Observed slump of sea land breeze in Brisbane under the 1 effect of aerosols from remote transport during 2019 2 Australia mega fire events

. The 2019 Australia mega fires were unprecedented considering its intensity and consistency. 8 There have been many researches on the environmental and ecological effects of this mega fires, most 9 of which focused on the effect of huge aerosol loadings and the ecological devastation. Sea land breeze 10 (SLB) is a regional thermodynamic circulation closely related to coastal pollution dispersion yet few 11 have looked into how it is influenced by different types of aerosols transported from either nearby or 12 remote areas. Mega fires provide an optimal scenario of large aerosol emissions. Near the coastal site 13 of Brisbane Archerfield during January in 2020 when mega fires were the strongest, reanalysis data 14 from Modern-Era Retrospective analysis for Research and Applications version 2 (MERRA-2) showed 15 that mega fires did release huge amounts of aerosols, making aerosol optical depth (AOD) of total 16 aerosols, black carbon (BC) and organic carbon (OC) approximately 240%, 425%, 630% of the 17 averages in other non-fire years. Using 20 years’ wind observations of hourly time resolution from 18 global observation network managed by National Oceanic and Atmospheric Administration (NOAA), 19 we found that the SLB day number during that month was only four, accounting for 33.3% of the 20 multi-years’ average. The land wind (LW) speed and sea wind (SW) speed also decreased by 22.3% 21 and 14.8% compared with their averages respectively. Surprisingly, fire spot and fire radiative power 22 (FRP) analysis showed that heating effect and aerosol emission of the nearby fire spots were not main 23 causes of local SLB anomaly while the remote transport of aerosols from the fire center was mainly 24 responsible for the decrease of SW, which was


35
There are different types of aerosols from various sources which have different climatological forcing 36 effects (Charlson, 1992;Yang et al., 2016). Aerosols differ in radiative forcing effects as their physical 37 and chemical properties vary, some of which may affect the earth-atmosphere system by bringing 38 changes to the lifespan of clouds (Albrecht, 1989;Zhao and Garrett, 2015).

39
Carbonaceous aerosol contains black carbon (BC) and organic carbon (OC) and serves as a major

43
There are also some studies trying to quantify the average radiative forcing effects of BC and OC while 44 they also emphasized the potential uncertainties with respect to the specific values .

45
At a planetary scale, the change of aerosols brings many uncertainties to radiation balance thus further 46 influences the magnitude of atmospheric circulation (Wang et al., 2015;Zhao et al., 2020). At a 47 synoptic scale, aerosols can affect tropical cyclone by enlarging its rainfall areas which is also related 48 to their radiative properties . At a regional scale, Han et al. (2020)  of the total fire-induced aerosols over the globe, which was estimated to be -1.0 W/m 2 on average. The 59 fire-induced aerosols could have more significant radiative effects with clouds than under clear sky 60 condition through cloud-aerosol interaction, whose global forcing effect could reach -1.16 W/m 2 61 (Chuang et al., 2002).

78
Sea land breeze (SLB) is a common circulation over coastal areas whose direct cause is the regional 79 temperature difference between land and sea (TDLS). Many studies have investigated this regional 80 circulation. On one hand, the complicated influencing factors of SLB have been studied from different 81 perspectives (Miller et al., 2013). Our previous studies pointed out that the change of TDLS is highly 82 related to the change of in situ downwelling solar radiation (Shen et al., 2021a, b; Shen and Zhao, 83 2020). We also found that the continuous increase of surface roughness in cities can reduce the SLB 84 speed in long term (Shen et al., 2019). The long-term significance and trends of SLBs over the globe 85 are driven by climate regimes which are related to climatological differences in both in situ 86 downwelling solar radiation and background wind fields. There are also many other studies on the 87 4 influencing factors of SLB in short periods. For example, based on the case analysis, Sarker et al. (1998) 88 found that UHI magnitude has a great impact on the encroachment range of sea wind (SW) frontal 89 surface. Using regional model simulation, Ma et al. (2013) found that UHI effect can greatly enhance 90 TDLS which would result in strengthened SLB circulation in a great metropolis. Miller et al. (2013) 91 reviewed the studies on SLB and pointed out that local topography such as the shape of the coastline, is 92 another important influencing factor of SLB. On the other hand, SLB's effect has also been extensively 93 investigated. For example, SLB has been reported as a direct controller of air pollutants which 94 transports air pollutants inland or to the vast ocean with the help of background meteorological field 95 (Nai et al., 2018;Shen and Zhao, 2020). SLB is also essential to the modification of the meteorological 96 conditions and local climate (Rajib and Heekwa, 2010). Moreover, SLB is a determinant factor of the 97 diurnal variation of the precipitation on the island since its direction and magnitude can affect the 98 location and magnitude of convective systems (Zhu et al., 2017).

99
Over the years, the cause and effect of aerosols, wild fires in typical areas, and SLBs have been learned 100 in detail respectively. The relationship between aerosols and other small scale circulations such as UHI 101 circulation has also been investigated from many aspects (Han et al. 2020). However, few studies have 102 investigated the effects of different types of aerosols on SLBs or looked into how local and remote 103 aerosol emissions during mega fires would affect local SLB with the help of meteorological 104 background field or other potential mechanisms. There was an updated and important study calling for 105 attention of the record-breaking aerosol emissions during 2019 Australia mega fires which led to 106 significant cooling effect on ocean temperature (Hirsch and Koren, 2021). Since in situ downwelling 107 solar radiation and SST, which are both important influential factors of SLB, are deeply affected by 108 different types of aerosols due to their different radiative properties, it is interesting to examine in detail 109 how the record-breaking mega fires would influence SLB by releasing large amounts of aerosols.

110
The paper is organized as follows. Section 2 describes the observation site, data and analysis methods.

111
Section 3 illustrates the characteristics of SLB, the variation of SLB days, the distribution and fire 112 radiative power (FRP) of wild fire spots, the anomaly of observed SW speed, land wind (LW) speed 113 and air temperature, the effects of different aerosols on SLB's variation, the analysis on background 114 wind field and the comparison between local fire spots' and the remote fire center's contributions.

115
Section 4 summarizes and discusses the findings of the study and proposes a mechanism of 116 aerosol-SLB interaction during the peak of 2019 Australia mega fires. SLB might not be accurate since it is likely to contain other wind disturbance at a small regional scale.

134
As shown in Fig. 1, we selected an urban site in Brisbane along the eastern coast of Australia as the 135 study site, which was due to several considerations. First, alongside the eastern coastal areas of 136 Australia which belong to monsoon climate, including Brisbane and areas to its south but to the north 137 of the fire center, the Australian monsoon system is not strong so that the OE-SLB can be verified from 138 a climatological perspective, which also means integrated SLB circulation can be found during all

325
In order to be more accurate, we carried out linear regression between temperature during PTL and LW 326 anomaly and found that they had negative linear relationship (p<0.02) with each other ( Figure 5). As

489
During nighttime, it has a warming effect on both atmosphere and Earth surface through longwave 490 radiation. As a result, it has a warming effect on the Earth-atmosphere system including the surface of 491 the regional land-sea system so that there was a temperature soar shown in Figure 4b. The soaring BC 492 during the mega fire heated the local atmosphere, which was like adding a 'heater' in the air. The 493 'heater' then gave out downward longwave radiation to the regional land-sea system. Just like the sun 494 during daytime, this could trigger a SW circulation anomaly, weakening LW circulation. Considering 495 the BC burst during mega fires, it is nothing weird about its dominant role in local land temperature 496 18 increase during nighttime. The mechanism proposed above can be summarized as follows. During 497 nighttime, the formation of LW originates from the process of heat release from both land and sea. As 498 they both lose heat with different paces due to different heat capacities, the TDLS is enlarged. During 499 the mega fires, the upper atmosphere of the regional land-sea system is heated so that the vertical 500 temperature gradient is weakened, which is unfavorable for heat release from both sea and land 501 surfaces. As a result, the TDLS is significantly weakened.

502
Another potential contributing accelerator is CO2 which is also the product of fires due to the 503 combustion of plants and trees. CO2 is a kind of greenhouse gas which is likely to be engaged in the 504 same mechanism as BC to reduce TDLS during nighttime except that CO2 cannot affect the 505 downwelling solar radiation. Details about this is not repeated again. However we should note that the 506 effect of CO2 is based on theoretical analysis rather than observational verification due to the lack of 507 accurate observation data. Both BC and CO2's warming effects increase TDLS during daytime, which 508 partially offset the strong negative radiative forcing effect of total aerosols, but their combined 509 warming effect is more significant during nighttime than during daytime. That is most likely the reason 510 (at least partially) that SW speed had negative anomaly but was less significant than LW speed.

511
What we discussed above are all factors whose influences were restrained to a small scale. Although 512 SLB is a small scale system, it can still be affected by the variations of signals in a large scale, since the 513 local temperature is affected by both regional forcing and the variation of large scale background 514 temperature field. In our previous study, we weighed their contributions qualitatively (Shen et al., 515 2019). In this study, we simply discuss the potential effect of the change in large scale SST. Hirsch and 516 Koren (2021) emphasized the effect of record-breaking aerosol emission from this mega fire on cooling 517 the oceanic areas. On a large scale, its average radiative forcing on sea surface was -1.0 ± 0.6 W/m 2 .

518
The temperature decrease of large scale sea surface could have negative forcing on the SST at a 519 regional scale, though the specific temperature variation of the sea surface where the SLB vertical 520 stream lies might not be the same.

521
We summarized all the influencing factors of TDLS at both regional and large scales in Table 2. Among 522 all these factors, aerosols, BC, OC and CO2 had direct forcing on TDLS by changing the solar radiation 523 reaching the regional sea-land system. In contrast, heating effect of fire spots and large scale SST 524 signal had forcing on land temperature and regional SST respectively thus further had different forcing 525 effects on TDLS during daytime and nighttime. During 2019 Australia mega fires, TDLS during 526 19 daytime and nighttime both decreased under their combined forcing effects, which could be inferred 527 from the anomalies of SLB speed. Clearly, the directions of all forcing effects of different factors were 528 the same during nighttime. That was why LW speed decreased much more significantly than SW speed 529 did. The negative radiative forcing effect of total aerosols was the determinant cause for TDLS 530 decrease during daytime, which could only be partially offset by other factors.

541
As Figures 6, 8, 11 and 12 show, the distributions of fire spots, TA-AOD, BC-AOD and OC-AOD were 542 quite similar as each other. In the fire center, both the density and FRP of fire spots were much higher 543 in January of 2020 than in January of other years, which are all based on distribution characteristics at a 544 large scale. In order to show the fire situation at the fire center more accurately, we magnified the FRP 545 map to restrain the areas to merely the fire center, which is shown in Figure 13. As shown, the fire spot 546 density was quite high in this region, especially along coastal areas. Compared with other areas, the fire 547 center had much more fire spots with higher FRP. The spots with FRP from 235 to 864 MW were 548 evenly distributed in all fire areas, surrounded by low FRP spots with high density. There were quite a 549 few spots with even higher FRP ranging from 864 to 2,194 MW, which could not be found in other 550 periphery areas (Figure 7a). In some areas at the fire center, we could even find fire spots with FRP

632
Since the period b was the longest among all No-SLB periods, it did not necessarily mean that the 633 wind's aerosol transport effect during this period was less than those during other periods although the 634 wind flows were not directly from the fire center. The moving paths of them were similar as those of 635 wind flows in Figure 15b, which all had an abrupt turning on the Pacific to the southeast of the site.

636
This is probably because that the south hemisphere's subtropical high developed to be quite strong 637 during the middle of summer, making the pressure gradient exist both at 500 m and 3 km (Figure 14). 638 Figure 16f shows the contribution of wind flows on monthly average, whose clustering number was 639 also four. There were four main directions of wind flows, whose contribution were 28.67%, 21.86%, 640 11.47% and 37.99% respectively. In order to make it clear, we define these four main wind flows as 3). Emissions from fire center were mainly responsible for the local positive aerosol anomaly during 675 mega fires. On average, the background wind fields from near surface to 3 km were unfavorable for 676 aerosol and CO2 transport. But there were likely aerosol and CO2 transports through large scale wind 677 field at single levels during shorter periods within January of 2020. Specifically, the wind flow 678 transport at 3 km was stronger than that at 500 m, which was particularly important for smoke transport 679 since the smoke from fires gathered at the same level. In general, free diffusion due to large 680 concentration gradient was mainly responsible for aerosol transport and the potential CO2 transport 681 while the effect of background wind field played a second role.

682
In order to make it clear and concise to the influencing factors of SLB, we summarized their potential 683 mechanisms in local sea-land system ( Figure 17). During daytime, negative anomaly of SW speed was 684 found at the site in January of 2020 when Australia mega fires were most intensive. The local cloud 685 fraction and COD were almost on an average level while there were much more aerosols during mega 686 fires, which mainly came from fire center by free diffusion. They significantly weakened the in situ 687 downwelling solar radiation thus further narrowed the TDLS, which was the direct cause of SW speed 688 decrease. BC and CO2 heated the atmosphere and warmed the earth-atmosphere system by longwave 689 radiation from the heated atmosphere. Warming effect of BC and CO2, the decrease of SST at a large 690 scale and the weak heating effect of nearby fire spots partially offset the effect of aerosols on narrowing 691 TDLS, making the negative SW speed anomaly not exceed the multi-years' oscillation range. During 692 nighttime, the heating effect of nearby fire spots was still weak but more significant than that during