Urbanization-induced land and aerosol impacts on sea breeze circulation and convective 1 precipitation 2 3

Urbanization-induced land and aerosol impacts on sea breeze circulation and convective 1 precipitation 2 3 Jiwen Fan1, *, Yuwei Zhang1, *, Zhanqing Li2, Jiaxi Hu3, and Daniel Rosenfeld4 4 5 1 Atmospheric Sciences and Global Change Division, Pacific Northwest National Laboratory, 6 Richland, WA, USA 7 2 Department of Atmospheric and Oceanic Science, University of Maryland, College Park, MD, 8 USA 9 3 University of Oklahoma, and NOAA/OAR National Severe Storms Laboratory, Cooperative 10 Institute for Mesoscale Meteorological Studies, Norman, OK, USA 11 4 Institute of Earth Sciences, The Hebrew University of Jerusalem, Jerusalem, Israel 12

2a, c). Based on LandAero, sensitivity tests are conducted to investigate the combined and 173 individual effects of urban land and anthropogenic aerosols. No_Aero is the simulation based on 174 LandAero, except that anthropogenic emissions are turned off and the initial and boundary 175 chemical and aerosol conditions are from the Domain 1 simulation without anthropogenic 176 aerosols considered (Fig. 2b). No_Land is also based on LandAero, except the Houston urban 177 land is replaced by the surrounding cropland and pasture (Fig. 2d). The aerosols used in 178 No_Land include the anthropogenic sources (Fig. 2a), which is analogous to the scenario of 179 downwind a big city (i.e., rural area with pollution particles transported from city). We also run a 180 simulation with both the urban land cover replaced by the surrounding cropland and the 181 anthropogenic aerosols excluded (Fig. 2b, d), which is referred to as "No_LandAero". That is, To apply the algorithm to both model simulation and NEXRAD observations 219 consistently in this study, we calculated liquid water path (LWP), a variable of model output 220 accounting for the column integrated liquid to replace VIL in MCIT for model simulation. We 221 track local maxima of LWP by identifying the two cells in consecutive radar scans that have 222 maximum common LWP. A cell is identified and tracked when the local maxima LWP exceeds 223 50 g m -2 . This value is selected because it allows us to start recognizing the deep convective cell 224 by filtering a lot of shallow clouds surrounded it. The storm area of the tracked cell is defined as 225 the grid area with LWP > 50 g m -2 . 226 To examine sea breeze circulation over the Houston region, the sea breeze wind intensity 227 at a specific time is calculated by averaging the horizontal wind speeds below 1-km altitude 228 along the black line UO in Fig. 4a. The cross section of the winds along this line is also analyzed 229 in the result section. 230 3 Results 231

Radar reflectivity, precipitation, and convective intensity 232
We first discuss the evaluation of the baseline simulation LandAero first. The simulation 233 is comprehensively evaluated in Zhang et al. (2020). Here the comparisons with observed radar 234 reflectivity and precipitation are included. The composite radar reflectivity at the time of the 235 peak reflectivity of the storm in Houston shows that LandAero captures the convective cell in 236 Houston, with the maximal radar reflectivity of 58 dBZ, very close to the observed 57 dBZ (Fig.  237 5a, b). The modeled convective cell in LandAero has a larger size compared with the radar 238 observations. The contoured frequency by altitude diagram (CFAD) over the major storm period 239 (1800 UTC 19 Jun to 0000 UTC 20 Jun) shows that the model overestimates the frequencies of 240 moderate reflectivity (i.e., 15-35 dBZ) over the entire vertical profile (Fig. 6a-b), but captures the 241 occurrence frequencies of high reflectivity (larger than 45 dBZ) reasonably well. The magnitude 242 of the surface rain rate averaged over the study area defined by the red box in Fig. 5 from 243 LandAero agrees with the retrieved value from the NEXRAD reflectivity, with a peak time about 244 40 min earlier than the observation (Fig. 7a). The probability density function (PDF) of rain rates 245 shows that LandAero reproduces the occurrence frequencies of low and mediate rain rates well 246 (left two columns in Fig. 7b) and overestimates the occurrence frequencies of high rain rates (> 247 10 mm h -1 ; right two columns in Fig. 7b). The accumulated precipitation over the time period 248 shown in Fig. 7a is about 7.2 mm from LandAero and 5.5 mm from observations, with a model 249 overestimated of ~ 30% because of the overestimation of occurrences of high rain rates and 250 longer precipitation period. 251 Without Houston urbanization (i.e., both effects of urban land and anthropogenic aerosol 252 are removed), the Houston convective cell is a lot smaller in area and has reflectivity values of ~ 253 7 dBZ lower in general compared with LandAero and the NEXRAD observation ( Fig. 5c vs. 5a-254 b). There is almost no radar reflectivity larger than 50 dBZ in No_LandAero (Fig. 6c), in 255 contrast with the significant occurrences of reflectivity larger than 50 dBZ in LandAero and the 256 NEXRAD observation. Those differences are more clearly shown in Fig. 6f. The peak surface 257 rain rate in No_LandAero is reduced by ~ 45% compared with LandAero and observations (Fig.  258 7a; black vs. red line), with the occurrences of large rain rates (> 15 mm h -1 ) reduced by nearly 259 an order of magnitude (Fig. 7b). In terms of updraft intensity, the CFAD plots in Fig. 8a-b show 260 that there is extremely low or no occurrence for updraft velocity larger than 15 m s -1 in 261 No_LandAero, while the occurrences of 30 m s -1 still exist in LandAero. There are less 262 occurrences of weak updraft velocities and more occurrences of relatively strong updraft 263 velocities over the vertical profile (Fig. 8e). These results indicate the urbanization (i.e., the joint 264 urban land and aerosol effects) drastically enhances the convective intensity and precipitation. 265 Now let's look at the individual effect from the Houston urban land and anthropogenic 266 aerosols. Fig. 5 shows that the urban land effect enlarges the storm area (Fig. 5d vs. 5b) but the 267 aerosol effect is more significant (Fig. 5e vs. 5b). The CFAD of radar reflectivity in Fig. 6 also 268 shows that changes of the PDF by the urban land effect is notably smaller than the anthropogenic 269 aerosol effect. For the occurrence frequencies of high reflectivity larger than 48 dBZ, the change 270 is mainly from the anthropogenic aerosol effect (Fig. 6f-h). 271 For precipitation, we do not see an important effect of urban land on the magnitudes of 272 precipitation rate and the PDF of rain rate ( Fig. 7a-b; No_Land vs LandAero). The accumulated 273 rain is about 6.9 mm, which is also not much different from 7.2 mm in LandAero. In contrary, 274 the anthropogenic aerosol effect increases the peak rate by ~ 30%. The frequency of large rain 275 rates (> 15 mm h -1 ) is increased by about 5 times ( Fig. 7b; No_Aero vs LandAero). The joint 276 effect of both urban land and aerosol increases the accumulated rain by ~ 26%, the peak rain 277 rates by 45%, and the frequency of large rain rates by an order of magnitudes (from 278 No_LandAero to LandAero), suggesting the interactions between the two factors amplify the 279 effect on precipitation, particularly on the large rain rates. Although the Houston urban land 280 alone does not much affect the magnitude of precipitation, the initial time of the rain is advanced 281 by ~ 30 min from No_Land to LandAero (Fig. 7a), indicating that the urban land effect speeds 282 up the rain formation. Aerosol effect delays the initial and peak rain by ~ 10 min (from No_Aero 283 to LandAero). This will be further discussed in Section 3.2 on convective evolution. 284 On convective intensity, the large increases in occurrence frequencies of the updraft 285 speeds greater than 10 m s -1 in the upper-levels by the joint effect is mainly contributed by the 286 anthropogenic aerosol effect (Fig. 8e, g). Below 6 km, both the urban land and aerosol effects 287 play evident roles in increasing the occurrences of relatively large updraft speeds ( Fig. 8e-

Convective evolution 297
The urban land effect initiates surface rain about 30 minutes earlier as discussed above, 298 suggesting that the convective cloud development is affected when urban land effect is 299 considered. We examine the convective evolution for the cell over Houston using the cell-300 tracking method described in Section 2. The time evolution of the tracked cell properties is 301 shown in Fig. 9a-b. Clearly, the urban land effect enhances the reflectivity and area for the 302 tracked cell over the lifetime (from the black dashed line to black solid line), and it also 303 accelerates the development to the peak reflectivity but slows down the dissipation after the peak 304 radar reflectivity is reached ( Fig. 9a-b). The anthropogenic aerosols also enhance the convective 305 cell reflectivity and area throughout the cell lifecycle (from the black dotted line to black solid 306 line), with a much larger effect compared with the urban land effect. The anthropogenic aerosol 307 effect does not affect the timing of peak reflectivity (dotted vs. solid black in Fig. 9a-b). The 308 overall reflectivity and cell area properties are shown in Fig. 9c-d, which presents a consistent 309 story as Fig. 9a-b. The baseline simulation LandAero tends to overestimate the frequency of big 310 cell sizes (200-300 km 2 ) and underpredict the frequency of small cell size (Fig. 9d). Since 311 LandAero predicts a similar rain intensity and rain rate PDF as observations as discussed above, 312 this means that a larger storm cell than observations are needed to predict a similar precipitation 313 intensity as observations. For this reason, No_LandAero which predicts much smaller cell size 314 agrees better with the observations compared with the other simulations purely based on cell size 315 Since the small and numerous shallow cumulus clouds are difficult to be tracked with cell 322 tracking algorithm and they are excluded from the above tracking, to examine how the 323 convective storm evolves from the initial shallow cumulus period, we chose the red box shown 324 in Fig. 5 which contains the Houston cell as the study area. Since the convective storm does not 325 spatially move much with time in this study, this is a valid way to look at the temporal evolution. 326 As we can see, there is a relatively long warm cloud period for this case (Fig. 10a). With 334 both urban land and anthropogenic aerosol effects removed, the cloud development from the 335 warm cloud to mixed-phase cloud is delayed by ~ 30 min (Fig. 10d vs. 10a), so is the 336 development from the mixed-phase cloud to deep cloud. Compared Fig. 10a with 10b and 10c, 337 we see that it is mainly the urban land effect that enhances the development of warm cloud to the 338 mixed-phase cloud by nearly 30 min, while aerosol effect does not affect it (Fig. 10a vs. 10c). 339 However, it is mainly the aerosol effect that accelerates the development from the mixed-phase 340 cloud to deep cloud by about 35 min. In the case of the urban land effect removed (i.e., 341 No_Land; Fig. 10b), the anthropogenic aerosol effect makes the duration of the mixed-phase 342 cloud very short -about 35 mins shorter relative to LandAero in which both effects are 343 considered and 75 min shorter relative to No_Aero in which aerosol effect is removed but the 344 urban land effect is considered. This is due to aerosol invigoration effect in the mixed-phase 345 cloud stage which will be elaborated later. For Question (1), Fig. 10a and Fig. 12a show that the development of the warm cloud to 358 mixed-phase cloud occurs when the sea breeze circulation reaches its strongest. Also, the 359 development corresponds to the fastest and largest increase of sea breeze intensity by the urban 360 land effect (Fig. 12a). Anthropogenic aerosol does not seem to affect sea breeze circulation. The 361 enhanced sea breeze circulation in the simulations with the urban land effect considered (i.e., 362 LandAero and No_Aero) compared with No_Land and No_LandAero corresponds to the 363 increases of surface sensible heat flux and air temperature at low levels (Fig. 12b, d), which is 364 so-called "urban heat island". The urban heating effect on temperature is significant up to 0.8-km 365 altitude at its strongest time that also corresponds to strongest sea breeze time (Fig. 13b) The stronger sea breeze circulation transports more water vapor to Houston (Fig. 14). At 373 the time 1930 UTC when the sea breeze is strongest and the enhancement is the largest (Fig.  374 12a), as well as the temperature contrast between the Houston urban area and Gulf of Mexico is 375 the largest (Fig. 13b), the low-level moisture in the urban area is clearly higher in LandAero 376 compared with No_Land (Fig. 14b, color contour), which would help enhance convection. As a result, the updraft speed of the Houston convective cell is much larger in LandAero compared 378 with No_Land (Fig. 14b, contoured line). The stronger convection continues even when sea 379 breeze dissipates (Fig. 14c) Fig. 11b shows the anthropogenic aerosol effect on updraft speeds becomes notable 389 at the mixed-phase cloud stage, the effect is doubled compared with the urban land effect at the 390 mixed-phase regime (6-9 km altitudes). This corresponds to the increased net buoyancy (Fig.  391 15a, black lines) at those levels from No_Aero to LandAero, which is mainly because of the 392 increased thermal buoyancy since condensate loading effect is small (Fig. 15a) as a result of 393 enhanced condensational heating (Fig. 15c, blue lines). The condensational heating increase is 394 most significant at 3-5 km and 6-9 km altitudes, corresponding to notably increased secondary 395 droplet nucleation of small aerosol particles which are not able to be activated at cloud base (Fig.  396 15e). In this case, aerosols with diameter smaller than 80 nm but larger than 39 nm (the smallest 397 size in the 4-sectional MOSAIC), which account for about two third of the total simulated 398 aerosols, are not activated around cloud bases. All of them can be activated in the strong updrafts 399 (Fan et al., 2018). This strong secondary nucleation leads to increased droplet number and mass by the anthropogenic aerosol effects (from No_Aero to LandAero; Fig. 16a, c). To recap, the 401 anthropogenic aerosols enhance updraft velocity at the mixed-phase cloud stage mainly through 402 enhanced condensation heating (i. e., "warm-phase invigoration"), as a result of nucleating small 403 aerosol particles below 60 nm which are transported to higher-levels. This mechanism has been 404 well documented previously (Fan et al., 2007(Fan et al., , 2013 At the deep cloud stage, the anthropogenic aerosol effect becomes more significant 410 compared with that in the mixed-phase cloud stage (Fig. 11c vs. 11b), particularly at the low-411 levels. We can still see the enhancement of convective intensity by the urban land effect 412 although the sea breeze difference is relatively smaller at this stage as explained above. The 413 larger aerosol effect at the deep cloud stage compared with the mixed-phase cloud stage is 414 because the secondary droplet nucleation above cloud base becomes larger (Fig. 15f). More 415 aerosols get activated is the result of higher supersaturation since (a) updrafts are stronger than 416 the mixed-phase cloud stage and (b) more rain forms and removes droplet surface area for 417 condensation (Fan et. al., 2018). As a result, the latent heating from condensation and then the 418 thermal buoyancy is increased in a larger magnitude (Fig. 15b, d), thus a larger aerosol impact is 419 seen at the deep cloud stage. The invigorated deep convection has up to 2 times more ice particle 420 number concentration and 30% larger ice particle mass mixing ratio (Fig. 16b, d)

Figure 14
Vertical cross sections of water vapor mixing ratio (g kg -1 ; shaded), updraft velocity 785 (contour lines are 2, 6, and 11 m s -1 ), and wind vectors along the line UO in Figure 4a

Figure 15
Vertical profiles of (a-b) buoyancy terms (m s -2 ; red for Thermal buoyancy, blue for 790 condensate loading and black for total buoyancy), (c-d) latent heating (K h -1 ) from condensation 791 (blue), deposition (red), drop freezing (orange), and riming (green), and (e-f) droplet nucleation 792 rate (mg -1 s -1 ) averaged over the top 25 percentiles (i.e., 75th to 100th) of the updrafts with value 793 greater than 2 m s −1 from the simulations LandAero and No_Aero in the study area during the 794 mixed-phase cloud (left) and deep cloud (right) stages. 795

Figure 16
Vertical profiles of (a-b) number mixing ratio (mg -1 ) and (c-d) mass mixing ratio (g 797 kg -1 ) of cloud droplets (blue), rain drops (red) and ice particles (green) averaged over the top 25 798 percentiles (i.e., 75th to 100th) of the updrafts with value greater than 2 m s −1 from the 799 simulations LandAero and No_Aero in the study area during the mixed-phase cloud (left) and 800 deep cloud (right) stages. 801