1 Estimating SMOS error structure using triple collocation

1 Estimating SMOS error structure using triple collocation Delphine Leroux, CESBIO, France Yann Kerr, CESBIO, France Philippe Richaume, CESBIO, France

2 Soil moisture products at global scale AMSR-E (VUA) TMI (VUA) How to evaluate SMOS ? ? ? SSM/I (VUA) SMOS ? ERSASCAT (TU Wien) Aquarius SMAP AMSR-E (NSIDC) Model output (ECMWF)

3 Inter comparison between SMOS soil moisture and … o Ground measurements (point scale) o Other global products (point scale) o Global scale ? Statistics -> triple collocation

4 Structure 1. Triple Collocation method -> Theory and requirements 2. Chosen datasets -> Characteristics and differences 3. Global maps of relative errors -> Maps of bias and scale factors

1) Triple Collocation Theory Requirements Triple Collocation – theory Starting equation (Caires et al. , 2003) Final equation Taking the anomalies Ø Maps of the std of the errors Ø Maps of the bias Ø Maps of the scale factors r: bias s: scale factor ε: error 5

1) Triple Collocation Theory Requirements 6 Triple Collocation - requirements o Strong assumptions : § Mutually independent errors (ε) § No systematic bias between the datasets -> choose properly the 3 datasets -> TC applied to the anomalies and not to the variables directly o Requirements : § 100 common dates (Scipal et al. , IGARSS 2010) o Results : § Relative errors -> including the 6 closest grid nodes

2) Datasets Chosen datasets Number of triplets 7 Datasets Frequency (GHz) SMOS AMSR-E Incidence angle (°) Instrument resolution (km) Crossing time (A/D) Grid resolution (km) 1. 4 0 -55 40 6 am / 6 pm 15 6. 9 – 10. 7 … 55 57 -6. 25 1: 30 pm/ 1: 30 am 25 AMSR-E soil moisture derived with the VUA algorithm (Vrije University of Amsterdam) ECMWF product from SMOS Level 2 product (at SMOS resolution and crossing time)

2) Datasets Chosen datasets Number of triplets for 2010 Difficulties for regions with mountains, forests, wetlands, … 8

3) Global maps of … relative errors bias scaling factors Std of SMOS errors Good results in North America, North Africa, Middle East, Australia. Land contamination in Asia (Richaume et al. , RAQRS, 2010). 9

3) Global maps of … relative errors bias scaling factors Std of AMSR-E(VUA) errors Good results in the same areas as SMOS. 10

3) Global maps of … relative errors bias Std of ECMWF errors scaling factors 11

3) Global maps of … relative errors bias scaling factors Comparison over continents ! RELATIVE ERRORS SMOS is often between or close to the two values except in Asia 12

3) Global maps of … relative errors bias scaling factors Bias : AMSR-E(VUA) - SMOS Very high bias for high latitudes (mainly due to the vegetation) Mean bias around 0. 1 13

3) Global maps of … relative errors bias Bias : ECMWF - SMOS High bias for high latitudes but more homogeneous Mean bias around 0. 2 -0. 3 scaling factors 14

3) Global maps of … relative errors bias scaling factors Scale factor AMSR-E(VUA) Scale >1 Scale <1 higher dynamic than SMOS lower dynamic than SMOS 15

3) Global maps of … relative errors bias scaling factors Scale factor ECMWF Unlike the bias maps, there is no obvious structure for the scale factor 16

17 Conclusions o As part of the validation process, triple collocation compares 3 different datasets at a global scale : SMOS, AMSR-E/VUA and ECMWF o SMOS and AMSR-E/VUA have the same performance areas, but ECMWF and VUA give the best results o SMOS algorithm is still improving and it can be considered as a good start o Further work : apply triple collocation to other triplets (SMOS-NSIDC-ASCAT, etc…) and apply it with 2011 data

18 Thank you for your attention Any questions ?