
Holben, B.N., T.F.Eck, I. Slutsker, D. Tanré, J.P. Buis, A. Setzer, E. Vermote, J.A. Reagan, Y.J. Kaufman,
13
T. Nakajima, F. Lavenu, I Jankowiak, and A. Smirnov, 1998: AERONET - A federated instrument network
and archive for aerosol characterization. Rem. Sens. Environ., 66, 1 - 16.
Gilgen, H., C.H. W hitlock, F. Koch, G. Müller, A. Ohmura, D. Steiger, R. W heeler, 1995: Baseline Surface
14
Radiation Network Technical Plan for BSRN Data Management (Version 2.1). W CRP W MO/TD-no. 443.
W orld Climate Research Programme, 2001: Report of the Sixth Science and Review W orkshop,
15
Melbourne, Australia, 1 - 5 May 2000 (Annex IV). W CRP Informal Report No. 17/2001.
53
7.0 Measurement of Aerosol Optical Depth
7.1 Introduction
The monitoring of aerosol optical depth (AOD) has been considered an important, but difficult, observation
that is necessary if there is to be an increase in understanding of the surface radiation budget. While
the optical depth provides information on spectral atmospheric extinction, a number of inversion algorithms
have been developed to use this information to produce data on the columnar aerosol number-size
distribution, volume distribution and concentration. Along with other measurements of the solar aureole
and almucantar the data has also been inverted to provide information on aerosol absorptivity. By
measuring, or more normally assuming, a change in aerosol concentration with height above the surface,
these optical properties can be used to better understand the radiative regime of the atmosphere. Over
time, more and more user communities have expressed the need for this type of data. Several global
and regional aerosol measurement networks have been established in response to this need. The
two most significant of these, with respect to the ongoing work of the BSRN are the Global Atmosphere
Watch (GAW) sunphotometer network, which is being established at global background stations, and
the NASA Aerosol Robotic Network (AERONET) .
13
At the inception of the BSRN the importance of aerosol optical properties on the radiation budget was
recognized and the BSRN Archive was designed to include values of spectral AOD . The sunphotometers
14
available in the early 1990s were found to be too unstable and the measurement protocols’ unacceptable
to determine changes in global and regional climates, as was the mandate of the BSRN. Instrument
instability was due in large part to the rapid aging of the interference filters used in the instruments,
especially at shorter wavelengths, while many measurement protocols consisted of only a few
measurements per day because they were designed around the use of handheld instruments. Changes
in instrument design, including the reduction of the exposure time of filters to UV radiation and the
stabilization of instrum ent temperature have improved the perform ance and stability of new instruments.
Accurate tracking system s have improved the repeatability of observations by decreasing pointing
uncertainty, while the increase in sampling frequency provides a means of tracking changes in instrument
and atmospheric conditions that could not be accomplished with only several or even tens of observations
per day.
At the 6 BSRN Science and Review Workshop the Working Group on Aerosol Optical Depth proposed,
th 15
and the BSRN accepted, specifications related to AOD measurements. The objective of the BSRN
measurement program is to provide AOD measurements at selected wavelengths between 360 and
1100 nm with an uncertainty no greater than 0.01 under near ideal conditions, decreasing to no greater
than 0.02 for all conditions when water clouds do not interfere with the direct line-of-site to the solar
disk.
The calculation of AOD, although straightforward in principle, varies because of the number of
approximations that are used in its calculation. Small differences in the methods used to calculate
sun-earth distance, Rayleigh optical depth, or the ozone absorption coefficients and scale height all
increase the overall uncertainty in the calculation of AOD. Individual differences between methods
of calculation alter the overall AOD. In combination, one set of algorithms, compared to a different
set will act as a bias. This will be of greater significance to the determination of small AODs, as measured
at background stations, than those in more turbid areas. The Working Group on Aerosol Optical Depth
Measurements determined that because of the possibility of such discrepancies, the submission of
AODs to the archive could not provide users with an accurate long-term, spatially diverse database.
Therefore, it was decided that spectral transmission data, with ancillary information, be transmitted
to the BSRN Archive. The archive would then calculate the aerosol optical depth using a common
set of algorithms. Once determined the algorithms will be made available to BSRN scientists, but all
Komentáře k této Příručce