Copula-Based Downscaling of Coarse-Scale Soil Moisture Observations With Implicit Bias Correction

Abstract

Soil moisture retrievals, delivered as a CATDS (Centre Aval de Traitement des Données SMOS) Level-3 product of the Soil Moisture and Ocean Salinity (SMOS) mission, form an important information source, particularly for updating land surface models. However, the coarse resolution of the SMOS product requires additional treatment if it is to be used in applications at higher resolutions. Furthermore, the remotely sensed soil moisture often does not reflect the climatology of the soil moisture predictions, and the bias between model predictions and observations needs to be removed. In this paper, a statistical framework is presented that allows for the downscaling of the coarse-scale SMOS soil moisture product to a finer resolution. This framework describes the interscale relationship between SMOS observations and model-predicted soil moisture values, in this case, using the variable infiltration capacity (VIC) model, using a copula. Through conditioning, the copula to a SMOS observation, a probability distribution function is obtained that reflects the expected distribution function of VIC soil moisture for the given SMOS observation. This distribution function is then used in a cumulative distribution function matching procedure to obtain an unbiased fine-scale soil moisture map that can be assimilated into VIC. The methodology is applied to SMOS observations over the Upper Mississippi River basin. Although the focus in this paper is on data assimilation apcations, the framework developed could also be used for other purposes where downscaling of coarse-scale observations is required.

SMOS soil moisture assimilation for improved hydrologic simulation in the Murray Darling Basin, Australia

“…As noted by Wood et al (2011), it can be expected that LSMs will be applied at fine spatial resolutions, while the remote sensing products will be delivered at significantly coarser spatial resolutions. Either the data assimilation algorithm will have to take into account this spatial mismatch (Sahoo, De Lannoy, Reichle, & Houser, 2013), or the satellite products will have to be pre-processed so their spatial resolution matches the spatial resolution of the hydrologic model (Merlin, Al Bitar, Walker, & Kerr, 2010;Verhoest et al, 2015). Another challenge is the fact that model simulations and satellite retrievals often exhibit differences in SM climatology, i.e.…”

Global Downscaling of Remotely-Sensed Soil Moisture using Neural Networks

“…Several studies have used this approach to develop soil moisture at fine spatial scales (Chakrabarti et al, 2016(Chakrabarti et al, , 2017Mascaro et al, 2011;Piles et al, 2011Piles et al, , 2014Srivastava et al, 2013;Verhoest et al, 2015).…”

Improving soil moisture and runoff simulations over Europe using a high-resolution data-assimilation modeling framework

“…Several studies evaluated the assimilation of satellite SM observations into land surface models to produce optimal soil moisture estimates (e.g., Reichle and Koster, 2005;Han et al, 2014;Lievens et al, 2015;De Lannoy and Reichle, 2016). For example, Rains et al (2017) assimilated SMOS data into CLM over Australia for drought monitoring purposes.…”

“…To address this issue, the spatial mismatch between observations and modeling approaches needs to be taken into 5 account either in the data assimilation algorithm (Sahoo et al, 2013;De Lannoy et al, 2012), or through pre-processing of satellite products to match the model resolution (Merlin et al, 2010;Verhoest et al, 2015). Another challenge is the availability of computational resources, since the computational burden increases (non-) linearly with increasing model resolution, the number of ensemble members in the data assimilation system as well as the complexity of simulated processes.…”