSMEX02: Successes and Challenges

Page 2: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Motivation and ObjectiveMotivation and ObjectiveMotivation: Data from NASA’s Advanced Microwave Scanning Radiometer (AMSR) are being used to estimate near-surface soil moisture. Analysis of the retrieval algorithms is needed to better understand the sensitivity of retrieved soil moisture to model parameters.

Objective: To utilize C- and X-band horizontally polarized brightness temperature data collected from an aircraft platform during the SMEX02 field experiment to evaluate the feasibility and limitations of soil moisture retrieval under conditions of significant vegetation biomass.

Page 3: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Soil Moisture Experiments in 2002 (SMEX02)Soil Moisture Experiments in 2002 (SMEX02)Location: Near Ames, Iowa Time: 24 June – 13 July 2002

Regional Area: ~5000 km2

Walnut Creek Watershed Area: ~ 400 km2

• 31 ground sampling sites for measuring gravimetric soil moisture, surface and soil temperatures (daily, AM), and vegetation properties (~weekly)• Surface energy flux stations, lidar, and radiosonde measurements• Aircraft microwave remote sensing data

Iowa Walnut CreekWatershed

Soybeans (40% of area)Corn (50% of area)

Page 4: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Polarimetric Scanning Radiometer (PSR)Polarimetric Scanning Radiometer (PSR)

10 km

Platform: NASA P-3BAltitude: 7300 m regional, 1800 m Walnut Creek Watershed

Incidence angle: 55o

3 dB footprint resolution (watershed): 750 m (C), 500 m (X)C-band frequencies (GHz):5.82 - 6.15, 6.32 - 6.65, 6.75 - 7.10, 7.15 - 7.50X-band frequency (GHz): 10.60 – 10.80

Walnut Creek Watershed coverage

PSR TB measurements

Corn

Soybeans

Page 5: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Daily mean volumetric soil moisture based on field site gravimetric ‘scoop’ measurementswith 2*standard error bars

Vegetation water content based on in situ destructive sampling

Vegetation Water ContentBased on in situ Measurements

0

1

2

3

4

5

6

176 178 180 182 184 186 188 190 1922002 Day of Year

kg

/m2

Corn

Soybeans

0.0

0.1

0.2

0.3

0.4

175 177 179 181 183 185 187 189 191 1932002 Day of Year

VW

C

PSRdays

Daily Soil Moisture and Vegetation Water ContentDaily Soil Moisture and Vegetation Water Content

Page 6: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Soil Moisture Retrieval AlgorithmSoil Moisture Retrieval Algorithm

VolumetricSoil Moisture

VolumetricSoil Moisture

Smooth SurfaceReflectivity

Smooth SurfaceReflectivity

EffectiveDielectricConstant

EffectiveDielectricConstant

VegetationTransmissivity

VegetationTransmissivity

SurfaceTemperature

SurfaceTemperature

SoilProperties

SoilProperties

BrightnessTemperature

BrightnessTemperature

Land CoverType

Land CoverType

SoilTemperature

SoilTemperature

Rough BareSurface Reflectivity

Rough BareSurface Reflectivity

SurfaceRoughness

SurfaceRoughness

Vegetated?Vegetated?

Yes

Apparent ReflectivityApparent Reflectivity

Effective TemperatureEffective TemperatureNo

Veg WaterContent

Veg WaterContent

Parameterizations:

Dobson Mixing Model

Choudhury et al. (1979) surface roughness correction

Choudhury et al. (1982) effective temperature

Jackson and Schmugge (1991) vegetation correction

Fresnel Reflectance

Page 7: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Experiment DesignExperiment Design

• Retrieve soil moisture: Using input data characterizing mean conditions over the SMEX02 Walnut Creek Watershed on each of 7 days, apply the retrieval algorithm for various points across a 3-dimensional parameter space (surface roughness, vegetation B parameter and single scattering albedo)• Validate: Using observed soil moisture conditions for each day, determine the parameter combinations that yield soil moisture estimates within a defined error range about the observed mean

Inputs representing mean conditions for given day:• Brightness temperature – Mean of PSR C- or X-band ‘postings’ • Surface and soil temperatures – Mean of observed temperatures • Vegetation water content – Mean of values measured at field sites (corn and soybeans), interpolated between ~weekly measurements

Page 8: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Veg-B 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.500.070.090.110.130.150.170.190.210.230.25

Parameter Space ComparisonParameter Space ComparisonC-band, H-PolC-band, H-Pol

Single Scattering Albedo = 0.03Single Scattering Albedo = 0.03

Shaded regions indicate combinations of surface roughness and vegetation B parameter that produce soil moisture values within 2 standard errors of the observed mean for the given day. The observed mean is for the 0-1 cm layer, intended to represent the effective C-band emitting depth. There is no overlap in valid soil moisture retrievals for July 9 and July 11.

July 1/July 11 (Wet) Overlap

July 1 (Very Dry) July 1/July 9 (Moderately Wet) Overlap

Roughness (cm)

Page 9: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Veg-B 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.500.070.090.110.130.150.170.190.210.230.25

Parameter Space ComparisonParameter Space ComparisonX-band, H-PolX-band, H-Pol

Single Scattering Albedo = 0.03Single Scattering Albedo = 0.03

Shaded regions indicate combinations of surface roughness and vegetation B parameter that produce soil moisture values within 2 standard errors of the observed mean for the given day. The observed mean is for the 0-1 cm layer, intended to represent the effective X-band emitting depth.

July 1/July 11 (Wet) Overlap

July 1 (Very Dry)

July 1/July 9 (Moderately Wet) Overlap

Roughness (cm)

All Days Overlap

Page 10: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to Surface RoughnessSensitivity of Retrieved VWC to Surface RoughnessC-band, H-polC-band, H-pol

SS Albedo = .03, Vegetation B = 0.16SS Albedo = .03, Vegetation B = 0.16

The range of roughness values for which retrieved soil moisture is within .04 of the observed values for each day is represented here by the horizontal bars.

0.0

0.1

0.2

0.3

0.4

0.5

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.50Roughness (cm)

VW

C

July 1July 11Observed Mean VWC +/- .04

Page 11: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to Surface RoughnessSensitivity of Retrieved VWC to Surface RoughnessC-band, H-polC-band, H-pol

SS Albedo = .03, Vegetation B = 0.16SS Albedo = .03, Vegetation B = 0.16

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.50Roughness (cm)

VW

C

June 25 June 27July 1 July 8July 9 July 11July 12

Soil moisture retrievals for wet, moderate and dry days have distinctly different sensitivity to roughness.

Page 12: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to Surface RoughnessSensitivity of Retrieved VWC to Surface RoughnessC-band, H-polC-band, H-pol

SS Albedo = .03, Various B Parameter ValuesSS Albedo = .03, Various B Parameter Values

0.0

0.1

0.2

0.3

0.4

0.5

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.50Roughness (cm)

VW

C

B = 0.07B = 0.10B = 0.13B = 0.16

Page 13: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to Surface RoughnessSensitivity of Retrieved VWC to Surface RoughnessX-band, H-polX-band, H-pol

SS Albedo = .03, Vegetation B = 0.16SS Albedo = .03, Vegetation B = 0.16

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40Roughness (cm)

VW

C

June 27 July 1

July 9 July 11

Page 14: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to VegetationSensitivity of Retrieved VWC to VegetationC-band, H-polC-band, H-pol

SS Albedo = .03, Roughness = 0.05 cmSS Albedo = .03, Roughness = 0.05 cm

0.0

0.1

0.2

0.3

0.4

0.5

0.07 0.10 0.13 0.16 0.19 0.22 0.25Vegetation B Parameter

VW

C

July 1July 11Observed Mean VWC +/- .04

The horizontal bars indicate the range of Vegetation B parameter values for which the retrieved soil moisture is within .04 VWC of the observed value.

Page 15: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to Single Scatter AlbedoSensitivity of Retrieved VWC to Single Scatter AlbedoC-band, H-polC-band, H-pol

Vegetation B = 0.16, Roughness = 0.05 cmVegetation B = 0.16, Roughness = 0.05 cm

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.00 0.03 0.06

Single Scatter Albedo

VW

C

June 25June 27July 1July 8July 9July 11July 12

Wet

Dry

Page 16: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to Surface RoughnessSensitivity of Retrieved VWC to Surface RoughnessC-band, V-polC-band, V-pol

SS Albedo = .03, Vegetation B = 0.07SS Albedo = .03, Vegetation B = 0.07

0.0

0.1

0.2

0.3

0.4

0.5

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.50Roughness (cm)

VW

C

July 1July 11Observed Mean VWC +/- .04

Page 17: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to VegetationSensitivity of Retrieved VWC to VegetationC-band, V-polC-band, V-pol

SS Albedo = .03, Roughness = 0.05 cmSS Albedo = .03, Roughness = 0.05 cm

0.0

0.1

0.2

0.3

0.4

0.5

0.07 0.10 0.13 0.16 0.19 0.22Vegetation B Parameter

VW

C

July 1July 11Observed Mean VWC +/- .04

Page 18: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Sensitivity of Retrieved VWC to Single Scatter AlbedoSensitivity of Retrieved VWC to Single Scatter AlbedoC-band, V-polC-band, V-pol

Vegetation B = 0.07, Roughness = 0.05 cmVegetation B = 0.07, Roughness = 0.05 cm

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.00 0.03 0.06

Single Scattering Albedo

VW

C

June 25 June 27July 1 July 8July 9 July 11July 12

Page 19: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Soil Moisture Retrievals UsingSoil Moisture Retrievals UsingConstant ParametersConstant Parameters

C-band, H-polC-band, H-pol

0.0

0.1

0.2

0.3

0.4

0.5

176 178 182 189 190 192 1932002 Day of Year

VW

C

B = .16, SS Albedo = .03, Roughness = 0.05 cmB = .13, SS Albedo = .03, Roughness = 0.40 cmB = .13, SS Albedo = .06, Roughness = 0.50 cmObserved Mean VWC +/- .04

Page 20: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Soil Moisture Retrievals UsingSoil Moisture Retrievals UsingConstant ParametersConstant Parameters

C-band, V-polC-band, V-pol

0.0

0.1

0.2

0.3

0.4

0.5

0.6

176 178 182 189 190 192 1932002 Day of Year

VW

C

B = .07, SS Albedo = .03, Roughness = 0.05 cmB = .07, SS Albedo = .00, Roughness = 0.05 cmB = .10, SS Albedo = .00, Roughness = 0.10 cmObserved Mean VWC +/- .04

Page 21: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Soil Moisture Retrievals UsingSoil Moisture Retrievals UsingConstant ParametersConstant Parameters

X-band, H-polX-band, H-pol

0.0

0.1

0.2

0.3

0.4

0.5

0.6

178 182 190 1922002 Day of Year

VW

C

B = .16, SS Albedo = .03, Roughness = 0.10 cmB = .16, SS Albedo = .03, Roughness = 0.20 cmB = .10, SS Albedo = .06, Roughness = 0.40 cmObserved Mean VWC +/- .04

Page 22: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

0.0

0.1

0.2

0.3

0.4

0.0 0.1 0.2 0.3 0.4Mean VWC

Ran

ge

of

B p

aram

eter

val

ues

0.0

0.1

0.2

0.3

0.4

0.5

0.0 0.1 0.2 0.3 0.4Mean VWC

Ran

ge

of

rou

gh

nes

s va

lues

(cm

)

July 1

July 9

July 8June 25

July 12

July 7

June 25

Acceptable Parameter Ranges asAcceptable Parameter Ranges asFunction of Soil MoistureFunction of Soil Moisture

C-band, H-polC-band, H-pol

Roughness = 0.05 cm

Vegetation B parameter = .16

Due to increasing sensitivity, valid parameter ranges tend to narrow with increasing VWC.

June 27

July 11

July 1July 9 July 12

July 11

June 27

July 8

Illustrations of the range of parameter values that produce retrievals within .04 of the observed VWC.

Single scattering albedo = .03.

Lower roughness values are indicated for wetter soils.

Page 23: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

Acceptable Parameter Ranges asAcceptable Parameter Ranges asFunction of Soil MoistureFunction of Soil Moisture

X-band, H-polX-band, H-pol

Due to increasing sensitivity, valid parameter ranges tend to narrow with increasing VWC. 0.0

0.1

0.2

0.3

0.4

0.0 0.1 0.2 0.3 0.4Mean VWC

Ran

ge

of

B p

aram

eter

val

ues

Roughness = 0.2 cmJuly 1

July 9 July 11June 27

0.0

0.1

0.2

0.3

0.4

0.5

0.0 0.1 0.2 0.3 0.4Mean VWC

Ran

ge

of

rou

gh

nes

s va

lues

(cm

)

July 1

July 9

Vegetation B parameter = .16

June 27

July 11

Illustrations of the range of parameter values that produce retrievals within .04 of the observed VWC.

Single scattering albedo = .03.

Lower roughness values are indicated for wetter soils.

Page 24: William Crosson, Ashutosh Limaye, Charles Laymon National Space Science and Technology Center Huntsville, Alabama, USA Soil Moisture Retrievals Using C-

ConclusionsConclusions

• Using a single-frequency, single-polarization soil moisture retrieval algorithm, there is no parameter vector (surface roughness, vegetation B parameter, single scattering albedo) that produces soil moisture values for aggregate corn and soybean fields that are within two standard errors of the observed values for each of the five days for which complete PALS L-band coverage is available.

• For three days that are relatively wet, excellent retrievals were obtained with a single parameter vector. For each of the two dry days, different sets of parameter vectors were required to produce acceptable results.

• The inability to obtain accurate soil moisture retrievals with a single parameter vector does not appear to be related to errors in brightness temperatures.

• Under wet conditions, retrieved soil moisture for corn is seen to be very sensitive to all three parameters. For dry conditions, sensitivity is lower, but not negligible.