Aerosols are thought to have
a large effect on the climate, both through the direct effect on radiation, and
through their interaction with clouds. In our atmosphere, aerosols refer to fine
solid particles or liquid droplets that are typically smaller than 1 ?m and can
be natural (dust, etc) or anthropogenic (haze, smoke, etc). Many aerosols can
act as cloud condensation nuclei (CCN). This is where we see cloud droplets able
to form on aerosol particles. Thus, any change in the aerosol number
concentration or aerosol properties may drastically influence cloud properties.
Several different indirect effects have been theorized in regions of increased
aerosols, including increases in cloud albedo (Twomey 1977), cloud fraction (Albrecht 1989), and decreases of
cloud top pressure (Koren et al. 2005). These effects
influence the radiative forcing of clouds and have been shown to be a
significant component of the total environmental anthropogenic radiative
However, the radiative forcing due to anthropogenic
aerosols is the most uncertain component of anthropogenic radiative forcing (Boucher 2013) with the interaction
between aerosols and clouds generating much of this uncertainty. It is thus
important and necessary to understand aerosol-cloud interactions to gain more
certain estimates of the radiative forcing they provide and thus a more certain
estimate of our future climate.
One person trying to tackle
this problem is Dr Edward Gryspeerdt. A recent seminar made by Gryspeerdt at
the University of Reading discussed his recent work on the role of aerosols in
cloud processes. In this report, an analysis of the work discussed by Gryspeerdt
will be made with a discussion on how this fits into the larger body of
research and suggesting where future work may be necessary.
Gryspeerdt and Stier (2012)
focussed on the relationship between aerosol concentration and cloud-albedo.
The “Twomey effect” theorises that increasing aerosol concentration at constant
cloud water content will decrease droplet size, which in turn increases cloud albedo
due to the increased scattering of smaller, more numerous droplets generating
an instantaneous radiative forcing (Twomey 1974). Previous global
satellite studies into the indirect aerosol effect have found positive relationship
between aerosol concentration and cloud albedo over ocean (Quaas et al. 2008). However, some find
a negative relationship over land, in contrast to models and theory (Quaas et al. 2009).
To study this further,
Gryspeerdt investigated the relationship in different cloud regimes. The findings
showed the strongest positive relationship in stratiform regimes over both land
and ocean. The Negative relationship previously observed over land was shown to
be due to the low cloud fraction (CF) regimes. This prompted Gryspeerdt to
first suggest that previous negative relationship findings are due to the
difficulty of retrieving cloud droplet number concentration (Nd) at
low cloud fractions.
Without accounting for the
different cloud regimes, the study found a negative sensitivity over land, in
the location of the strongest anthropogenic aerosol perturbations. Gryspeerdt and
Stier (2012) provided a way of compensating for the sampling bias by providing
a weighting to each cloud regime by its frequency. The weighting strengthens
the relationship over land, in some regions changing the sign to positive,
putting it in better agreement with the theory. The Increased sensitivity over
land is significant given that the short lifetime of many aerosols leads to
them to be concentrated near sources, often over land. The results first
highlighted the importance of regime based analysis when studying aerosol effects
due to the differing interaction strengths of the different regimes allowing
more accurate analysis of aerosol anthropogenic forcing.
Following these results, Gryspeerdt et al. (2014) aimed to further
constrain the influence of aerosols in different cloud regimes. Rather than the
“snapshot” study in Gryspeerdt and Stier (2012), this study looked at the
influence of aerosols on cloud regime development. To do this, Gryspeerdt et
al. (2014) made use of multiple temporally-spaced satellite retrievals to
observe the development of cloud regimes. The observation of cloud regime
development allowed the ability to account for the influences of CF and meteorological
factors on the aerosol-satellite retrieval, thus, reducing errors. This work
was important as hypothesised aerosol effects that modify the CF in some manner
cannot easily be separated from CF-related errors in the aerosol retrieval. By
accounting for the aerosol-CF relationship, the influence of meteorological
correlations compared to ”snapshot” studies were reduced, finding that simple
correlations over-estimate any aerosol effect on CF by a factor of two.
Furthermore, Gryspeerdt Investigated
the transitions between cloud regimes to allow a direct view of cloud
development in different aerosol environments. The results found an increased
occurrence of transitions into stratocumulus regimes over ocean with increased
aerosols, consistent with the hypothesis that aerosols increase stratocumulus
persistence. An increase in transitions into the deep convective regime over
land was also observed, consistent with the aerosol invigoration hypothesis (Koren et al. 2005). The increased
transitions between regimes with increasing aerosol, are consistent with
previously hypothesized effects of aerosols on cloud development (Rosenfeld et al. 2006), even after accounting
for the influence of CF and meteorological parameters at the time of aerosol
retrieval. These results Illustrated the importance of accounting for CF and
meteorological influences on aerosol retrieval.
The study demonstrates the
powerful possibilities of studying cloud development for controlling for
meteorological effects when investigating for aerosol cloud interactions.
Further work necessary to understand the relationship between development
methods and “snapshot” studies so that the advantages of each method can be
used in conjunction with each other. While the so the results likely upper
bound on the effect of aerosols on cloud development and CF. Whilst the results
of this study reduced the error due to meteorological and CF effects on the
aerosol retrieval, meteorological covariation with the cloud and aerosol
properties is harder to remove, so estimates were likely the upper bound.
Due to the difficulty in separating aerosol effects on
clouds from correlations generated by local meteorology, there was still
significant uncertainty. Gryspeerdt et al. (2016) attempts to
constrain the influence of aerosols on cloud fraction to further reduce this
uncertainty. The relationship between aerosol and CF is particularly important
to determine, due to the strong correlation of CF to other cloud properties and
its large impact on radiation. Strong correlations between aerosols and CF have
been made (Shinozuka et al. 2015) however the magnitude of the influence in previous studies were
also still highly uncertain.
Gryspeerdt et al. (2016)
used a new method of analysing the relationship (Pearl 1994) using the cloud
droplet number concentration (CDNC). Using this method, the impact of the
meteorological covariations was significantly reduced allowing for a more
certain aerosol-CF relationship. The method shows much of the aerosol-CF
correlation is explained by relationships other than that mediated by CDNC. The
strength of the global mean aerosol-CF relationship was reduced by 80% suggesting
that most of the aerosol-CF relationship is due to meteorological covariations,
especially in the shallow cumulus regime. The method provided an estimate of anthropogenic
radiative forcing from an aerosol influence of -0.48 Wm-2 although uncertainty
remained due to possible biases in the CDNC retrievals in broken cloud cases.
By using extra information
about cloud properties, the impact of meteorological covariations was shown to
have been reduced. In doing so, an improved observational estimate of the
possible aerosol influence on CF was determined that was in better agreement
with estimates from models and inverse studies. This study demonstrated how a
method for determining causality in complex systems can reduce the impact of
meteorological covariations on the aerosol-CF relationship. It was first shown
to have many advantages when investigating aerosol-cloud interactions, due to
the complex nature of the relationships involved and the difficulty in
explicitly accounting for every possible meteorological covariation. The
implications of this provide future studies with a tested method for isolating
aerosol influences on cloud properties and the resulting radiative forcing.
As aerosols can serve as CCN,
they can have a strong influence on cloud droplet number concentration (Nd).
In previous studies, the sensitivity of Nd to aerosol properties had
been used as a constraint on the strength of the radiative forcing from
aerosol-cloud interactions (Feingold et al. 2003). In said studies
(and ones discussed previously in this report), AOD has been used as a proxy
for CCN concentration. However, (Shinozuka et al. 2015) called this
assumption into question showing a disconnect between AOD and CCN. Considering
this, (Gryspeerdt et al. 2017)
continued his work with aerosol-cloud interactions attempting to further
constrain the relationship between aerosols and cloud albedo. To do this, new
techniques previously unavailable were taken advantage of to combat these
problems. An ensemble of global aerosol-climate models was used to demonstrate
how joint histograms between Nd and aerosol-properties can account
for issues highlighted by previous studies.
The accuracy of using
different aerosol proxies for diagnosing the aerosol anthropogenic forcing was
investigated, confirming that using the AOD significantly underestimates the
strength of the aerosol-cloud interactions in satellite data throwing into
question the results from previous studies
using this method. These results imply that information about the aerosol
size and distribution makes a dominant contribution to the accuracy of the
predictions made. Use of the Aerosol Index (AI) showed significant gains over using
AOD, similar to Stier (2016). Using AI has the
advantage over using CCN as it Is currently retrieved by satellite instruments.
This suggests that AI is potentially a useful parameter to use when calculating
observational constraints for future studies and model inputs.
A revised aerosol
anthropogenic forcing estimate is made of around -0.4 Wm-2 which is
lower than the estimate provided in Gryspeerdt et al. (2016) of -0.48 Wm-2, although there is a large diversity between model
estimates, ranging from -0.18 Wm-2 to -1.01 Wm-2 which is
a larger range than the Gryspeerdt et al. (2016) estimate range of -0.1 Wm-2 to -0.64 Wm-2. This
may seem like a step backwards however, the results from this study provide a
new technique to constrain the aerosol cloud relationship that has the
potential to be refined in future work.
Climate change is the one of
the biggest problems that faces humanity in the modern day. The first step to
combatting climate change is to understand the various interactions and
forcing’s in play. One aspect is made up by influence of anthropogenic aerosols
through direct influence on radiation and their interaction with clouds. The
radiative forcing from anthropogenic aerosols remains the one of the most
uncertain components of the total anthropogenic radiative forcing. Thus,
gaining a more certain view of aerosol-cloud interactions allows a more certain
estimate of the total anthropogenic forcing for our future climate.
The work done by Gryspeerdt
has made significant advances in furthering the field of knowledge of
aerosol-cloud interactions and, reducing the uncertainty on the anthropogenic
aerosol radiative forcing. The need to account for differing cloud regimes was
highlighted by showing increased agreement with models and the theory when
doing so. Previous studies results showing negative relationships between
aerosol concentration and cloud albedo were shown to be reversed to a positive
one when accounting for cloud regime. Furthermore, Gryspeerdt was able to
constrain aerosols influence on cloud regime development. Both results show the
powerful possibilities of using cloud regime in the analysis of aerosol-cloud
interactions to reduce the effect of meteorological and CF influences on cloud,
reducing overall uncertainty.
Gryspeerdt also showed the
possibilities in using further information about cloud properties, in this case
CDNC, to further reduce the uncertainty on anthropogenic aerosol radiative
forcing. In addition, AOD was shown to be a bad proxy for aerosol
concentration, showing the satellite retrieved AI was a much better candidate.
These results help reduce the large error currently assigned to anthropogenic
radiative forcing. The work provides future studies with the tools and the
knowledge to study aerosol-cloud interactions more accurately and providing
future climate prediction better estimates.
Into the future, further
refining of aerosol retrieval techniques will provide studies an ever-improving
ability to estimate aerosol-cloud interactions. The contradictions to previous
studies presented by Gryspeerdt that agree more with the theory show the need
to return to previous studies data.