PossiblE imPACTs of PolluTioN AErosols oN …PossiblE imPACTs of PolluTioN AErosols oN AusTrAliAN ClimATE AND wEAThEr well as the direct effects of aerosols on solar radiation, indirect

26 CleanAirandEnvironmentalQualityVolume41No.3.August2007

PossiblEimPACTsofPolluTioNAErosolsoNAusTrAliANClimATEANDwEAThEr

Possible impacts of pollution aerosols on Australian climate and weather: A short reviewL. D. Rotstayn

AbsTrACT

This review considers possible impacts of pollution aerosols on Australian climate and weather. The main focus is on recent analyses of 20th Century simulations with a low-resolution version of the CSIRO global climate model. Two ensembles of simulations were performed, with and without anthropogenic-aerosol forcing, in order to determine the impact of aerosols. These simulations suggest large effects on tropical Australian rainfall due to anthropogenic aerosol forcing from the Northern Hemisphere. In particular, the observed rainfall increase in north-western Australia in recent decades may be attributable to the large Asian aerosol haze. The haze cools Asia and the surrounding oceans, thus altering the temperature gradient between Asian and Australia, and this affects the monsoonal winds and rainfall. Other impacts of anthropogenic aerosol forcing in the model include an improved simulation of temperature in the Indian Ocean, and a tendency for aerosols to make the sea-surface temperature pattern in the Pacific become more La Niña-like with time. These results suggest that more research into aerosols and their forcing of climate may be crucial to improving our understanding of climate change in Australia, and the broader Indo-Pacific region.

Other studies, which have considered the possibility of climatic or meteorological effects due to Australian-sourced aerosol, are also briefly reviewed. Areas of interest are (a) recent data suggesting a substantial radiative forcing due to aerosol from biomass burning in northern Australia during the dry season, (b) an assertion, which remains contentious, that local aerosol pollution can significantly suppress rain formation in Australia, and (c) evidence from the 2003 Canberra bushfire that the dense smoke layer stabilised the lower atmosphere in the days after the fire, and prevented strong north-westerly winds from penetrating to the surface. The latter study suggests that inclusion of smoke aerosol in weather-prediction models could substantially improve critical weather forecasts during bushfires.

iNTroDuCTioN

Aerosol effects are considered to be the most uncertain of the radiative forcing components by which human activity is altering our climate (Solomon et al. 2007). This is especially true of the indirect effects,

whereby aerosols modify cloud properties by acting as cloud condensation nuclei (CCN) (e.g., Twomey 1974). Despite this, it has been understood for over a decade that aerosol pollution produced by human activity can partly offset the warming effect of greenhouse gases, and that inclusion of these anthropogenic aerosols in climate models can improve the agreement between models and observations (e.g., Mitchell et al. 1995). In recent years, a number of studies have suggested the possibility of large climatic effects due to aerosols in the Northern Hemisphere, primarily due to the strong spatial variations in the distribution of aerosols. This spatial inhomogeneity can lead to substantial changes in atmospheric circulation and rainfall, especially in the Tropics, where circulation is known to be sensitive to large-scale horizontal temperature gradients (Rotstayn and Lohmann 2002a). The best-known example of this phenomenon is the El Niño Southern Oscillation (ENSO), whereby changes in sea-surface temperatures (SSTs) in the eastern tropical Pacific affect rainfall over Australia. Possible impacts of anthropogenic aerosols include summertime floods and droughts in China (Menon et al. 2002), droughts in the Sahel (Rotstayn and Lohmann 2002a), and a weakening of the South Asian Monsoon (Ramanathan et al. 2005). Aerosols are also thought to have contributed to the “global dimming” phenomenon, which refers to the decrease in measured global (direct plus diffuse) solar radiation at the surface from about 1960 to 1990 (Liepert et al. 2004; Nazarenko and Menon 2005). Recent reductions in aerosol pollution in many parts of the Northern Hemisphere may have contributed to the widespread reversal of the dimming trend since 1990 (Wild et al. 2005). If aerosols contributed to global dimming then aerosols are also likely to have contributed to observed decreases in pan evaporation in heavily polluted areas (Roderick and Farquhar 2002). Collectively, these studies suggest large impacts of anthropogenic aerosols on the climate of the Northern Hemisphere.

Aerosol levels are much lower in the Southern Hemisphere than in the Northern Hemisphere. Until recently, there has been little evidence of substantial aerosol effects on the climate of the Southern Hemisphere (or of Australia), other than the vague idea that aerosols act as a kind of “negative greenhouse gas”. However, recent climate modelling at CSIRO (Rotstayn et al. 2007,

hereafter R2007) has suggested that aerosols (principally from the Northern Hemisphere) may be important for understanding late 20th Century trends in the large-scale dynamics of the Australasian region. A key finding of this study was that the observed increase of rainfall over north-western and central Australia in recent decades could be substantially driven by the Asian aerosol haze.

According to the model, the haze cools the Asian continent and surrounding oceans, thus altering the meridional temperature gradient between Asia and Australia. The change in temperature gradient affects the monsoonal winds, which carry moisture towards northern Australia during the wet season. Analysis of these simulations has been extended by Cai et al. (2006) and Cai et al. (2007), who have shown the fundamental importance of the oceans in the climate system’s response to aerosol forcing. The main purpose of this article is to review the above work and its implications for our understanding of past and future climate change in our region. Other studies, which have considered the possibility of climatic or meteorological effects due to Australian-sourced aerosol, will also be briefly reviewed.

simulatedeffectsofNorthernhemisphereaerosolonAustralianclimate

R2007 used a low-resolution version of the Mk3 CSIRO Global Climate Model (GCM) to perform ensembles of coupled ocean-atmosphere simulations of the 20th Century, with and without the effects of aerosols. The atmospheric model has 18 vertical levels and a horizontal resolution of approximately 5.6° in longitude and 3.2° in latitude, and uses a time step of 30 minutes. It is based on an earlier high-resolution version of the Mk3 atmospheric GCM (Gordon et al. 2002), which was substantially upgraded by inclusion of a comprehensive, interactive aerosol scheme and an updated radiation scheme (since the existing radiation scheme could not handle aerosols). The aerosol species treated interactively in the model are sulfate, particulate organic matter (POM), black carbon (BC), mineral dust and sea salt. Historical emission inventories are used for sulfur, POM and BC derived from the burning of fossil fuels and biomass. Natural sources are also included for sulfur (sulfur dioxide from non-eruptive volcanoes and biogenic emissions of dimethyl sulfide (DMS) from oceans) and secondary organic aerosol from terpenes emitted by vegetation. As

CleanAirandEnvironmentalQualityVolume41No.3.August2007 27


well as the direct effects of aerosols on solar radiation, indirect effects of aerosols on cloud properties are also treated (Rotstayn and Liu 2003, 2005). All of the above aerosol types, except for sea salt, are treated prognostically by the model, which means that equations are included for their emission, transformation, transport and removal at each time step. In all, there are 11 prognostic variables in the aerosol scheme: DMS, sulfur dioxide, sulfate, hydrophobic and hydrophilic forms of BC and POM, and four different sizes of mineral dust. A further two modes of sea salt are diagnosed as a function of wind speed in the marine boundary layer, but are not treated prognostically (in the sense that they are not transported, or otherwise processed, by the model). Further details of the aerosol schemes are given by Rotstayn and Lohmann (2002b) and R2007. The oceanic component of the model is based on the Cox-Bryan code (Cox 1984), and has the same horizontal resolution as the atmospheric model, with 21 vertical levels. Further details of the oceanic component and the coupling procedures have been given previously by Gordon and O’Farrell (1997) and Hirst et al. (2000). The CSIRO GCM has participated in numerous international intercomparisons, such as the Coupled Model Intercomparison Project (CMIP), and has generally been found to perform well relative to other models (e.g., Covey et al. 2006).

Simulations were performed for the period 1871-2000. As well as the aerosols described above, forcings included were those due to changes in long-lived greenhouse gases, ozone, volcanic eruptions and solar variations. An eight-run ensemble with all of these forcings (ALL ensemble) and a further eight runs with all forcings except those related to anthropogenic aerosols (AXA ensemble) were performed. The AXA ensemble only differs from the ALL ensemble in that the anthropogenic emissions of sulfur, POM and BC were fixed at their 1870 levels. To isolate the effects of Asian aerosols, another eight-run ensemble was performed, identical to the AXA ensemble, except that Asian anthropogenic emissions of sulfur, POM and BC were allowed to vary with time. R2007 showed that the change in global-mean temperature was much more realistic in the ALL ensemble than in the AXA ensemble, which is consistent with the finding from other groups that inclusion of aerosol forcing improves the simulation of 20th Century temperature trends in GCMs. Also, it is worth noting that R2007 estimated the net (direct plus indirect) global-mean aerosol radiative forcing as -1.2 W m-2 for the period 1870-1990, which agrees well with the best estimate of -1.2 W m-2 from the Fourth Assessment of the Intergovernmental Panel on Climate Change (IPCC; Solomon et al. 2007). Although the time period considered by IPCC was longer (1750-2005), the similarity of the estimates gives some confidence that the CSIRO GCM does not seriously overestimate the global effects of aerosols.

Modelling of aerosols in a GCM is highly challenging and uncertain, and many simplifications are necessary. A

critical assessment of the model’s design and performance was given by R2007. For example, comparisons of modelled distributions of mid-visible aerosol optical depth (AOD) with a satellite-based composite compiled by Kinne et al. (2006) suggested that the model tends to underestimate AOD. However, the satellite retrievals themselves are prone to overestimate AOD (due to contamination by high clouds), and this can exaggerate the impression of an underestimate in the GCM. For a complementary view, it is useful to compare the modelled clear-sky mid-visible AOD with another new climatology, derived by Kinne (2007). This climatology prioritises high-quality measurements from the surface-based Aerosol Robotic Network (AERONET) (Holben et al. 2001), but falls back onto the median output fields from the AeroCom models (Kinne et al. 2006) if AERONET data are unavailable or considered too poor in quality. Thus, the climatology can be called an “AERONET - AeroCom blend”

Figure 1 confirms, for the June-August (JJA) season, the broad finding that this version of the model underestimates AOD for the modern climate. As discussed by R2007, the underestimate in areas strongly affected by anthropogenic aerosol is substantially due to underestimated emissions of carbonaceous aerosols in the historical inventory. Over remote oceans, the underestimate of AOD may be due to the simple treatment of sea-salt aerosol, which assumes that sea salt is well-mixed in the marine boundary layer, but has zero concentration above the boundary layer. However, another possible explanation is that the global production of secondary organic aerosol from natural sources is seriously underestimated in current models (Goldstein and Galbally 2007). In a regional context, it is also worth noting the model’s underestimate of AOD over tropical Australia and Indonesia, which are affected by biomass burning in JJA. The main cause of the problem seems to be an underestimate of these biomass-burning emissions in the historical inventory (Ito and Penner 2005) used by the model. Thus, this version of the model is probably not an effective tool to evaluate any possible impacts of these aerosols on Australian climate.

With the above caveats in mind, it is instructive to look at a map of the simulated trends in small-particle AOD during 1951-1996 from the ALL ensemble (adapted from R2007). Small particles are assumed to be sulfate, POM and BC, so this quantity is a good proxy for anthropogenic AOD (e.g., Kaufman et al. 2005). Figure 2 (adapted from R2007) shows the predominance of anthropogenic aerosol trends in the Northern Hemisphere relative to those in the Southern Hemisphere, and in particular of those over Asia relative to those over Australia. The reason we chose 1996 as the end-point for the analysis is that, based on the emissions inventories we used, 1996 was the year that the average anthropogenic AOD peaked over the region that we defined as “Asia” for the purpose of the study, although the veracity of this assumption is uncertain.

Figure 3 p.30 shows observed and modelled December-February (DJF) rainfall trends for the period 1951-1996. More details, including levels of statistical significance, and annual data, were shown by R2007. The observed trends in Fig. 3a clearly show an increase over the northwest and centre of Australia, which is mainly a summertime phenomenon. The AXA ensemble (Fig. 3b), forced mainly by increasing greenhouse gases, gives a decrease of summer rainfall in the northwest, in common with many other coupled ocean-atmosphere climate models forced by greenhouse gases (e.g., Whetton et al. 2001, who noted that seven out of eight GCMs actually gave a decrease of summer rainfall in the northwest). Inclusion of aerosol forcing in the ALL ensemble (Fig. 3c) gives an increasing trend of rainfall over much of the continent, and greatly improves the simulated rainfall trends in the northwest. The differences between the rainfall trends in the ALL and AXA ensembles (Fig. 3d) show that inclusion of aerosols has also improved the simulated rainfall trend in the northeast, even though the model still gives an increasing trend there (whereas the observations show decreasing rainfall over most of the northeast).

As discussed by R2007, the increase of rainfall in the northwest when aerosol forcing is included correlates with a change in the sign of the trend in meridional surface-temperature gradient in the Indian-Ocean sector. As shown in Fig. 2, aerosol from Asia essentially remains in the Northern Hemisphere: the impacts on Australian tropical rainfall are not due to Asian aerosol being advected over Australia, but occur because the aerosol cools the Asian continent and surrounding oceans, thus altering the meridional temperature gradient between Asia and Australia. (The cooling effect of aerosols is in contrast to the effect of greenhouse-gas forcing, under which the large Asian continent warms strongly.) The physical interpretation of the results over the northeast is less clear. As shown by R2007, the aerosol-induced “improvement” in the modelled trends disappears when annual means are considered, instead of summertime data. On the other hand, there is little statistical significance in the northeast (in either observed or modelled trends), and the impression of an observed drying trend there is exaggerated by the fact that the 1950s was a wet decade over much of eastern Australia. As noted by Nicholls (2006), quantitative studies into possible causes of the rainfall decrease in eastern Australia are currently lacking.

One view of the atmospheric dynamics underlying the modelled rainfall trends is shown by the DJF trends in 220 hPa velocity potential from the model (Figure 4 p.30). Noting that negative values of velocity potential denote regions of divergence, and that 220 hPa is roughly the level of convective outflow in the upper troposphere, we see that a trend towards more negative values corresponds to increasing convection and rainfall. Thus, in the ALL ensemble, the model simulates a shift of convection



and rainfall from near India towards the northwest of Australia. Conversely, in the AXA ensemble, the model simulates reduced convection over most of southeast Asia, the maritime continent, and northwest Australia, consistent with the decreasing rainfall shown for the northwest in Figure 3 p.30. In the ALL ensemble, the trend of reduced convection over India and the northern Indian Ocean is broadly consistent with earlier assessments of the effects of the Indo-Asian aerosol haze (e.g., Ramanathan et al. 2005). To determine whether the response of the CSIRO model to aerosol forcing was due to aerosols from Asia, R2007 performed a sensitivity experiment in which only Asian aerosol sources were allowed to increase with time; this ensemble gave similar rainfall trends over Australia to those in the ALL ensemble, confirming that the effect was indeed due mainly to aerosols from Asia (in the model, at least). Another intriguing result from the simulations performed by R2007 is that in the AXA ensemble, the Pacific SST pattern became more El Niño-like with time, whereas in the ALL ensemble it became more La Niña-like. A possible mechanism to explain this effect was described in the paper. Although this result should be viewed with caution, it clearly merits further study, due to the large impacts of ENSO on rainfall in Australia and many other countries.

Analysis of the above simulations has been extended by Cai et al. (2006) and Cai et al. (2007), with a major focus on oceanic processes involved in the response to aerosol forcing. Cai et al. (2006) found that, even though most of the aerosol forcing occurs in the Northern Hemisphere, the aerosol-induced change in ocean heat content hardly differed between the two hemispheres during the 20th Century. This was because aerosol-forced cooling of the Northern Hemisphere induced a pan-oceanic northward heat transport, with most of it taking place in the Atlantic Ocean. Figure 5 p.32 (reproduced from Cai et al. 2006) shows aerosol-induced trends in vertically-averaged oceanic currents over 0-750 metres depth (vectors) and full-depth averaged temperature (contours). Note that the northward heat transport by the Atlantic “Conveyor” leads to a marked depth-averaged warming (in relative terms) of the subtropical and mid-latitude southern Indian Ocean.

Cai et al. (2007) compared the output from the CSIRO GCM with a new compilation of historical temperature profiles for the Indian Ocean, the Indian Ocean Thermal Archive (IOTA; Alory et al. 2007). They found that the strengthening of the Atlantic Conveyor in the simulation with aerosols increased the warming rate in the southern subtropical Indian Ocean, and by taking heat out of the near-equatorial region, generated a sub-surface cooling there. In both respects, the effects of aerosol forcing improved the agreement of the model with the observed trends from Alory et al. (2007). Figure 6a p.32 shows the observed zonally averaged temperature trends, which are captured much better in the ALL ensemble (Fig. 6b) than in the AXA ensemble (Fig. 6c). This counter-intuitive result is an excellent

Figure 1. Clear-sky aerosol optical depth at 550 nm for June-August from (a) the climatology (AERONET-AeroCom blend), and (b) the CSIRO GCM, using emissions for the year 2000.

Figure 2. 1951-1996 trends in small-particle aerosol optical depth at 550 nm (AOD units per century) from the ALL ensemble. Adapted from Rotstayn et al. (2007) by permission of the American Geophysical Union.



illustration of the global nature of aerosol-induced changes in atmospheric or oceanic circulation. Given that Indian Ocean dynamics and SSTs are known to have important effects on Australian rainfall variability, these effects are potentially of great importance for our understanding of regional climate change.

Using empirical orthogonal function (EOF) analysis, Shi et al. (2007) studied the modes of variability associated with the increasing trend of north-western Australian rainfall in observations and in the low-resolution simulations described above. They found that the process that gives rise to the increasing rainfall trend in the model differs somewhat from the equivalent process in the observations. The key problem they identified in the model is that, because the model’s ENSO is shifted westward by about 20° in longitude, the tropical Pacific warm pool extends into the eastern Indian Ocean. (See, for example, Fig. 14 of R2007). Consequently, the tropical eastern Indian Ocean behaves as a part of the Pacific warm pool, and there is an unrealistic relationship in the Southern-Hemisphere summer between Australian rainfall and the eastern Indian Ocean SST. This emphasises the importance of repeating the above simulations with a higher-resolution model, which has a better simulation of ENSO and its teleconnections to Australian rainfall. A version of the high-resolution (1.9° by 1.9°) CSIRO Mk3.5 model, similar to the one submitted for the IPCC Fourth Assessment, is currently being tested and “tuned”, after inclusion of the aerosol treatments described above. It remains to be seen whether the findings from R2007 are reproduced by this model (or by independent models from other groups). Aside from the resolution of the model, there are probably even larger uncertainties regarding the treatment of aerosol processes, and the historical emissions used to force the model.

PossibleeffectsofAustralian-sourcedaerosol

Various studies have investigated the possibility of effects on climate or local meteorology from aerosols sourced from within Australia, but quantitative modelling is mostly lacking at present. These studies are briefly reviewed here.

Several authors have considered the direct radiative effects of the smoke aerosol produced by biomass burning during the dry season in northern Australia. O’Brien and Mitchell (2003) combined satellite retrievals with surface-based sun-photometer observations to calculate the aerosol radiative heating rates at Jabiru in Kakadu National Park, and found that the radiative heating was substantial, even in comparison with the heating produced by water vapour. These emissions were considered further in a two-part study by Meyer et al. (2007) and Luhar et al. (2007). Meyer et al. constructed a high-resolution inventory of aerosol emissions for the Top End of the Northern Territory for the 2004 dry season (April-November), and validated the emissions using a three-dimensional transport model (TAPM; Hurley et al. 2005) coupled with

a variety of field measurements. Luhar et al. used these emission rates in TAPM, and then compared the modelled particle concentrations and AODs with satellite and ground-based measurements, obtaining generally satisfactory results. Their estimated top-of-atmosphere radiative forcing due to aerosol shows a seasonal mean of -1.8 W m-2, with a region of strong enhancement over the western portion of the Top End. Given that this radiative effect occurs during the dry season, it is unclear whether these emissions have a meaningful effect on regional climate or rainfall, but further study seems warranted. In particular, in view of the deficiencies noted above with regard to the biomass-burning emissions used by the CSIRO GCM over northern Australia, the possibility of extending the emissions inventory of Meyer et al. (2007) to a wider domain seems attractive.

Cloud-physics theory has long suggested the possibility of rainfall suppression in clouds affected by pollution aerosols that act as CCN, since this may lead to smaller cloud droplets and reduced coalescence of droplets (e.g., Rogers and Yau 1988). A number of authors have considered the possibility of rainfall suppression in Australian clouds affected by anthropogenic pollution. Warner and Twomey (1967) analysed data taken upwind and downwind of massive sugar-cane fires in Queensland, and showed that the cloud droplet concentration was clearly increased in clouds affected by the smoke aerosol from the fires. Motivated by this result, Warner (1968) analysed 60 years of rainfall records, and sought a decrease of rainfall downwind of the sugar-cane fires that mirrored the history of expanding cane production. He concluded that such a signal appeared to be evident in the data, but noted that other causative factors could not be eliminated. Warner (1971) performed a more comprehensive analysis, which led him to conclude that the available evidence did not support an association between cane fires and rainfall.

More recently, Rosenfeld (2000) presented satellite retrievals of cloud-droplet effective radius (Re) and precipitation formation for two days over south-eastern Australia, and argued that these showed suppression of precipitation in clouds affected by aerosol pollution. In particular, he suggested that observed decreases in precipitation in the Snowy Mountains during the 20th Century could be due to aerosol pollution. However, Nicholls (2000) showed that an apparent strong downward trend in District Average Rainfall for the Snowy Mountains since 1913 was an artificial result of changes in the mix of stations used to produce the district average. Further, Nicholls (2005) showed that although the snow season in the Australian Alps was shortening, with less snow remaining early in spring, this was due to warming rather than any substantial decline in precipitation. It is also worth noting that IOCI (2002) considered local aerosol pollution in its detailed study of the rainfall decline in southwest Western Australia, and concluded that it was unlikely to be a major contributing factor. New aspects of

Rosenfeld’s analysis were also disputed by Ayers (2005), who noted that atmospheric conditions over south-eastern Australia on the two days considered by Rosenfeld (2000) were not conducive to widespread, significant rainfall. Ayers also showed that the regions of reduced Re shown by Rosenfeld for 21 October 1998 did not correspond well with the modelled distribution of aerosol number concentration generated by TAPM, using an aerosol source function based on a previously established relationship for Australian cities and towns (8 x 1013 particles per second per person; Ayers et al. 1982). Based on this result (and others shown in his paper) Ayers (2005) argued that Rosenfeld’s conclusions were invalid. Rosenfeld et al. (2006) responded to the criticisms of Ayers (2005) in some detail, and noted (for example) that the population-based aerosol source function used in TAPM by Ayers would seriously underestimate emissions from the power stations in the Latrobe Valley. He argued that this aerosol source could explain an area of reduced Re in eastern Victoria (the region denoted as “B” in his Figures 1 and 2). However, as pointed out by Ayers (personal communication, 2007), region B was not downwind of the Latrobe Valley on the day in question, since the wind was from the south-west. In summary, the assertion that aerosol pollution has significantly suppressed rainfall in Australia remains contentious.

In an interesting study, Mitchell et al. (2006) calculated the radiative impact of the aerosol generated by the Canberra firestorm of January, 2003. They estimated mean radiative forcings of -50 W m-2 at the top of the atmosphere and -172 W m-2 at the surface during the week following the firestorm. Note that these are huge numbers, compared to (for example) the global-mean radiative forcing of +1.66 W m-2 due to anthropogenic CO2 estimated by the IPCC (Solomon et al. 2007). For plausible atmospheric profiles, the large optical depths caused heating rates peaking at more than 10ºK per day near the top of the smoke layer. The smoke layer stabilised the lower troposphere through a combination of reduced surface heating and a positive gradient in heating rate through most of the layer. The enhanced stability suppressed surface temperatures and prevented the north-westerly winds from penetrating to the surface during the week following the firestorm. This reduced the risk of renewed fire danger, and suggests an important role for inclusion of aerosol effects in weather prediction models. Herring and Hobbs (1994) described similar dynamical feedbacks due to the smoke plume from the 1991 Kuwait oil fires. Anecdotally, a similar effect may have also occurred during the recent Victorian bushfires, when forecast strong winds didn’t arrive on 10 December 2006, leading Peter McHugh, of the Department of Sustainability and Environment, to express frustration at the inaccurate weather forecast: “We set ourselves up on the basis of expected strong north-westerly winds and high temperatures, which haven’t materialised. So, a lot of energy went into preparing for that, and the weather didn’t come...” (reported on

colour pages



ABC AM, 11 December 2006; http://www.abc.net.au/am/content/2006/s1808501.htm). Although bushfires are episodic by nature, the large radiative effects reported by Mitchell et al. (2006) suggest that this topic deserves further research, and that inclusion of smoke aerosol in weather-prediction models could substantially improve critical weather forecasts during bushfires.

CoNClusioNs

The paucity of research on the effects of aerosols on Australian climate and weather is in stark contrast to the situation in the Northern Hemisphere. In recent years, a large number of studies on aerosol-climate interactions have been published by groups based in the Northern Hemisphere (e.g.,

see the review by Menon 2004), but few of these consider effects on Australian climate. Several Australian-based studies reviewed in the preceding paragraphs suggest myriad possibilities regarding how aerosols may affect our climate and weather. However, in terms of quantifying these effects, they barely scratch the surface. Because the sky is generally blue in Australia, the effects of global aerosol pollution on our climate may have been under-recognised until recently. However, as shown in the studies by Rotstayn et al. (2007), and Cai et al. (2006, 2007), our climate is potentially susceptible to aerosol forcing from outside our immediate region. The specific attribution of the increase in north-western Australian rainfall to the Asian aerosol haze should be regarded with caution. Large uncertainties exist regarding our understanding of key aerosol processes, their simplified treatment in GCMs, historical emissions inventories and the low resolution of the model. Also, the question of whether the observed increase in rainfall could be due to natural variability has not been resolved. However, the broad finding that Australian rainfall trends are sensitive to the inclusion of aerosol forcing from the Northern Hemisphere (and especially Asia) is likely to be robust. Some of the postulated effects of Australian-sourced aerosol on climate or local weather are similarly intriguing, and possibly important. Hopefully, a coordinated and adequately funded research program will provide more definite answers to these questions in the next few years.

ACkNowlEDgEmENTs

This work was funded in part by the Australian Greenhouse Office, through the Australian Climate Change Science Program. The author thanks Peter Manins and Greg Ayers for their constructive comments on the manuscript, and Martin Dix for his help with the calculations required for Figure 4.

Figure 4. December-February modelled trends in velocity potential at 220 hPa (106 m-2 per century) from (a) the ALL ensemble and (b) the AXA ensemble during 1951-1996. Cool (warm) colours denote regions of increasing (decreasing) convection.

Figure 3. December-February seasonal rainfall trends (mm per century) over the Australian region for the period 1951 to 1996 from observations and from the CSIRO GCM. Modelled trends shown are (b) the AXA ensemble mean, (c) the ALL ensemble mean, and (d) the difference between the ALL and AXA ensemble means. Observed trends are based on a least-squares fit to gridded data from CRU TS 2.1 (Mitchell and Jones 2005).



rEfErENCEs

Alory, G., S. Wijffels, and G. Meyers, 2007: Observed temperature trends in the Indian Ocean over 1960-1999 and associated mechanisms. Geophysical Research Letters, 34, L02606, doi: 10.1029/2006GL028044.

Ayers, G. P., 2005: Air pollution and climate change: Has air pollution suppressed rainfall over Australia? Clean Air and Environmental Quality, 39(2), 51-57.

Ayers, G. P., E. K. Bigg, D. E. Turvey, and M. J. Manton, 1982: Urban influences on condensation nuclei over a continent. Atmospheric Environment, 16, 951-954.

Cai, W., D. Bi, J. Church, T. Cowan, M. R. Dix, and L. D. Rotstayn, 2006: Pan-oceanic response to increasing anthropogenic aerosols: impacts on the Southern Hemisphere oceanic circulation. Geophysical Research Letters, 33, L21707.

Cai, W., T. Cowan, M. R. Dix, L. D. Rotstayn, J. Ribbe, G. Shi, and S. Wijffels, 2007: Anthropogenic aerosol forcing and the structure of temperature trends in the Southern Indian Ocean. Geophysical Research Letters, 34, doi: 10.1029/2007GL030380. In press.

Covey, C., P. J. Gleckler, T. J. Phillips, and D. C. Bader, 2006: Secular trends and climate drift in coupled ocean-atmosphere general circulation models. Journal of Geophysical Research, 111, D3107, doi:10.1029/2005JD006009.

Cox, M. D., 1984: A primitive equation, three-dimensional model of the ocean. Technical Report 1, Geophysical Fluid Dynamics Laboratory Ocean Group, Princeton, NJ, USA. 143 pp.

Goldstein, A. H., and I. E. Galbally, 2007: Known and unexplored organic constituents in the earth’s atmosphere. Environmental Science and Technology, 41, 1514-1521.

Gordon, H. B., and S. P. O’Farrell, 1997: Transient climate change in the CSIRO coupled model with dynamic sea ice. Monthly Weather Review, 125, 875-907.

Gordon, H. B., L. D. Rotstayn, J. L. McGregor, M. R. Dix, E. A. Kowalczyk, S. P. O’Farrell, L. J. Waterman, A. C. Hirst, S. G. Wilson, M. A. Collier, I. G. Watterson, and T. I. Elliott, 2002: The CSIRO Mk3 climate system model. Technical Paper No. 60, CSIRO Atmospheric Research, Aspendale, Vic., Australia. 134 pp. Available online at http://www.cmar.csiro.au/e-print/open/gordon_2002a.pdf.

Herring, J., and P. Hobbs, 1994: Radiatively driven dynamics of the plume from 1991 Kuwait oil fires. Journal of Geophysical Research, 99, 18,809-18,826.

Hirst, A. C., S. P. O’Farrell, and H. B. Gordon, 2000: Comparison of a coupled ocean-atmosphere model with and without oceanic eddy-induced advection. Part I: ocean spinup and control integrations. Journal of Climate, 13, 139-163.

Holben, B. N., D. Tanré, A. Smirnov, T. F. Eck,

I. Slutsker, N. Abuhassan, W. W. Newcomb, J. S. Schafer, B. Chatenet, F. Lavenu, Y. J. Kaufman, J. V. Castle, A. Setzer, B. Markham, D. Clark, R. Frouin, R. Halthore, A. Karneli, N. T. O’Neill, C. Pietras, R. T. Pinker, K. Voss, and G. Zibordi, 2001: An emerging ground-based aerosol climatology: Aerosol optical depth from AERONET. Journal of Geophysical Research, 106, 12,067-12,098.

Hurley, P. J., W. L. Physick, and A. K. Luhar, 2005: TAPM: a practical approach to prognostic meteorological and air pollution modelling. Environmental Modelling and Software, 20, 737-752.

IOCI, 2002: Climate variability and change in southwest Western Australia. Technical report, Indian Ocean Climate Initiative Panel, Perth, Western Australia. 43 pp. Available online at http://www.ioci.org.au/publications/pdf/IOCI_TechnicalReport02.pdf.

Ito, A., and J. E. Penner, 2005: Historical emissions of carbonaceous aerosols from biomass and fossil fuel burning for the period 1870-2000. Global Biogeochemical Cycles, 19, GB2028, doi:10.1029/2004GB002374.

Kaufman, Y. J., O. Boucher, D. Tanré, M. Chin, L. A. Remer, and T. Takemura, 2005: Aerosol anthropogenic component estimated from satellite data. Geophysical Research Letters, 32, L17804, doi:10.1029/2005GL023125.

Kinne, S., 2007: Aerosol direct radiative forcing with an AERONET touch. Atmospheric Environment. submitted.

Kinne, S., M. Schulz, C. Textor, S. Guibert, Y. Balkanski, S. E. Bauer, T. Berntsen, T. F. Berglen, O. Boucher, M. Chin, W. Collins, F. Dentener, T. Diehl, R. Easter, J. Feichter, D. Fillmore, S. Ghan, P. Ginoux, S. Gong, A. Grini, J. Hendricks, M. Herzog, L. Horowitz, I. Isaksen, T. Iversen, A. Kirkevåg, S. Kloster, D. Koch, J. E. Kristjansson, M. Krol, A. Lauer, J. F. Lamarque, G. Lesins, X. Liu, U. Lohmann, V. Montanaro, G. Myhre, J. Penner, G. Pitari, S. Reddy, O. Seland, P. Stier, T. Takemura, and X. Tie, 2006: An AeroCom initial assessment - optical properties in aerosol component modules of global models. Atmospheric Chemistry and Physics, 6, 1815-1834.

Liepert, B. G., J. Feichter, U. Lohmann, and E. Roeckner, 2004: Can aerosols spin down the water cycle in a warmer and moister world? Geophysical Research Letters, 31, L06207.

Luhar, A. K., R. M. Mitchell, C. P. M. Meyer, S. Campbell, J. L. Gras, and D. Parry, 2007: Biomass burning emissions over northern Australia constrained by aerosol measurements: II - model validation, and impacts on air quality and climate. Atmospheric Environment. Submitted.

Menon, S., 2004: Current Uncertainties in Assessing Aerosol Effects on Climate. Annual Review of Environment and Resources, 29(1), 1-30.

Menon, S., J. Hansen, L. Nazarenko, and Y. Luo, 2002: Climate effects of black carbon aerosols in China and India. Science, 297, 2250-2253.

Meyer, C. P. M., A. K. Luhar, and R. M. Mitchell, 2007: Biomass burning emissions over northern Australia constrained by aerosol measurements: I - modelling the distribution of hourly emissions. Atmospheric Environment. Submitted.

Mitchell, J. F. B., T. C. Johns, J. M. Gregory, and S. F. B. Tett, 1995: Climate response to increasing levels of greenhouse gases and sulphate aerosols. Nature, 376, 501-504.

Mitchell, R. M., D. M. O’Brien, and S. K. Campbell, 2006: Characteristics and radiative impact of the aerosol generated by the Canberra firestorm of January 2003. Journal of Geophysical Research, 111, 2204, doi:10.1029/2005JD006304.

Mitchell, T. D., and P. D. Jones, 2005: An improved method of constructing a database of monthly climate observations and associated high-resolution grids. International Journal of Climatology, 25, 693-712.

Nazarenko, L., and S. Menon, 2005: Varying trends in surface energy fluxes and associated climate between 1960-2002 based on transient climate simulations. Geophysical Research Letters, 32, L22704.

Nicholls, N., 2000: An artificial trend in District Average Rainfall in the Snowy Mountains. Australian Meteorological Magazine, 49, 255-258.

Nicholls, N., 2005: Climate variability, climate change and the Australian snow season. Australian Meteorological Magazine, 54, 177-185.

Nicholls, N., 2006: Detecting and attributing Australian climate change: a review. Australian Meteorological Magazine, 55, 199-211.

O’Brien, D. M., and R. M. Mitchell, 2003: Atmospheric heating due to carbonaceous aerosol in northern Australia - confidence limits based on TOMS aerosol index and sun-photometer data. Atmospheric Research, 66, 21-41.

Ramanathan, V., C. Chung, D. Kim, T. Bettge, L. Buja, J. T. Kiehl, W. M. Washington, Q. Fu, D. R. Sikka, and M. Wild, 2005: Atmospheric brown clouds: Impacts on South Asian climate and hydrological cycle. Proceedings of the National Academy of Sciences, 102, 5326-5333.

Roderick, M. L., and G. D. Farquhar, 2002: The cause of decreased pan evaporation over the past 50 years. Science, 298, 1410-1411.

Rogers, R. R., and M. K. Yau, 1988: A short course in cloud physics. Pergamon Press, Oxford, third edition. 293 pp.

Rosenfeld, D., 2000: Suppression of rain and snow by urban and industrial air pollution. Science, 287, 1793-1796.

Rosenfeld, D., I. M. Lensky, J. Peterson, and A. Gingis, 2006: Potential impacts of air pollution aerosols on precipitation in Australia. Clean Air and Environmental Quality, 40(2), 43-49.

Rotstayn, L. D., and Y. Liu, 2003: Sensitivity of



radiello®

Quick Diffusive Air Sampler for fast Industrial Hygiene and Ambient Air Sampling.

sigma-aldrich.com/radielloradiello® samplers are now globally available from Sigma-Aldrich/Supelco.

The patented radiello® sampler of theFondazine Salvatore Maugeri, Padua/Italy provides:

• High sampling ratesdue to radial design 3-8x faster sampling rates compared commonaxial diffusive sampler

• High capacitymore precise results

• Economic usereusable hardware

• Robustfor both indoor and outdoor sampling

SIGMA-ALDRICH PTY LTD • PO BOX 970 CASTLE HILL NSW 1765 • PHONE 1800 800 097 • FREE FAX 1800 800 096 • EMAIL ausmai l@s ia l .comSIGMA-ALDRICH NEW ZEALAND • PO BOX 106-406 AUCKLAND 1030 • PHONE 0800 936 666 • FAX 0800 936 777 • EMAIL ausmai l@s ia l .com

radiellothe first indirect aerosol effect to an increase of cloud droplet spectral dispersion with droplet number concentration. Journal of Climate, 16, 3476-3481.

Rotstayn, L. D., and Y. Liu, 2005: A smaller global estimate of the second indirect aerosol effect. Geophysical Research Letters, 32, L05708, doi:10.1029/2004GL021922.

Rotstayn, L. D., and U. Lohmann, 2002a: Tropical rainfall trends and the indirect aerosol effect. Journal of Climate, 15, 2103-2116.

Rotstayn, L. D., and U. Lohmann, 2002b: Simulation of the tropospheric sulfur cycle in a global model with a physically based cloud scheme. Journal of Geophysical Research, 107, 4592, doi:10.1029/2002JD002128.

Rotstayn, L. D., W. Cai, M. R. Dix, G. D. Farquhar, Y. Feng, P. Ginoux, M. Herzog, A. Ito, J. E. Penner, M. L. Roderick, and M. Wang, 2007: Have Australian rainfall and cloudiness increased due to the remote effects of Asian anthropogenic aerosols? Journal of Geophysical Research, 112, D09202, doi:10.1029/2006JD007712.

Shi, G., W. Cai, T. Cowan, J. Ribbe, L. Rotstayn, and M. Dix, 2007: Variability and trend of the northwest Western Australia rainfall: observations and coupled climate modeling. Journal of Climate. Submitted.

Solomon, S., D. Qin, M. Manning, R. B. Alley, T. Berntsen, N. L. Bindoff, Z. Chen, A. Chidthaisong, J. M. Gregory, G. C. Hegerl, M. Heimann, B. Hewitson, B. J. Hoskins, F. Joos, J. Jouzel, V. Kattsov, U. Lohmann, T. Matsuno, M. Molina, N. Nicholls, J. Overpeck, G. Raga, V. Ramaswamy, J. Ren, M. Rusticucci, R. Somerville, T. F. Stocker, P. Whetton, R. A. Wood, and D. Wratt, 2007: Technical Summary. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H. L. Miller, Eds., Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

Twomey, S., 1974: Pollution and the

planetary albedo. Atmospheric Environment, 8, 1251-1256.

Warner, J., 1968: A reduction in rainfall associated with smoke from sugar-cane fires - An inadvertent weather modification. Journal of Applied Meteorology, 7, 247-251.

Warner, J., 1971: Smoke from sugar-cane fires and rainfall. Proceedings, International Conference on Weather Modification, Canberra, American Meteorological Society, 191-192.

Warner, J., and S. Twomey, 1967: The production of cloud nuclei by cane fires and the effect on cloud droplet concentration. Journal of the Atmospheric Sciences, 24, 704-706.

Whetton, P. H., J. J. Katzfey, K. J. Hennessy, X. Wu, J. L. McGregor, and K. Nguyen, 2001: Developing scenarios of climate change for southeastern Australia: an example using regional climate model output. Climate Research, 16, 181-201.

Wild, M., H. Gilgen, A. Roesch, A. Ohmura,

C. N. Long, E. G. Dutton, B. Forgan, A. Kallis, V. Russak, and A. Tsvetkov, 2005: From Dimming to Brightening: Decadal Changes in Solar Radiation at Earth’s Surface. Science, 308, 847-850.

AuThor

Dr Leon David Rotstayn, CSIRO Marine and Atmospheric Research, Private Bag 1, Aspendale, Victoria 3195, Australia E-mail: [email protected]

Figure 5. Aerosol-induced trends in vertically-averaged oceanic currents over 0-750 m (vectors; maximum vector, shown at lower right, is 1 cm-2 per century) and vertically averaged temperature (contours, K per century). “Aerosol-induced” refers to the difference between the ALL and AXA ensembles, during 1951- 2000. Reproduced from Cai et al. (2006) by permission of the American Geophysical Union.

Figure 6. Trends over 1951-2000 in zonal-mean temperature of the upper 800 metres of the Indian Ocean from (a) IOTA observations, (b) the ALL ensemble, and (c) the AXA ensemble, in units of K per 50 years. Adapted from Cai et al. (2007) by permission of the American Geophysical Union.

PossiblE imPACTs of PolluTioN AErosols oN …PossiblE imPACTs of PolluTioN AErosols oN AusTrAliAN ClimATE AND wEAThEr well as the direct effects of aerosols on solar radiation, indirect

Documents