Triple Collocation Analysis for Two Error-Correlated Datasets: Application to L-Band Brightness Temperatures over Land

Verónica González-Gambau,Estrella Olmedo,Justino Martínez,Manuel Martín-Neira,Cristina González-Haro,Antonio Turiel,Roger Oliva

doi:10.3390/rs12203381

Verónica González-Gambau, Estrella Olmedo + Show 5 more

Open Access

https://doi.org/10.3390/rs12203381

Copy DOI

Abstract

The error characterization of satellite observations is crucial for blending observations from multiple platforms into a unique dataset and for assimilating them into numerical weather prediction models. In the last years, the triple collocation (TC) technique has been widely used to assess the quality of many geophysical variables acquired with different instruments and at different scales. This paper presents a new formulation of the triple collocation (Correlated Triple Collocation (CTC)) for the case of three datasets that resolve similar spatial scales, with two of them being error-correlated datasets. Besides, the formulation is designed to ensure fast convergence of the error estimators. This approach is of special interest in cases such that finding more than three datasets with uncorrelated errors is not possible and the amount of data is limited. First, a synthetic experiment has been carried out to assess the performance of CTC formulation. As an example of application, the error characterization of three collocated L-band brightness temperature (TB) measurements over land has been performed. Two of the datasets come from ESA (European Space Agency) SMOS (Soil Moisture and Ocean Salinity) mission: one is the reconstructed TB from the operational L1B v620 product, and the other is the reconstructed TB from the operational L1B v620 product resulting from application of an RFI (Radio Frequency Interference) mitigation technique, the nodal sampling (NS). The third is an independent dataset, the TB acquired by a NASA (National Aeronautics and Space Administration) SMAP (Soil Moisture Active Passive) radiometer. Our analysis shows that the application of NS leads to TB error reduction with respect to the current version of SMOS TB in 80% of the points in the global map, with an average reduction of approximately 1 K over RFI-free regions and approximately 1.45 K over strongly RFI-contaminated areas.

Highlights

Triple collocation (TC) analysis has been widely used for the quality assessment of many remotely sensed geophysical variables, such as ocean winds [1,2], soil moisture [3,4,5,6], and sea surface salinity [7,8]
Bias is the difference between the average of all valid estimates of the error standard deviations σδi and the value used for the generation of the dataset
A new formulation of the triple collocation analysis has been developed for the specific case of three datasets resolving similar spatial scales and presenting correlated errors for two of them

Summary

Introduction

Triple collocation (TC) analysis has been widely used for the quality assessment of many remotely sensed geophysical variables, such as ocean winds [1,2], soil moisture [3,4,5,6], and sea surface salinity [7,8]. Triple collocation is a powerful tool first introduced by [1] to estimate the standard deviations of the random errors of three spatiotemporally collocated measurements of the same target parameter. These estimations referred to the dynamic range of the system chosen as a reference. Major assumptions of TC are that errors must be uncorrelated with the target variable and that the errors of the different datasets must be uncorrelated between them. The latter is the major drawback of TC in its original conception. Pierddica et al proposed a quadruple collocation assuming uncorrelated errors between all four datasets, which leads to a more robust estimation of unknown error standard deviations and addresses the problem of having a common reference [4]

Objectives

Results

Conclusion