New Climatology

The new climatology is defined by monthly maps for (mid-visible) the aerosol characteristics of AOD, ro0, and AnP. The climatology prioritizes quality data of local statistics, which are merged onto globally and temporally complete background fi elds, from modeling. To be more specifi c, monthly maps of

AeroCom model median values (Kinne et al. 2006) define complete and consistent background fields, and monthly statistics from sun/sky photometers are the quality data. Thus, this climatology does not directly involve data from satellite remote sensing. The quality data involve sky/sun photometer monthly statistics of more than 150 AERONET and 7 SKYNET sites, with a complete annual cycle and more than 200 AERONET sites with limited annual coverage. The merging of the sun-/sky photometer data onto the model median background fields occurs in five consecutive steps (Kinne 2008):

Rate all sun photometer sites individually in terms of quality and regional representation.

Combine sun photometer local monthly statistics onto the regular grid (1°X1°) of modeling.

Define a global ratio map (backward field) based on globally spreading available ratios between sun photometer data and model data.

Establish local domain fields around each sun photometer site.

Establish "effective weights" (apply the domain weight to the ratio map).

To define the new aerosol climatology, the effective weights fields are applied to the background fields from modeling to yield modified monthly maps for AOD, ro0, and AnP.

Figures 3.3-3.5 compare maps of annual averages between the modeling background, AERONET samples, and the merged product. The data of the sky or sun photometer samples have been enlarged in the fi gures for legibility, and as a site's domain varies according to its assigned regional representation, changing colors are possible in the case of nearby sites. With respect to the merged product of ro0 in Figure 3.4, some of the lower values from sky photometry are not shown, because their associated AOD fell below the confidence threshold (mid-visible AOD < 0.3) for reliable ro0 inversions.

Aerosol fields

0 0

Post a comment