Physically based models for remote sensing of the dynamics of atmospheric aerosols

Dust storms in the earth’s major desert regions significantly influence microphysical weather processes, the CO2-cycle and the global climate in general. Increases in the spatio-temporal resolution of satellite instruments have created new opportunities to understand these phenomena. However, both big data scales and inherent stochasticity of the process pose significant challenges. In this interdisciplinary project, we develop a statistical model of atmospheric transport that relies on latent Gaussian Markov random fields for inference. We propose a “Bayesian hierarchical model” (BHM) that maps “Spinning Enhanced Visible and Infrared Imager” (SEVIRI) measurements to a predictor of the dust density. Case studies demonstrate that, as compared to linear methods, our “latent signal mapping” (LSM) approach mitigates effects of signal intrinsic noise on further processing steps. Furthermore, an extensive cross-validation study is employed to show that LSM successfully adapts to intra-daily changes of the infrared data and yields outstanding dust detection accuracy. Physically, the dust density and its transport process are tied together by the continuity equation. We link optical flow methods and the formulation of the transport process as a latent field in a generalized linear model using the “integrated nested Laplace approximation” (INLA) for inference. This framework is specified such that it satisfies physical constraint equations. The importance of allowing compressible motion and treating the problem in a statistical manner is emphasized by simulation and case studies showing a significant reduction in errors of the estimated flow field. In addition, we demonstrate how our methodology provides uncertainty quantification, dust storm forecasts and estimation of emission sources. Our methodology is able to both accurately and coherently detect sources of dust and it’s temporal evolution. For the first time we are able to solve these critical problems in this field.

 

Name and contact of project responsible(s):

PD Dr. C. Garbe (Interdisciplinary Center for Scientific Computing, Heidelberg University)

Additionally involved scientists and partners

Fabian Bachl (Interdisciplinary Center for Scientific Computing, Heidelberg University)
Matthias Klinger (Interdisciplinary Center for Scientific Computing, Heidelberg University)
Prof. Dr. Tilmann Gneiting (Institute of Applied Mathematics, Heidelberg University)
Prof. Dr. R. Rannacher (Institute of Applied Mathematics, Heidelberg University)

Publications:

F.E. Bachl, A. Lenkoski, T.L. Thorarinsdottir, C.S. Garbe,
Bayesian Motion Estimation for Dust Aerosols,
Annals of Applied Statistics, (2014), [in revision]

F.E. Bachl, C.S. Garbe, P. Fieguth,
Bayesian Inference on Integrated Continuity Fluid Flows and their Application to Dust Aerosols,
in: IEEE International Geoscience and Remote Sensing Symposium, IEEE, (2013), pp. 2246-2249. DOI = 10.1109/IGARSS.2013.6723264

F.E. Bachl, C.S. Garbe, P. Fieguth, A Bayesian Approach to Spaceborn Hyperspectral Optical Flow Estimation on Dust Aerosols,
in: IEEE International Geoscience and Remote Sensing Symposium, IEEE, (2012), pp. 256-259.
DOI = 10.1109/IGARSS.2012.6351589