A Numerical Study of the Evolving Convective Boundary Layer and Orographic Circulation A Numerical Study of the Evolving Convective Boundary Layer and Orographic Circulation around the Santa Catalina Mountains in Arizona. Part II: Circulation with Deep Convectionaround the Santa Catalina Mountains in Arizona. Part II: Circulation with Deep Convection

J. Cory Demko J. Cory Demko (coryuw@uwyo.edu)(coryuw@uwyo.edu) Bart GeertsBart Geerts (geerts@uwyo.edu)(geerts@uwyo.edu)

Department of Atmospheric Science, University of WyomingDepartment of Atmospheric Science, University of WyomingAbstraAbstractct The Weather, Research, and Forecasting (WRF) modeling system is used for

several IOPs during the Cumulus Photogrammetric, In situ and Doppler Observations (CuPIDO) campaign, conducted in summer 2006 around the Santa Catalina Mountains in southeast Arizona (Damiani et al. 2008), with the purpose to examine the interplay between boundary-layer convergence and orographic thunderstorms in a weakly-sheared environment subject to strong surface heating. This study builds on Part I (Demko et al. 2009b, presented as paper 10.4 at this conference, 4:45 pm on Tuesday), which examines thermally-forced orographic boundary-layer circulations without deep convection. The study is motivated by the fact that operational models poorly capture the timing and intensity of orographic convection, mostly due to inaccurate coupling between boundary-layer processes and cumulus convection over complex terrain. Model Setup and Data Model Setup and Data

SourcesSourcesDetailed simulations using WRF-NMM model version 3.0.1.1 have Detailed simulations using WRF-NMM model version 3.0.1.1 have been conducted on several days during CuPIDO. Model output been conducted on several days during CuPIDO. Model output validation data sources used in this study include ISFF stations validation data sources used in this study include ISFF stations located around the mountain, KTUS soundings, and WRF output, located around the mountain, KTUS soundings, and WRF output, specifically the 24 hr period from 06 UTC to 06 UTC the following specifically the 24 hr period from 06 UTC to 06 UTC the following day.day.

Horizontal grid resolution (km) 9 3 1Vertical levels 43 43 43Microphysics Lin et al. schemeLongwave radiation Rapid Radiative Transfer ModelShortwave radiation Dudhia schemeSurface layer Eta similiarityLand surface Noah Land Surface ModelPlanetary boundary layer Mellor-Yamada-Janjić schemeCumulus parameterization Kain-Fritsch scheme, none, none

Fig. 1:Fig. 1: Map (right) illustrates the ISFF station Map (right) illustrates the ISFF station locations as well as WRF’s closest grid point to locations as well as WRF’s closest grid point to that station used for validation and calculation that station used for validation and calculation of mountain-scale convergence. Specific of mountain-scale convergence. Specific location of canyons and ridges also shown.location of canyons and ridges also shown. Fig. 2:Fig. 2: Schematic transect (right) of Schematic transect (right) of the boundary-layer flow (labeled the boundary-layer flow (labeled vvnn), ), convective boundary layer depth (convective boundary layer depth (zzii, , dashed line), temperature dashed line), temperature distribution and Cu development distribution and Cu development over an isolated, heated mountain over an isolated, heated mountain without large-scale wind, based upon without large-scale wind, based upon CuPIDO observations (Demko et al. CuPIDO observations (Demko et al. 2009a) and WRF simulations (Demko 2009a) and WRF simulations (Demko et al. 2009b). Part of the solenoidal et al. 2009b). Part of the solenoidal circulation feeds the orographic circulation feeds the orographic cumulus cloud (solid squiggly lines). cumulus cloud (solid squiggly lines). The dashed squiggly lines indicate The dashed squiggly lines indicate possible convergent flow above the possible convergent flow above the CBL, associated with moist CBL, associated with moist convectionconvection.

Table 1Table 1: Configurations of the WRF model, version 3.0.1.1 for : Configurations of the WRF model, version 3.0.1.1 for simulations of 06 August 2006 shown in this study. Multiple values simulations of 06 August 2006 shown in this study. Multiple values indicate domains 1 (outer) through 3 (inner).indicate domains 1 (outer) through 3 (inner).

1.1.Deep moist convection overwhelms and interrupts the thermally-driven Deep moist convection overwhelms and interrupts the thermally-driven anabatic surface flow, due to cold pool dynamicsanabatic surface flow, due to cold pool dynamics. After the Cb dissipates . After the Cb dissipates or moves of the mountain, the mountain-scale surface convergence or moves of the mountain, the mountain-scale surface convergence reestablishes, possibly resulting in multiple convection cycles. WRF V.3 reestablishes, possibly resulting in multiple convection cycles. WRF V.3 accurately simulates this evolution, with two peaks of convective accurately simulates this evolution, with two peaks of convective intensity on 6 Aug, in the early and the late afternoon. intensity on 6 Aug, in the early and the late afternoon.

2.2.WRF simulations indicate that enhanced MSC does develop 1-2 hours WRF simulations indicate that enhanced MSC does develop 1-2 hours prior to orographic deep convection. prior to orographic deep convection. Deeper convection is not Deeper convection is not necessarily associated with stronger precursor convergence, but does necessarily associated with stronger precursor convergence, but does produce a more intense cold pool and stronger surface divergence. produce a more intense cold pool and stronger surface divergence.

3.3.Even though the surface flow is divergent after a Cb outbreakEven though the surface flow is divergent after a Cb outbreak, the , the anomalous low may persist over the mountain, which reestablishes the anomalous low may persist over the mountain, which reestablishes the PGF, and therefore, the anabatic flow.PGF, and therefore, the anabatic flow.

4.4.The mass budget (Fig. 9) The mass budget (Fig. 9) indicates a relationship between the amount of indicates a relationship between the amount of moist convection and imbalance. Not shown are hours without moist moist convection and imbalance. Not shown are hours without moist convection which shows smaller residuals. These large imbalances are convection which shows smaller residuals. These large imbalances are the subject to further analysis and may be attributed to the subject to further analysis and may be attributed to evaporative/condensational cooling/warming and errors associated with evaporative/condensational cooling/warming and errors associated with finite differencing.finite differencing.

ReferencesReferencesDamiani, R., J. Zehnder, B. Geerts, J. Demko, S. Haimov, J. Petti, G.S. Poulos, A. Razdan, J. Hu, M. Leuthold, and J. French, 2008: Damiani, R., J. Zehnder, B. Geerts, J. Demko, S. Haimov, J. Petti, G.S. Poulos, A. Razdan, J. Hu, M. Leuthold, and J. French, 2008:

Cumulus Photogrammetric, In-situ and Doppler Observations: the CuPIDO 2006 experiment. Cumulus Photogrammetric, In-situ and Doppler Observations: the CuPIDO 2006 experiment. Bull. Amer. Meteor. Soc.,Bull. Amer. Meteor. Soc., 89, 57– 89, 57–73.73.

Demko, J. C., B. Geerts, Q. Miao, and J. Zehnder, 2009a: Boundary-layer energy transport and cumulus development over a heated Demko, J. C., B. Geerts, Q. Miao, and J. Zehnder, 2009a: Boundary-layer energy transport and cumulus development over a heated mountain: an observational study. mountain: an observational study. Mon. Wea. Rev. , Mon. Wea. Rev. , 137, 447–468.137, 447–468.

Demko, J. C., B. Geerts, Q. Miao, 2009b: A Numerical Study of the Evolving Convective Boundary Layer and Orographic Demko, J. C., B. Geerts, Q. Miao, 2009b: A Numerical Study of the Evolving Convective Boundary Layer and Orographic Circulation around the Santa Catalina Mountains in Arizona. Part I: Circulation without Deep Convection. Circulation around the Santa Catalina Mountains in Arizona. Part I: Circulation without Deep Convection. Mon. Wea. RevMon. Wea. Rev, , submitted. (submitted. (presented as paper 10.4 at this conference, 4:45 pm on Tuesday)

Case 1: 06 August 2006Case 1: 06 August 2006

Fig. 7Fig. 7: (a) WRF’s Cloud top chronology (CTC) and the observed CTC with : (a) WRF’s Cloud top chronology (CTC) and the observed CTC with various other WRF derived stability parameters for 6 August. These various other WRF derived stability parameters for 6 August. These parameters (LCL, LFC, CBL, MLCAPE, and MLCIN) were computed over a parameters (LCL, LFC, CBL, MLCAPE, and MLCIN) were computed over a 30x30 km30x30 km22 box. The cloud top chronology tracks the highest cloud liquid box. The cloud top chronology tracks the highest cloud liquid water and ice having a value at least 0.01 gkmwater and ice having a value at least 0.01 gkm-3-3. The geographic location of . The geographic location of the highest cloud element in WRF is shown in panel bthe highest cloud element in WRF is shown in panel b .

ConclusiConclusionsons

Fig. 3Fig. 3: Conceptual view of the diurnal evolution of a weakly-capped CBL and : Conceptual view of the diurnal evolution of a weakly-capped CBL and thermally-forced circulation over an isolated mountain under negligible mean wind thermally-forced circulation over an isolated mountain under negligible mean wind and enough moisture for shallow to mediocre cumulus development (the focus of and enough moisture for shallow to mediocre cumulus development (the focus of part I, Demko et al. 2009b). The horizontal (vertical) dimensions of the west-east part I, Demko et al. 2009b). The horizontal (vertical) dimensions of the west-east cross section are ~50 km (~5 km). The times shown are: (a) near sunrise; (b) shortly cross section are ~50 km (~5 km). The times shown are: (a) near sunrise; (b) shortly before orographic cumulus development; (c) orographic cumulus phase, typically before orographic cumulus development; (c) orographic cumulus phase, typically around solar noon; (d) near sunset. Red contours are dry isentropes, purple contours around solar noon; (d) near sunset. Red contours are dry isentropes, purple contours indicate variations of the height of the 850 hPa surface (Zindicate variations of the height of the 850 hPa surface (Z850850) and thus the direction ) and thus the direction of the pressure gradient, the bold grey contour is the top of the CBL or the nocturnal of the pressure gradient, the bold grey contour is the top of the CBL or the nocturnal boundary-layer (NBL), and the black arrowed contours indicate the mean secondary boundary-layer (NBL), and the black arrowed contours indicate the mean secondary circulation. The sign of the surface sensible heat (SH) flux is indicated by the circulation. The sign of the surface sensible heat (SH) flux is indicated by the squiggly arrow near the surface, pointing upward for a positive heat flux.squiggly arrow near the surface, pointing upward for a positive heat flux.

The schematic on The schematic on the left shows the the left shows the diurnal evolution diurnal evolution of the thermally-of the thermally-forced circulation forced circulation with mediocre with mediocre moist convection moist convection (Part I). In this (Part I). In this poster (Part II) we poster (Part II) we examine the two-examine the two-way interaction way interaction between the BL between the BL circulation and circulation and deep convection.deep convection.

Fig. 4Fig. 4:: 21 21 vertical cross vertical cross sectional sectional average of average of θθ’’

west,east west,east , u, u’’, w,, w, and PBL and PBL height for 12 height for 12 (a), 15 (b), 18 (a), 15 (b), 18 (c), and 21 (d) (c), and 21 (d) UTC 06 UTC 06 August 2006. August 2006. Also shown Also shown are profiles of are profiles of the mean the mean θθwestwest

(solid), (solid), θθeasteast

(dashed) (dashed) (profiles on (profiles on the left) and u the left) and u (profile on the (profile on the right).right).

By 21 UTC, the PBL has deepened to ~675 hPa over the mountain. The warm dome is By 21 UTC, the PBL has deepened to ~675 hPa over the mountain. The warm dome is approximately 15 km wide, extends over the majority of the CBL, and has strength of up to approximately 15 km wide, extends over the majority of the CBL, and has strength of up to 2 K. The anabatic flow is rather symmetric on both sides with maxima 2 K. The anabatic flow is rather symmetric on both sides with maxima u’u’ located near the located near the surface and on the western slopes. The anabatic circulation collides slightly downwind surface and on the western slopes. The anabatic circulation collides slightly downwind (west) of the crest, and is the main focus of precipitation and a deep Cb burst ((west) of the crest, and is the main focus of precipitation and a deep Cb burst (23 UTC 06 Aug 23 UTC 06 Aug

– 01 UTC 07 Aug– 01 UTC 07 Aug). ).

Fig. 5Fig. 5:: Time versus height Time versus height plot of mountain-scale (30x30 plot of mountain-scale (30x30 kmkm22) convergence (color ) convergence (color shaded), mean vertical shaded), mean vertical velocity as inferred from the velocity as inferred from the convergence field (black convergence field (black solid and dotted lines), and solid and dotted lines), and PBL height (black dashed). PBL height (black dashed). Note the daytime deepening Note the daytime deepening of the PBL in sync with the of the PBL in sync with the deepening of MSC. Deep deepening of MSC. Deep convection starting at 23 convection starting at 23 UTC produces low-level UTC produces low-level subsidence and near-surface subsidence and near-surface divergence (cold pool divergence (cold pool dynamics). Local solar noon dynamics). Local solar noon is at 19:30 UTC, sunrise at is at 19:30 UTC, sunrise at 12:35 UTC, sunset at 02:25 12:35 UTC, sunset at 02:25 UTC.UTC.

Fig. 6Fig. 6: Hydrostatic pressure : Hydrostatic pressure difference between Mt. Lemmon difference between Mt. Lemmon and closed boxes (black, left axis) and closed boxes (black, left axis) and subsequent surface and subsequent surface convergence (grey, right axis) for convergence (grey, right axis) for (a) 100, (b) 400, (c) 900, and (d) (a) 100, (b) 400, (c) 900, and (d) 1600 km1600 km22 boxes for period 06 UTC boxes for period 06 UTC 06 August through 06 UTC 07 06 August through 06 UTC 07 August. The vertical lines show the August. The vertical lines show the time of minimum & maximum time of minimum & maximum values, when the PGF becomes values, when the PGF becomes directed toward the mountain, and directed toward the mountain, and when surface flow becomes when surface flow becomes convergent. The average height convergent. The average height MSL of each box is shown in the MSL of each box is shown in the upper left side of each figure.upper left side of each figure.

Fig. 8Fig. 8: 2 m potential temperature (color shaded surface), 10 m wind (barbs), and cloud : 2 m potential temperature (color shaded surface), 10 m wind (barbs), and cloud liquid or frozen water isosurface of 0.01 gkmliquid or frozen water isosurface of 0.01 gkm-1-1 for WRF forecast hours 19, 20, 21 for WRF forecast hours 19, 20, 21 (top/left to right), 22, 23 UTC 06 August, and 00 UTC 07 August (bottom/left to right). (top/left to right), 22, 23 UTC 06 August, and 00 UTC 07 August (bottom/left to right). Also illustrated are 1 hr accumulated precipitation contours over terrain in intervals of Also illustrated are 1 hr accumulated precipitation contours over terrain in intervals of 0.01, 0.1, 0.25, 0.5 inches respectively. Cloud isosurfaces are only showed within the 0.01, 0.1, 0.25, 0.5 inches respectively. Cloud isosurfaces are only showed within the 900 km900 km22 box centered on Mt. Lemmon to reduce image clutter. Notice how the box centered on Mt. Lemmon to reduce image clutter. Notice how the majority of convection and subsequent precipitation is centered upon Pusch ridge, majority of convection and subsequent precipitation is centered upon Pusch ridge, where the anabatic circulation belts collide slightly downwind of the peaks generating where the anabatic circulation belts collide slightly downwind of the peaks generating a narrow column of upward motion.a narrow column of upward motion.

Fig. 9Fig. 9: Mass budget analysis for hours 19 UTC 06 August – 00 UTC 07 August for 3 : Mass budget analysis for hours 19 UTC 06 August – 00 UTC 07 August for 3 closed boxes performed over the 900 kmclosed boxes performed over the 900 km22 area with vertical dimensions of box 1 = (z area with vertical dimensions of box 1 = (zmid-mid-

PBLPBL to z to zsfcsfc), box 2 = (z), box 2 = (zPBL-topPBL-top to z to zmid-PBLmid-PBL), and box 3 =(z), and box 3 =(zcloud topcloud top to z to zPBL-topPBL-top) which are ) which are represented by the horizontal planes. Flux values are in 10represented by the horizontal planes. Flux values are in 1066 kgs kgs-1-1 and arrows indicate and arrows indicate flux direction. Cloud isosurfaces of 0.01 gkmflux direction. Cloud isosurfaces of 0.01 gkm-1-1 are also shown. are also shown.