Progress report on work at Progress report on work at the CHRIS-PROBA study the CHRIS-PROBA study site on Thorney Island, UKsite on Thorney Island, UK

Ted Milton, University of Southampton, UKTed Milton, University of Southampton, UK

Karen Anderson, University of Exeter, UKKaren Anderson, University of Exeter, UK

Location of Thorney IslandLocation of Thorney Island

CHRIS / PROBA : 16 Sept 2003

CHRIS PROBA data from Thorney IslandCHRIS PROBA data from Thorney Island

Date Comments

14th Sept 2003 CASI and ATM data collected two days later.

17th April 2004 Some cloud cover.

7th October 2004 Excellent quality.

14th September 2003

CHRIS PROBA data from Thorney IslandCHRIS PROBA data from Thorney Island

Date Comments

14th Sept 2003 CASI and ATM data collected two days later.

17th April 2004 Some cloud cover.

7th October 2004 Excellent quality.

14th September 2003

Ground spectra collected with a dual field-of-view GER1500 spectroradiometer and single beam ASD FieldSpec Pro

Measuring aerosol optical thickness using a Microtops sunphotometer.

Intertidal vegetation mappingIntertidal vegetation mapping

14th September 2003• Land cover survey during

summer 2004.• String of 3 images co-

registered using GCPs.• Per-pixel classification of

images from each view angle.

Intertidal vegetation mappingIntertidal vegetation mapping

14th September 2003• Land cover survey during

summer 2004.• String of 3 images co-

registered using GCPs.• Per-pixel classification of

images from each view angle.

• Atmospheric correction.• Per-pixel classification of

the multi-angle data set.• Soft classification /

parcel based methods.

QUASAR methodology (Teillet et al., 2001)

Atmospheric model Atmospheric model

Radianceat TOA

Average reflectance of ground calibration target

Atmospheric correction : FLAASH

CASI : 16th April 2003

ground

casiconcrete

ground

casiasphalt

Reflectance of concrete calibration targetReflectance of concrete calibration target

High precision dual-beam mobile spectroradiometerHigh precision dual-beam mobile spectroradiometer Two matched GER1500 spectroradiometersTwo matched GER1500 spectroradiometers

Long-term change in reflectance of bright targetLong-term change in reflectance of bright target

Afternoon spectra significantly lower reflectance, esp. in the visible region

Each sequence shows a small variation with solar zenith angle.Not large enough to account for the difference between am and pm.

Neither is the spatial uncertainty of location...

First clue to what is affecting the concrete reflectance is found in the met. data...

Onset of sea breeze in late morning is confirmed by wind direction measurements from a buoy in Chichester

harbour.

But how can a change in wind direction cause a statistically significant change in reflectance ?

Cutting a long story short ...

Estimating reflectance of bright targetEstimating reflectance of bright target

Towards a model relating concrete reflectance to Towards a model relating concrete reflectance to meteorological data.meteorological data.

Meteorologicaldata

Precipitation

Wind

Air temp.

Rel humidity

D:G ratio

Surface Moisture

Biotic factors

Surface temperature

Surface properties

Surface composition

NadirReflectance

Dew Point

Time sincerainfall

30th April 2004

9th Sept 2004

16th Sept 2004

22nd Nov 2004

17th Dec 2004

Vicarious Calibration siteRailroad Valley, Nevada

© Sira Technology, 2004

© Sira Technology, 2004

CH

RIS

on

PR

OB

A :

7th O

ct 2

004

An exemplar data set for testing An exemplar data set for testing atmospheric correction methodsatmospheric correction methods

CHRIS data from 14CHRIS data from 14thth September 2003. September 2003. CASI and ATM from 16CASI and ATM from 16thth September 2003. September 2003. Spectral ground data from 16Spectral ground data from 16 thth September 2003 September 2003

Spectra from ground calibration targets:Asphalt : spectral radiance, 300-1,000 nm, every 30 s during the flights.Spectral reflectance 300-1,000 nm immediately after the flights.Concrete and Grass : Spectral reflectance 300-1,000 nm(immediately after the flights).Atmospheric data:Water vapour (equivalent thickness)Aerosol optical thickness (AOT) : 5 spectral bands.Direct beam irradiance : 5 spectral bands.Global and diffuse quantum flux (PAR) : every 5 sGlobal broadband irradiance : every 5 sGlobal spectral irradiance : 300-2,400 nm, every 30 sRelative humidity, air temperature and barometric pressure, every 30 s.

CHRIS PROBA data from Thorney IslandCHRIS PROBA data from Thorney Island17th April 2004

CH_040417_3FE3_31.hdf CH_040417_3FE4_31.hdf

CH_040417_3FE5_31.hdf

CH_040417_3FE6_31.hdf

CHRIS PROBA data from Thorney IslandCHRIS PROBA data from Thorney Island

CH_041017_479F_31.hdf CH_041007_47A0_31.hdf

CH_041017_47A1_31.hdfCH_041017_47A2_31.hdf

CH_041017_47A3_31.hdf

Thorney Island calibration siteThorney Island calibration site

Data provided by NERC ARSF, 18/6/02