Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Publications at this Location » Publication #313464
Title: Optimal averaging of soil moisture predictions from ensemble land surface model simulations

Author
item Crow, Wade
item SU, CHUN-HSU - University Of Melbourne
item RYU, D. - University Of Melbourne
item YILMAZ, M.T. - Collaborator

Submitted to: Water Resources Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/15/2015
Publication Date: 11/28/2015
Citation: Crow, W.T., Su, C., Ryu, D., Yilmaz, M. 2015. Optimal averaging of soil moisture predictions from ensemble land surface model simulations. Water Resources Research. 51:9273–9289. doi: 10.1002/2015WR016944.

Interpretive Summary: Increasingly, soil water balance models are being operationally applied to provide soil moisture estimates within agricultural landscapes. These estimates are of value for a number of key agricultural applications including: drought forecasting, irrigation scheduling, and optimizing fertilizer usage. However, various water balance models can vary significantly with regards to the details of their soil water balance calculations and - as a direct result - soil moisture estimates can vary widely between various competing modeling approaches. As a result, new statistical tools are needed to combine multiple model simulations into a single optimized prediction of soil moisture availability. This paper describes a mathematical approach which utilizes cross-comparisons between these model-based soil moisture products and an (independently-acquired) remotely-sensed soil moisture product in order to determine the required statistical information needed to merge the model-based soil moisture products into an optimized prediction of soil moisture variability. This technique can potentially be used by operational drought monitors to enhance the quality of soil water availability estimates used in agricultural decision support systems.

Technical Abstract: The correct interpretation of ensemble information obtained from the parallel implementation of multiple land surface models (LSMs) requires information concerning the LSM ensemble’s mutual error covariance. Here we propose a new technique for obtaining such information using an instrumental variable (IV) regression approach and comparisons against a long-term surface soil moisture dataset obtained from satellite remote sensing. Application of the approach to multi-model ensemble soil moisture output from the North American Land Data Assimilation System (NLDAS-2) and European Space Agency (ESA) Soil Moisture (SM) Essential Climate Variable (ECV) dataset allows for the calculation of optimal weighting coefficients for individual members of the NLDAS-2 ensemble and a biased-minimized estimate of uncertainty in a deterministic soil moisture analysis derived via such optimal weighted averaging. As such, it provides key information required to accurately condition soil moisture expectations using information gleaned from a multi-model LSM ensemble. However, existing continuity and rescaling concerns surrounding the generation of long-term, satellite-based soil moisture products must likely be resolved before the proposed approach can be applied with full confidence.