Skip to Main content Skip to Navigation

Différence d'échelle spatiale entre les mesures satellitaires et in situ d'humidité du sol : analyse par des approches spatio-temporelles

Abstract : Soil moisture is an essential climate variable that is globally monitored with the help of satellite borne microwave instruments. The accuracy of satellite soil moisture estimations is assessed by direct comparison to in situ measurements. While satellite estimates have a resolution ranging between 30 and 100 km, in situ sensors typically measure over an area of a few centimetres (point resolution). This spatial scale mismatch between satellite and in situ soil moisture estimates impairs the validation process and the respective summary statistics to an extent that is not currently known. This thesis aims at improving the knowledge of the spatial scale mismatch, as well as providing methods for its assessment applicable to any validation area. To this end, the connection between the SM spatial and time scales was investigated. Modelled and measured soil moisture series at different regions of the globe were decomposed into time scales ranging from 0.5 to 128 days, using wavelet transforms. The spatial representativeness of the point measurements was then assessed, on a per time scale basis, with 4 different approaches: temporal stability analysis, triple collocation, the percentage of correlated areas (CArea) and a new approach that uses wavelet- based correlations (WCor). Moreover, one of the components of the mismatch, the sampling uncertainty, has been assessed separately with bootstrap and Monte Carlo simulations of point-support series. It was found that the average of the spatial representativeness values tends to increase with the time scales but so does their dispersion. This implies that some stations had large representativeness areas at seasonal scales, while others do not. At sub-weekly scales, all stations exhibited very small representativeness areas. Finally, the sampling uncertainty has been observed to have a considerable impact on satellite validation statistics.
Document type :
Complete list of metadatas
Contributor : Abes Star :  Contact
Submitted on : Friday, November 30, 2018 - 4:43:06 PM
Last modification on : Thursday, October 15, 2020 - 3:19:03 AM
Long-term archiving on: : Friday, March 1, 2019 - 3:20:56 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01941077, version 1



Beatriz Molero Molero Rodenas. Différence d'échelle spatiale entre les mesures satellitaires et in situ d'humidité du sol : analyse par des approches spatio-temporelles. Hydrologie. Université Paul Sabatier - Toulouse III, 2017. Français. ⟨NNT : 2017TOU30264⟩. ⟨tel-01941077⟩



Record views


Files downloads