https://epub.ub.uni-greifswald.de/frontdoor/index/index/docId/14588

*Authors: *Felix Wrana

*24 April 2026*

*Abstract*
In this work, a new algorithm to retrieve the size distribution of
stratospheric aerosol from the SAGE III instrument's satellite solar
occultation measurements was introduced. Utilizing the instrument's wide
spectral range, all three parameters of an assumed monomodal lognormal
distribution could be derived, eliminating a large source of ambiguity
compared to previous satellite retrievals. This results in one of the most
reliable satellite-based stratospheric aerosol size data sets to date.

Utilizing the balance this data set strikes between the size resolution of
the retrieved PSD and the spatial and temporal coverage, an unexpected
signal was found in both the SAGE III/M3M (2002-2005) and the SAGE III/ISS
(2017 - ongoing) data sets. Some volcanic eruptions were found to lead to a
decrease of the average size of stratospheric aerosol rather than an
increase, which is what would have been expected based on past studies,
e.g. of the 1991 Pinatubo eruption. Furthermore, this size decrease was
found to last up to 9 months, i.e. longer than expected. These were all
eruptions of tropical volcanoes with relatively small amounts of SO2 (less
than 1 Tg) injected into the stratosphere, namely Ruang and Reventador in
2002, Manam in 2005, Ambae in 2018, Ulawun in 2019 and La Soufrière in
2021. Model simulations with the MAECHAM5-HAM model could reproduce the
qualitative signal found, but not its longevity. On the other hand, the
2019 Raikoke eruption (48°N, 1.37 Tg SO2) and the 2022 Hunga eruption
(20.5°S, ~ 0.4-0.5 Tg SO2) were found to have led to an increase of the
average aerosol size.

This raises the question of which factors are needed for a volcanic
eruption to lead to a size decrease. The ambient temperature likely plays
an important role (and therefore season of the eruption, volcano latitude
and injection altitude) as low temperatures favor the nucleation of new
particles over the condensation onto existing particles. However, the
coagulation rate must also be an important factor, especially since the
size reduction was found to last so long in some cases. Furthermore, the
injected SO2 amount and the background PSD will also play a role. A lot of
further research is needed in order to better understand the interaction of
all these factors, both by modeling and observational experts.

The presented retrieval data sets were compared and validated against other
observational aerosol size data sets, such as from balloon-borne in situ
measurements and satellite limb scattering observations. While such
comparisons are in detail very difficult to interpret due to differences in
viewing geometry, observed air volumes, quality of collocations and
differences in assumptions and retrieval algorithms, ultimately good
agreements were found. This is encouraging especially in the case of the in
situ measurements, that represent the most reliable aerosol size data
available to date.

Finally, a general issue raised before by von Savigny and Hoffmann (2020),
which is an important source of uncertainty, was applied to the presented
method: The assumption of a simplified analytical function, in this case a
monomodal lognormal distribution, to describe the stratospheric aerosol
PSD. Since the larger particles of an aerosol population will individually
scatter more strongly than the smaller particles, a divergence of the true
PSD from the assumed PSD shape can introduce large errors into the aerosol
size retrievals from any remote sensing instrument. In this work, this
affects the number density the strongest.

The presented results are relevant for climate modeling and for the better
understanding of the physical processes in our atmosphere in general.
Furthermore, they add two major uncertainties to the predictability of
possible *geoengineering experiments in the form of solar radiation
management through the injection of sulfur species into the stratosphere:*
Firstly, we currently do not sufficiently understand under which
circumstances the aerosol size will increase or decrease, which will affect
their lifetime and radiative effect. Secondly, all remote sensing
observations that we have used in the past to understand the stratospheric
aerosol PSD include major uncertainties due to the limited size resolution
of the size retrievals and due to the smaller particles contributing very
little to the measured signals.

*Source: Universität Greifswald, Mathematisch-Naturwissenschaftliche
Fakultät*

-- 
You received this message because you are subscribed to the Google Groups 
"geoengineering" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/geoengineering/CAHJsh9-mTOG-qNGCHPYcEk2aOHL1hi-B_qN3A43yqnyJotHh2Q%40mail.gmail.com.

Reply via email to