Land Surface Temperature Land Surface Temperature Development and Validation Development and Validation

for GOES-R Mission for GOES-R Mission

Yunyue Yu (STAR)Yunyue Yu (STAR)

Peng Yu, Yuling Liu, Kostya Vinnikov (UMD/CICS)Peng Yu, Yuling Liu, Kostya Vinnikov (UMD/CICS)

Rob Hale (CSU/CIRA)Rob Hale (CSU/CIRA)

Dan Tarpley (Short & Associates)Dan Tarpley (Short & Associates)

GOES-R AWGGOES-R AWG

March 20131Yunyue.Yu@noaa.govYunyue.Yu@noaa.gov

Topics: Future Satellite Mission

2. Introduction2. Introduction

Land Surface Temperature (LST) is one of baseline products for the Land Surface Temperature (LST) is one of baseline products for the GOES-R MissionGOES-R Mission

LST algorithm has been developed and tested at STAR, and is LST algorithm has been developed and tested at STAR, and is implementing at Vender (Harris Corporation)implementing at Vender (Harris Corporation)

Issues in the LST dataIssues in the LST data» Satellite data issues: Satellite data issues: observation geometryobservation geometry, , Instrument noise/stabilityInstrument noise/stability

» Ground data issues:Ground data issues: emissivity variation, Instrument noise/stability emissivity variation, Instrument noise/stability

» Others: Others: temporal and spatial variability, cloud impactstemporal and spatial variability, cloud impacts

Validation needsValidation needs» discrimination among the above problems as much as possiblediscrimination among the above problems as much as possible

» use real-time cal/val info from other products to identify problem use real-time cal/val info from other products to identify problem cascades (instrument noise > cloud detection > LST)cascades (instrument noise > cloud detection > LST)

» need parallel cal/val system for ground observationsneed parallel cal/val system for ground observations

2

3. 3. Methodology/Expected Methodology/Expected

OutcomesOutcomes

Validation approach of the LST data has been Validation approach of the LST data has been developed at STAR: developed at STAR: » MODIS, SEVIRI data used as proxyMODIS, SEVIRI data used as proxy

» Utilize existing ground station dataUtilize existing ground station data– Stations under GOES-R Imager coverageStations under GOES-R Imager coverage– Stations under MSG/SEVIRI coverageStations under MSG/SEVIRI coverage

» Ground site characterizationGround site characterization» Stringent cloud filteringStringent cloud filtering» Multiple comparisons: satellite vs satellite, satellite vs ground Multiple comparisons: satellite vs satellite, satellite vs ground

stationstation..» Direct and indirect comparisonsDirect and indirect comparisons» International cooperationInternational cooperation

Validation Outcomes: Validation Outcomes: Routing validation, Deep-dive Routing validation, Deep-dive validation validation 3

March 20134

Deep DivingDeep DivingEndEnd

ProblemsProblems

STARTSTART

User input: Sensor, stations, period

User input: Sensor, stations, period

Ground data reader and cloud

filtering

Ground data reader and cloud

filtering

Geo-location matchup

Geo-location matchup

Time matchupTime matchup

LST calculation and analysis

LST calculation and analysis

Read in TPW, Emissivity, etc.Read in TPW,

Emissivity, etc.Satellite data

reader and cloud filtering

Satellite data reader and cloud

filtering

MODIS SURFRAD

Output graphics and statistics

SEVIRI

Others

CRN

Others

Additional Cloud filtering

Additional Cloud filtering

Ancillary

Preprocessed data package

Visualization

Flow chart of the LST validation system

NO

Yes

4. Results 4. Results

5

Routine Validation tool applied for comparing GOES-R LST (top-right, derived from MODIS as proxy), and the MODIS LST (top-left). Map of Difference (bottom-left) and histogram of the difference (bottom-right) are also displayed.

Validation Tool Widget

Routine Validation Routine Validation -- SURFRAD data results-- SURFRAD data results

6

Comparison results of GOES-8 LST (as proxy) using six SURFRAD ground station data, in 2001.

Month

Site 1 Site 2 Site 3 Site 4 Site 5 Site 6

Day Night Day Night Day Night Day Night Day Night Day Night

1 48 17 55 64 107 107 58 48 100 67 100 51

2 43 30 38 38 73 57 55 34 41 54 64 48

3 0 0 51 71 94 62 93 79 41 63 123 34

4 94 18 39 80 81 34 62 61 81 57 139 35

5 71 22 27 59 127 65 75 83 82 45 168 75

6 50 26 82 102 82 51 65 50 86 64 187 60

7 6 4 91 77 40 8 27 38 40 43 173 74

8 43 30 129 113 41 38 82 137 56 45 107 52

9 115 57 95 108 124 49 79 110 116 85 189 98

10 103 38 62 95 184 39 90 79 74 58 115 61

11 114 34 43 129 146 70 64 53 116 53 131 88

12 40 38 67 66 124 71 73 70 107 74 113 56

Total 727 314 779 1002 1223 651 823 842 940 708 1609 732

Numbers (Table, left) and scatter plots (right) of the match-up LSTs derived from GOES-8 Imager data vs. LSTs estimated from SURFRAD stations in year 2001. Data sets in plots are stratified for daytime (red) and night time (blue) atmospheric conditions

7

T(x,y,t)

T(x0,y0,t0)

ASTER pixel The site pixelMODIS pixel

Quantitatively characterize the sub-pixel heterogeneity and evaluate whether a ground site is adequately representative for the satellite pixel. The sub-pixels may be generated from pixels of a higher-resolution satellite.

For pixel that is relatively homogeneous, analyze statistical relationship of the ground-site sub-pixel with the surrounding sub-pixels:

{T(x,y) } ~ T(x0,y0) Establish relationship between the objective pixel and its sub-pixels (i.e.,

up-scaling model), e.g., Tpixel = T(x,y) + T (time dependent?)

Site characterization analysis using ASTER data— an integrated approach for understanding site representativeness and for site-to-pixel model development

Surface heterogeneity is shown in a 4km x 4km Google map (1km x 1km, in the center box) around the Bondville station area

The Synthetic pixel/sub-pixel model

SiteMODIS Pixel-

SURFRADSynthetic Pixel-

SURFRADNearest Aster Pixel

– Synthetic pixelMean StdDev Mean StdDev Mean StdDev

Desert Rock, NV -0.44 1.84 2.09 1.69 0.04 0.44 Boulder,CO -0.49 2.08 1.49 2.90 -0.09 0.67 Fort Peck, MT 0.35 2.52 0.78 1.95 0.38 1.15 Bondville, IL -0.17 1.51 1.04 1.48 0.03 0.90 Penn State, PA -1.53 1.91 0.61 1.96 0.04 1.07

Site-to-Pixel Statistical Relationship for 5 SURFRAD sites

””Deep-Dive” ValidationDeep-Dive” Validation

EOGC 2009, May 25-29, 20098

Goodwin Creek, MS, observation pairs are about 510. View Zenith of GOES-8/-10: 42.68 0/61.890

”Deep-Dive” Validation Tools-- Directional effect study

Due to the satellite LST directional properties (surface components, topography, shadowing etc.), the satellite LST can be significantly different from different view angles. Deep dive validation tools may be used for case studies and improved algorithms.

5.5. Possible Path to Possible Path to OperationsOperations

The validation tools should be considered as non-The validation tools should be considered as non-operational, or semi-operational. Rather, it is for LST operational, or semi-operational. Rather, it is for LST developers and users.developers and users.

A prototype development for the validation toolA prototype development for the validation tool» Scientific approaches Scientific approaches » Test data sets: satellite proxy and ground data Test data sets: satellite proxy and ground data

Case studies for testing the improvementCase studies for testing the improvement Technical Detail Documentation developmentTechnical Detail Documentation development Software design and architecture, coding standardSoftware design and architecture, coding standard

9

6. Future Plans6. Future Plans

Improvement of site characterization modelImprovement of site characterization model Provide LST improvement approachProvide LST improvement approach Case study of emissivity variation impact in LST and Case study of emissivity variation impact in LST and

correction methodcorrection method Validation visualization tool improvementValidation visualization tool improvement Additional Cloud filtering methodAdditional Cloud filtering method Field data collection and processingField data collection and processing

10

7. Publication List7. Publication List

Project PublicationsProject PublicationsK. Vinnikov, K. Vinnikov, Y. Yu,Y. Yu, M. Goldberg, D. Tarpley, P. Romanov, I. Laszlo, M. Chen, “ Angular Anisotropy of Land Surface M. Goldberg, D. Tarpley, P. Romanov, I. Laszlo, M. Chen, “ Angular Anisotropy of Land Surface

Temperature”, Temperature”, Geophysical Research Lett. Geophysical Research Lett. VOL. 39, L23802, doi:10.1029/2012GL054059, 2012VOL. 39, L23802, doi:10.1029/2012GL054059, 2012

H. Xu, H. Xu, Y. Yu, D. Tarpley, Y. Yu, D. Tarpley, F-MF-M. . Göttsche and F-S. Olesen “Evaluation of GOES-R Land Surface Temperature Algorithm Using Göttsche and F-S. Olesen “Evaluation of GOES-R Land Surface Temperature Algorithm Using SEVIRI Satellite Retrievals with SEVIRI Satellite Retrievals with in-situin-situ Measurements”. Measurements”. IEEE Geoscience and Remote Sensing, IEEE Geoscience and Remote Sensing, in revision, 2012 (Sept).in revision, 2012 (Sept).

Y. Liu, Y. Liu, Y. Yu,Y. Yu, D. Sun, D. Tarpley, L. Fang, “Effect of Different MODIS Emissivity Products on Land-Surface Temperature D. Sun, D. Tarpley, L. Fang, “Effect of Different MODIS Emissivity Products on Land-Surface Temperature Retrieval From GOES Series”, Retrieval From GOES Series”, IEEE Geoscience and Remote Sensing, IEEE Geoscience and Remote Sensing, in press, 2012in press, 2012

D. Sun, D. Sun, Y. Yu,Y. Yu, H. Yang, Q. Liu, J. Shi, “Comparison between GOES East and West for Land Surface Temperature Retrieval H. Yang, Q. Liu, J. Shi, “Comparison between GOES East and West for Land Surface Temperature Retrieval from a Dual-Window Algorithm”, from a Dual-Window Algorithm”, IEEE Geoscience and Remote Sensing Lett, IEEE Geoscience and Remote Sensing Lett, in press, 2012in press, 2012

Yu, YYu, Y; Tarpley, D.;   Privette, J. L.;   Flynn, L. E.;   Xu, H.;   Chen, M.;   Vinnikov, K. Y.;   Sun, D.;   Tian, Y., ; Tarpley, D.;   Privette, J. L.;   Flynn, L. E.;   Xu, H.;   Chen, M.;   Vinnikov, K. Y.;   Sun, D.;   Tian, Y., Validation of GOES-R Validation of GOES-R Satellite Land Surface Temperature Algorithm Using SURFRADGround Measurements and Statistical Estimates of Error Satellite Land Surface Temperature Algorithm Using SURFRADGround Measurements and Statistical Estimates of Error Properties,Properties, IEEE Trans. Geosci. Remote Sens.IEEE Trans. Geosci. Remote Sens., , vol. 50vol. 50, No. 3, pp. 704-713, 2012, No. 3, pp. 704-713, 2012DOI: 10.1109/TGRS.2011.2162338DOI: 10.1109/TGRS.2011.2162338

Hale, Robert; Gallo, Kevin; Tarpley, Dan; Hale, Robert; Gallo, Kevin; Tarpley, Dan; Yu, YunyueYu, Yunyue, , Characterization of variability at in situ locations for calibration/validation of Characterization of variability at in situ locations for calibration/validation of satellite-derived land surface temperature  datasatellite-derived land surface temperature  data, REMOTE SENSING LETTERS Vol. 2, Issue: 1, 2011 DOI: , REMOTE SENSING LETTERS Vol. 2, Issue: 1, 2011 DOI: 10.1080/01431161.2010.490569 10.1080/01431161.2010.490569

Gallo, Kevin, Robert Hale, Dan Tarpley, Gallo, Kevin, Robert Hale, Dan Tarpley, Yunyue Yu,Yunyue Yu, Evaluation of the Relationship between Air and Land Surface Temperature Evaluation of the Relationship between Air and Land Surface Temperature under Clear- and Cloudy-Sky Conditionsunder Clear- and Cloudy-Sky Conditions, Journal of Applied Meteorology and Climatology, Volume: 50 Issue: 3 Pages: 767-, Journal of Applied Meteorology and Climatology, Volume: 50 Issue: 3 Pages: 767-775, 2011 . DOI: 10.1175/2010jamc2460.1 775, 2011 . DOI: 10.1175/2010jamc2460.1

Vinnikov, K. Y.; Vinnikov, K. Y.; Yu, YYu, Y.; Goldberg, M. D.; et al, .; Goldberg, M. D.; et al, Scales of temporal and spatial variability of midlatitude land surface temperature,Scales of temporal and spatial variability of midlatitude land surface temperature, Journal of Geophysical Research-Atmospheres, Volume: 116 , 2011 DOI: D02105 10.1029/2010jd014868Journal of Geophysical Research-Atmospheres, Volume: 116 , 2011 DOI: D02105 10.1029/2010jd014868

Yu, Y.Yu, Y., D. Tarpley, J. L. Privette, M. K. Rama Varna Raja, K. Vinnikov, H. Xue, , D. Tarpley, J. L. Privette, M. K. Rama Varna Raja, K. Vinnikov, H. Xue, Developing algorithm for operational GOES-R land Developing algorithm for operational GOES-R land surface temperature productsurface temperature product, , IEEE Trans. Geosci. Remote Sens.IEEE Trans. Geosci. Remote Sens., , vol. 47vol. 47, no. 3, pp. 936-951., 2009, no. 3, pp. 936-951., 2009DOI: 10.1109/tgrs.2008.2006180 DOI: 10.1109/tgrs.2008.2006180

11