Operational Dust Prediction - Mineral Dust: A Key Player in the Earth System - page 239

Fig. 10.2 Example of ICAP-MME 24 h dust forecasts of aerosol optical depth at 550 nm
wavelength valid for 2 January 2013. Models are ordered as follows: NAAPS ( top left ), GEOS-5
( middle left ), MASINGAR ( bottom left ), NMMB/BSC-CTM ( top right ), MACC/ECMWF ( middle
right ) and NCEP NGAC ( bottom right )

products describing spread (standard deviation and range of variation) are calculated
and are available daily at http://sds-was.aemet.es/ . An example of products is shown
in Fig. 10.3 . In order to generate them, the model outputs are bilinearly interpolated
toacommongridmeshof0.5 ı 0.5 ı . The daily SDS-WAS NA-ME-E multi-model

median (together with the individual models) is continuously evaluated against data
from the Aerosol Robotic Network (AERONET; see also Chap. 7 ) .

10.4 Data Assimilation for Dust Prediction

10.4.1 Introduction

Data assimilation offers a mathematical framework to incorporate observational
information into models. When dust prediction was started at various operational